Stoke Rehabilitation NICE Guidelines
Stoke Rehabilitation NICE Guidelines
Stoke Rehabilitation NICE Guidelines
FinalFullGuideline
StrokeRehabilitation
Longtermrehabilitationafterstroke
Clinicalguideline162
Methods,evidenceandrecommendations
29May2013
FinalDraft
CommissionedbytheNationalInstitutefor
HealthandCareExcellence
StrokeRehabilitation
Contents
Disclaimer
HealthcareprofessionalsareexpectedtotakeNICEclinicalguidelinesfullyintoaccountwhen
exercisingtheirclinicaljudgement.However,theguidancedoesnotoverridetheresponsibilityof
healthcareprofessionalstomakedecisionsappropriatetothecircumstancesofeachpatient,in
consultationwiththepatientand/ortheirguardianorcarer.
Copyright
NationalClinicalGuidelineCentre,2013.
Funding
NationalInstituteforHealthandCareExcellence
NationalClinicalGuidelineCentre,2013.
StrokeRehabilitation
Contents
Contents
Guidelinedevelopmentgroupmembers.......................................................................................11
Acknowledgments......................................................................................................................13
1
Introduction..........................................................................................................................14
Developmentoftheguideline...............................................................................................16
2.1
WhatisaNICEclinicalguideline?.......................................................................................16
2.2
Remit...................................................................................................................................16
2.3
Whodevelopedthisguideline?..........................................................................................17
2.4
Whatthisguidelinecovers..................................................................................................17
2.5
Whatthisguidelinedoesnotcover....................................................................................17
2.6
RelationshipsbetweentheguidelineandotherNICEguidance.........................................17
Guidelinesummary...............................................................................................................20
3.1
Keyprioritiesforimplementation.......................................................................................20
3.1.1 Strokeunits............................................................................................................20
3.1.2 Thecoremultidisciplinarystroketeam..................................................................20
3.1.3 Healthandsocialcareinterface.............................................................................20
3.1.4 Transferofcarefromhospitaltocommunity........................................................20
3.1.5 Settinggoalsforrehabilitation...............................................................................21
3.1.6 Intensityofstrokerehabilitation............................................................................21
3.1.7 Cognitivefunctioning.............................................................................................21
3.1.8 Emotionalfunctioning............................................................................................21
3.1.9 Swallowing.............................................................................................................21
3.1.10 Returntowork.......................................................................................................21
3.1.11 Longtermhealthandsocialsupport.....................................................................22
3.2
Fulllistofrecommendations..............................................................................................22
3.3
Keyresearchrecommendations.........................................................................................34
Methods................................................................................................................................35
4.1
Developingthereviewquestionsandoutcomes................................................................35
4.2
Searchingforevidence........................................................................................................41
4.2.1 Clinicalliteraturesearch.........................................................................................41
4.2.2 Healtheconomicliteraturesearch.........................................................................42
4.3
Evidenceofeffectiveness....................................................................................................42
4.3.1 Inclusion/exclusioncriteria....................................................................................42
4.3.2 Methodsofcombiningclinicalstudies...................................................................43
4.3.3 Typeofstudies.......................................................................................................44
4.3.4 Typeofanalysis......................................................................................................44
4.3.5 Appraisingthequalityofevidencebyoutcomes...................................................44
NationalClinicalGuidelineCentre,2013.
4
StrokeRehabilitation
Contents
4.3.6 Gradingthequalityofclinicalevidence.................................................................46
4.3.7 Studylimitations....................................................................................................46
4.3.8 Inconsistency..........................................................................................................47
4.3.9 Indirectness............................................................................................................47
4.3.10 Imprecision.............................................................................................................47
4.4
Evidenceofcosteffectiveness............................................................................................50
4.4.1 Literaturereview....................................................................................................51
4.4.2 Undertakingnewhealtheconomicanalysis..........................................................52
4.4.3 Costeffectivenesscriteria......................................................................................53
4.5
PostconsultationprotocolincludingmodifiedDelphimethodology.................................53
4.6
Developingrecommendations............................................................................................57
4.6.1 Researchrecommendations..................................................................................57
4.6.2 Validationprocess..................................................................................................57
4.6.3 Updatingtheguideline...........................................................................................58
4.6.4 Disclaimer...............................................................................................................58
4.6.5 Funding...................................................................................................................58
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke..................59
5.1
Strokeunits.........................................................................................................................59
5.1.1 EvidenceReview:Inpeopleafterstroke,doesorganisedrehabilitationcare
(comprehensive,rehabilitationandmixedrehabilitationstrokeunits)
improveoutcome(mortality,dependency,requirementforinstitutionalcare
andlengthofhospitalstay)?..................................................................................59
5.1.2 Recommendationsandlinkstoevidence..............................................................77
5.2
Thecoremultidisciplinarystroketeam..............................................................................78
5.2.1 EvidenceReview:Whatshouldbetheconstituencyofamultidisciplinary
rehabilitationteamandhowshouldtheteamworktogethertoensurethe
bestoutcomesforpeoplewhohavehadastroke?...............................................78
5.2.2 Delphistatementswhereconsensuswasachieved...............................................79
5.2.3 Delphistatementwhereconsensuswasnotreached...........................................80
5.2.4 RecommendationsandlinkstoDelphiconsensussurvey.....................................82
5.3
Healthandsocialcareinterface..........................................................................................84
5.3.1 Delphistatementswhereconsensuswasachieved...............................................84
5.3.2 RecommendationsandlinkstoDelphiconsensussurvey.....................................85
5.4
Transferofcarefromhospitaltocommunity.....................................................................87
5.4.1 Earlysupporteddischarge......................................................................................87
5.4.2 EvidenceReview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofearlysupporteddischargeversususualcare?.............................87
5.4.3 Recommendationsandlinktoevidence..............................................................113
5.4.4 Transferofcarefromhospitaltocommunity......................................................115
5.4.5 EvidenceReview:Whatplanningandsupportshouldbeundertakenbythe
NationalClinicalGuidelineCentre,2013.
5
StrokeRehabilitation
Contents
multidisciplinaryrehabilitationteambeforeapersonwhohadastrokeis
dischargedfromhospitalortransferstoanotherteam/settingtoensurea
successfultransitionofcare?...............................................................................115
5.4.6 Delphistatementswhereconsensuswasachieved.............................................116
5.4.7 Delphistatementwhereconsensuswasnotreached.........................................117
5.4.8 RecommendationsandlinkstoDelphiconsensussurvey...................................119
6
Planninganddeliveringstrokerehabilitation......................................................................123
6.1
Screeningandassessment................................................................................................123
6.1.1 EvidenceReview:Inplanningrehabilitationforapersonafterstrokewhat
assessmentsandmonitoringshouldbeundertakentooptimisethebest
outcomes?............................................................................................................123
6.1.2 Delphistatementswhereconsensuswasachieved.............................................123
6.1.3 Delphistatementwhereconsensuswasnotreached.........................................126
6.1.4 RecommendationsandlinkstoDelphiconsensussurvey...................................127
6.2
Settinggoalsforrehabilitation..........................................................................................130
6.2.1 EvidenceReview:Doestheapplicationofpatientgoalsettingaspartof
planningstrokerehabilitationactivitiesleadtoanimprovementin
psychologicalwellbeing,functioningandactivity?..............................................130
6.2.2 Economicevidencesummary...............................................................................140
6.2.3 Evidencestatements............................................................................................141
6.2.4 Economicevidencestatements...........................................................................142
6.2.5 Recommendationsandlinkstoevidence............................................................142
6.2.6 Delphistatementswhereconsensuswasachieved.............................................144
6.2.7 Delphistatementswhereconsensuswasnotachieved......................................145
6.2.8 RecommendationsandlinkstoDelphiconsensussurvey...................................147
6.3
Planningrehabilitation......................................................................................................148
6.3.1 Delphistatementswhereconsensuswasachieved.............................................148
6.3.2 Delphistatementwhereconsensuswasnotreached.........................................150
6.3.3 RecommendationsandlinkstoDelphiconsensussurvey...................................151
6.4
Intensityofstrokerehabilitation......................................................................................153
6.4.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofintensiverehabilitationversusstandardrehabilitation?...........153
6.4.2 Recommendationsandlinktoevidence..............................................................166
Supportandinformation.....................................................................................................170
7.1
Providingsupportandinformation...................................................................................170
7.1.1 Evidencereview:Whatistheclinicalandcosteffectivenessofsupported
informationprovisionversusunsupportedinformationprovisiononmood
anddepressioninpeoplewithstroke?................................................................170
7.1.2 Recommendationsandlinktoevidence..............................................................179
Cognitivefunctioning..........................................................................................................181
8.1
Visualneglect....................................................................................................................181
NationalClinicalGuidelineCentre,2013.
6
StrokeRehabilitation
Contents
8.1.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofcognitiverehabilitationversususualcaretoimprovespatial
awarenessand/orvisualneglect?.......................................................................181
8.1.2 Recommendationsandlinktoevidence..............................................................194
8.2
Memoryfunction..............................................................................................................195
8.2.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofmemorystrategiesversususualcaretoimprovememory.......196
8.2.2 Recommendationsandlinktoevidence..............................................................201
8.3
Attentionfunction.............................................................................................................202
8.3.2 Recommendationsandlinktoevidence..............................................................210
Emotionalfunctioning.........................................................................................................213
9.1
Psychologicaltherapies.....................................................................................................213
9.1.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofpsychologicaltherapiesprovidedtothefamily(includingthe
patient)?...............................................................................................................213
9.1.2 Recommendationsandlinktoevidence..............................................................222
10 Vision..................................................................................................................................225
10.1 Eyemovementtherapy.....................................................................................................225
10.1.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofeyemovementtherapyforvisualfieldlossversususualcare? 225
10.1.2 Recommendationsandlinktoevidence..............................................................233
10.2 Diplopiaorotherongoingvisualsymptomsafterstroke.................................................234
10.2.1 Evidencereview:Howshouldpeoplewithvisualimpairmentsincluding
diplopiabebestmanagedafterastroke?...........................................................235
10.2.2 Delphistatementswhereconsensuswasachieved.............................................235
10.2.3 Delphistatementwhereconsensuswasnotreached.........................................235
10.2.4 RecommendationsandlinkstoDelphiconsensussurvey...................................237
11 Swallowing..........................................................................................................................238
11.1.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofinterventionsforswallowingversusalternativeinterventions
/usualcaretoimprovedifficultyswallowing(dysphagia)?.................................238
11.1.2 EconomicLiteraturereview.................................................................................245
11.1.3 Evidencestatements............................................................................................245
11.1.4 Recommendationsandlinktoevidence..............................................................247
12 Communication...................................................................................................................249
12.1 Aphasia..............................................................................................................................249
12.1.1 EvidenceReview:Inpeoplewhohaveaphasiaafterstrokeisspeechand
languagetherapycomparedtonospeechandlanguagetherapyorplacebo
(socialsupportandstimulation)effectiveinimproving
language/communicationabilitiesand/orpsychologicalwellbeing?..................249
12.2 Dysarthria..........................................................................................................................279
12.2.1 EvidenceReview:Inpeopleafterstrokeisspeechandlanguagetherapy
NationalClinicalGuidelineCentre,2013.
7
StrokeRehabilitation
Contents
comparedtosocialsupportandstimulationeffectiveinimproving
dysarthria?...........................................................................................................279
12.2.2 Recommendationsandinktoevidence...............................................................282
12.3 Speechandlanguagetherapiesfordysarthriaandapraxiaofspeech.............................286
12.3.1 Whatinterventionsimprovecommunicationinpeopledysphasia,dysarthria
andapraxiaofspeech?.........................................................................................286
12.3.2 Delphistatementswhereconsensuswasachieved.............................................286
12.3.3 Delphistatementwhereconsensuswasnotreached.........................................287
12.3.4 RecommendationsandlinkstoDelphiconsensussurvey...................................291
12.4 Intensityofspeechandlanguagetherapy........................................................................292
12.4.1 Evidencereview:Inpeopleafterstrokewithcommunicationdifficultieswhat
istheclinicalandcosteffectivenessofintensivespeechtherapyversus
standardspeechtherapy?....................................................................................292
12.4.2 Recommendationsandlinktoevidence..............................................................305
12.5 Listeneradvice..................................................................................................................307
12.5.1 Whatlisteneradviceskills/trainingorinformationwouldhelpfamily
members/carersimprovecommunicationinpeoplewithaphasiaafter
stroke?..................................................................................................................307
12.5.2 Recommendationsandlinktoevidence..............................................................312
13 Movement..........................................................................................................................313
13.1 Strengthtraining...............................................................................................................313
13.1.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofstrengthtrainingversususualcareonimprovingfunction
andreducingdisability?.......................................................................................314
13.1.2 Recommendationsandlinktoevidence..............................................................336
13.2 FitnessTraining.................................................................................................................338
13.2.1 Inpeopleafterstroke,doescardiorespiratoryorresistancefitnesstraining
improveoutcome(fitness,function,qualityoflife,mood)andreduce
disability?.............................................................................................................338
13.2.2 Recommendationsandlinkstoevidence............................................................395
13.3 Handandarmtherapies:orthosesfortheupperlimb.....................................................397
13.3.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessoforthosesforpreventionoflossofrangeofmovementinthe
upperlimbversususualcare?..............................................................................397
13.3.2 Recommendationsandlinktoevidence..............................................................403
13.4 Electricalstimulation:upperlimb....................................................................................404
13.4.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofelectricalstimulation(ES)forhandfunctionversususual
care?.....................................................................................................................404
13.4.2 Recommendationsandlinktoevidence..............................................................437
13.5 Constraintinducedmovementtherapy............................................................................438
13.5.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofconstraintinducedtherapyversususualcareonimproving
NationalClinicalGuidelineCentre,2013.
8
StrokeRehabilitation
Contents
functionandreducingdisability?.........................................................................438
13.5.2 Recommendationsandlinktoevidence..............................................................456
13.6 Shoulderpain....................................................................................................................458
13.6.1 Howshouldpeoplewithshoulderpainafterstrokebemanagedtoreduce
pain?.....................................................................................................................458
13.6.2 Delphistatementswhereconsensuswasachieved.............................................458
13.6.3 Delphistatementwhereconsensuswasnotreached.........................................459
13.6.4 RecommendationsandlinkstoDelphiconsensussurvey...................................460
13.7 Repetitivetasktraining.....................................................................................................461
13.7.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofrepetitivetasktrainingversususualcareonimproving
functionandreducingdisability?.........................................................................461
13.7.2 Recommendationsandlinktoevidence..............................................................472
13.8 Walkingtherapies:treadmillandtreadmillwithbodyweightsupport...........................473
13.8.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofalltreadmillversususualcareonimprovingwalking?..............474
13.8.2 Evidencereview:Inpeopleafterstrokewhocanwalk,whatistheclinicaland
costeffectivenessoftreadmillplusbodysupportversustreadmillonlyon
improvingwalking?..............................................................................................474
13.8.3 Recommendationsandlinktoevidence..............................................................496
13.9 Electromechanicalgaittraining........................................................................................498
13.9.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofelectromechanicalgaittrainingversususualcareon
improvingfunctionandreducingdisability?........................................................498
13.9.2 Recommendationsandlinktoevidence..............................................................517
13.10Anklefootorthoses..........................................................................................................518
13.10.1Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofAnkleFootorthosesofalltypestoimprovewalkingfunction
versususualcare?................................................................................................518
13.10.2Recommendationsandlinktoevidence..............................................................527
14 Selfcare..............................................................................................................................530
14.1 Intensityofoccupationaltherapyforpersonalactivitiesofdailyliving...........................530
14.1.1 Evidencereview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofintensiveoccupationaltherapyfocusedspecificallyon
personalactivitiesofdailyliving(dressing/others)versususualcare?.............530
14.1.2 RecommendationsandLinktoEvidence.............................................................540
15 Communityparticipationandlongtermrecovery................................................................543
15.1 Returntowork..................................................................................................................543
15.1.1 EvidenceReview:Inpeopleafterstrokewhatistheclinicalandcost
effectivenessofinterventionstoaidreturntoworkversususualcare?.............543
15.1.2 Clinicalevidence...................................................................................................544
15.1.3 Recommendationsandlinktoevidence..............................................................548
NationalClinicalGuidelineCentre,2013.
9
StrokeRehabilitation
Contents
15.2 Longtermhealthandsocialsupport................................................................................551
15.2.1 Whatongoinghealthandsocialsupportdothepersonafterstrokeandtheir
carer(s)requiretomaximisesocialparticipationandlongtermrecovery?........551
15.2.2 Delphistatementswhereconsensuswasachieved.............................................551
15.2.3 Delphistatementwhereconsensuswasnotreached.........................................553
15.2.4 RecommendationsandlinkstoDelphiconsensussurvey...................................555
16 Acronymsandabbreviations...............................................................................................558
17 Glossary..............................................................................................................................560
18 Referencelist......................................................................................................................573
NationalClinicalGuidelineCentre,2013.
10
StrokeRehabilitation
Guidelinedevelopmentgroupmembers
Guidelinedevelopmentgroupmembers
Name
Organisation
Dr.DianePlayford(Chair)
Readerinneurologicalrehabilitation
UCLInstituteofNeurology
HonoraryConsultantNeurologist
NationalHospitalforNeurologyandNeurosurgery,UCLHNHSFoundation
Trust
Dr.KhalidAli
SeniorLecturerinGeriatrics
BrightonandSussexMedicalSchool
Mr.MartinBird
Carermember
Mr.RobinCant
Patientmember
Ms.SandraChambers
ClinicalSpecialist
StrokeandNeurorehabilitation,PhysiotherapyDepartment,GuysandSt.
ThomasHospitalNHSFoundationTrust
Ms.LouiseClark
TraineeConsultantPractitionerinNeurology(Stroke)
NHSSouthCentral
SeniorOccupationalTherapistspecialisinginStroke
Dr.AvrilDrummond
DeputyDirector,
TrentLocalResearchNetworkforStroke
(ResignedfromtheGuidelineDevelopmentGroupinOctober2012)
Prof.AnneForster
ProfessorofStrokeRehabilitation
InstituteofHealthSciences,UniversityofLeedsandBradford
InstituteforHealthResearch
(ResignedfromtheGuidelineDevelopmentGroupinMarch2013)
Dr.KathrynHead
PrincipalSpeechandLanguageTherapist
Strokeservice,CwmTafHealthBoard,SouthWales
Ms.PamelaHolmes
Representative
SocialCareInstituteforExcellence
Ms.HelenE.Hunter
ClinicalSpecialistNeurophysiotherapist
NorthumberlandCareTrust
Dr.NajmaKhanBourne
ConsultantClinicalNeuropsychologist
ClinicalLeadforNeuropsychologicalNeurorehabilitation
KingsCollegeHospital,KingsCollegeHospitalNHSFoundationTrust
Dr.KeithMacDermott
GeneralPractitioner(RetiredfromGeneralPracticeinApril2010)
Drs.Priceandpartners,York
Dr.RoryOConnor
HonoraryConsultantinRehabilitationMedicine
CommunityRehabilitationUnit,LeedsCommunityHealthcareNHSTrust
LeedsHonoraryConsultantinRehabilitationMedicine
NationalDemonstrationCentreinRehabilitation,LeedsTeachingHospitals
NHSTrust,Leeds
Ms.SueThelwell
StrokeServicesCoordinator
UniversityHospitalsCoventryandWarwickshireNHSTrust
NationalClinicalGuidelineCentre,2013.
11
StrokeRehabilitation
Guidelinedevelopmentgroupmembers
Cooptees/ExpertAdvisors
Name
Organisation
Dr.CharlieDavie
ConsultantNeurologistattheRoyalFreeLondonNHSFoundationTrust
ProgrammeDirectorforNeuroscienceatUniversityCollegeLondon
Partners
Ms.JuliaParnaby
HeadofStrokeInformationServices
StrokeAssociation
Ms.CarolePound
Researcheraphasiatherapyandsupportservices
CentreforResearchandRehabilitation,
BrunelUniversity
Dr.FionaRowe
SeniorLecturerinOrthoptics
UniversityofLiverpool
Mr.MirekSkrypak
ClinicalCoordinatorandManager,CamdenEarlySupportedDischargeand
StrokeNavigationServices
Mr.RonaldBarneyWhite
SeniorOrthotist
SandwellandWestBirminghamHospitalsNHSTrust
NCGCStaffmembersontheguidelinedevelopmentgroup
Name
Role
Ms.GillRitchie
GuidelineLead
Ms.TamaraDiaz
ProjectManager
Dr.KatharinaDworzynski
SeniorResearchFellow
Ms.ElisabettaFenu
SeniorHealthEconomist
Ms.LinaGulhane
JointHeadofInformationScience
Dr.JonathanNyong
ResearchFellow
Dr.AngelaCooper
SeniorResearchFellow untilJuly2010
Dr.PaulineTurner
ResearchFellow
untilAugust2010
Dr.AntoniaMorga
HealthEconomist
untilApril2011
Ms.LolaAdedokun
HealthEconomist
untilJune2012
Dr.GrammatiSarri
SeniorResearchFellow untilJuly2012
Ms.KateLovibond
SeniorHealthEconomist untilAugust2012
NationalClinicalGuidelineCentre,2013.
12
StrokeRehabilitation
Acknowledgments
Acknowledgments
Thedevelopmentofthisguidelinewasgreatlyassistedbythefollowingpeople:
NCGC:
Role
IanBullock
ChiefOperatingOfficer
SerenaCarville
SeniorResearchFellow/ProjectManager
RalphHughes
HealthEconomist
RosaLau
ResearchFellow
SharanginiRajesh
ResearchFellow
JaymeeniSolanki
Projectcoordinator
PhilippeLaramee
HealthEconomist
RichardWhitome
InformationScientist
DavidWonderling
HeadofHealthEconomics
HatiZorba
Projectcoordinator
External
Role
JacobyPatterson
ResearchFellow
ClaireTurner
NICECommissioningManagerfromJuly2010
SarahWillett
NICECommissioningmanageruntilJuly2010
NationalClinicalGuidelineCentre,2013.
13
StrokeRehabilitation
Introduction
1 Introduction
StrokeisamajorhealthproblemintheUK.EachyearinEngland,approximately110,000people230,
inWales11,000andinNorthernIreland4,000peoplehaveafirstorrecurrentstroke250.Most
peoplesurviveafirststroke,butoftenhavesignificantmorbidity.Morethan900,000peoplein
Englandarelivingwiththeeffectsofstroke.StrokemortalityratesintheUKhavebeenfalling
steadilysincethelate1960s25.Thedevelopmentofstrokeunitsfollowingthepublicationofthe
StrokeUnitTrialistsCollaborationmetaanalysisofstrokeunitcare1,andthefurtherreorganisation
ofservicesfollowingtheadventofthrombolysishaveresultedinfurthersignificantimprovementsin
mortalityandmorbidityfromstroke(asdocumentedintheNationalSentinelAuditforStroke123).
However,theburdenofstrokemayincreaseinthefutureasaconsequenceoftheageing
population.
Despiteimprovementsinmortalityandmorbidity,strokesurvivorsneedaccesstoeffective
rehabilitationservices.Over30%ofpeoplehavepersistingdisabilityandtheyneedaccesstostroke
serviceslongterm.Strokerehabilitationisamultidimensionalprocess,whichisdesignedtofacilitate
restorationof,oradaptationto,thelossofphysiologicalorpsychologicalfunctionwhenreversalof
theunderlyingpathologicalprocessisincomplete.Rehabilitationaimstoenhancefunctional
activitiesandparticipationinsocietyandthusimprovequalityoflife.
Astrokerehabilitationservicecomprisesamultidisciplinaryteamofpeoplewhoworktogether
towardsgoalsforeachpatient,involveandeducatethepatientandfamily,haverelevantknowledge
andskillstohelpaddressmostcommonproblemsfacedbytheirpatients276Keyaspectsof
rehabilitationcareincludemultidisciplinaryassessment,identificationoffunctionaldifficultiesand
theirmeasurement,treatmentplanningthroughgoalsetting,deliveryofinterventionswhichmay
eithereffectchangeorsupporttheindividualinmanagingpersistingchange,andevaluationof
effectiveness.
AssessmentistypicallyundertakenusingtheWorldHealthOrganisation(WHO)International
ClassificationofFunctioning,DisabilityandHealth(ICF)whichprovidesabiopsychosocialmodelof
disability.AswellassupportingcomprehensiveassessmenttheICFcanbeusedingoalsetting&
treatmentplanningandmonitoring,aswellasoutcomemeasurement.Treatmentsarelargely
deliveredviaphysiotherapists,occupationaltherapists,speechandlanguagetherapists,nursesand
psychologists.Othercomponentsofrehabilitationincludethelearningofnewskillstocircumvent
thoselost;adaptationtolossbyboththepatientandfamily;theapplicationofnewtechnologies,
appliancesandenvironmentalmodifications;andthedevelopmentofnewservicedeliverysystems.
Therehabilitationprocessaimstomaximisetheparticipationofthepatientinhisorhersocial
setting,includingsupportingpeopletoestablishrolesandoccupations,andminimisethepainand
distressexperiencedbythepatientandtheirfamilycarers276.
Clearstandardsexistforstrokerehabilitation,forinstanceasdescribedbothintheNationalClinical
GuidelineforStrokedevelopedbytheIntercollegiateStrokeWorkingParty122.Thesearereflectedin
theNICEqualitystandards189andtheNationalStrokeStrategy61.Overallthereislittledoubtthat
therehabilitationapproachiseffective;whatindividualinterventionsshouldtakeplacewithinthis
structureislessclear.
Advancesintheneurosciencesincludinggreaterunderstandingofthemechanismsofimpairment
willleadtonoveltreatments.Thereisawealthofevidencesuggestingthatcentralnervoussystem
reorganisationunderliesmuchoftheimprovementinimpairmentthatisfrequentlyseen.
Experimentsshowthatsomeregionsinthenormaladultbrain,particularlythecortex,havethe
capacitytochangestructureandconsequentlyfunctioninresponsetoenvironmentalchange,a
processdescribedasplasticity.Inadditionfunctionallyrelevantadaptivechangeshavebeen
demonstratedfollowingfocaldamagetothebrain.Itissuggestedthatrehabilitationtherapies
interactswiththeseplasticchanges,thusreducingimpairmentviaactivitydependentplastic
NationalClinicalGuidelineCentre,2013.
14
StrokeRehabilitation
Introduction
change.280Examplesofsuchtherapiesalreadyexistinrehabilitationpracticesuchasupperorlower
limbsensorimotorfunctionbytaskrelatedtrainingusingconstraintinducedtherapy173,treadmill
training109,andprismadaptation(toreversevisualneglect)87,109.
Theaimofthisguidelinedevelopmentgroupwastoreviewthestructure,processesand
interventionscurrentlyusedinrehabilitationcare,andtoevaluatewhethertheyimproveoutcomes
forpeoplewithstroke.Suchstudiesarecomplexandresearchmethodologiesneedtoberobust.
Evaluationofclinicaleffectivenessneedsstudiesthathaverobusttheoreticalunderpinnings,capture
changesthatarerelevanttothetreatmentevaluatedandreflectwhatisimportanttopatients,and
belargeenoughtoallowreliabledatainterpretation.Thisguidelinereviewssomeoftheavailable
interventionsthatcanbeusedinstrokerehabilitation,andhighlightswheretherearegapsinthe
evidence.Itisnotintendedtobecomprehensive.
Allinterventionsshouldtakeplaceinthecontextofacomprehensivestrokepathwaywhich
recognisesthatearlymanagement,whilecritical,isacomponentofaprocesswhichaimsto
amelioratethelongtermconsequencesoflivingwithstrokeforindividualsandtheirfamiliesandto
enablethemtoliveathome,abletoparticipateinasmanyactivitiesastheyareable.Atthepointof
dischargethepersonwhohashadastrokemayneedsupportfromarangeofotheragenciessuchas
housing,JobcentrePlus,socialservicesandstrokevoluntaryorganisations.Randomisedcontrolled
trialevidence,althoughthegoldstandardforinterventionstudiesmaynotbeavailableor
appropriateforexaminingrehabilitationprocesses.AmodifiedDelphisurveywasconductedto
obtainformalconsensusaroundareassuchasservicedeliveryandcareplanning.Itneedstobe
recognisedthatevenwheretheevidencebaseisclear,rehabilitationinterventionsneedtobe
targetedandrelevanttotheindividual.Someindividualsmaydeclinetreatmentwhichhealthcare
professionalsseeasimportant.Insuchcircumstancesissuessuchascapacityandconsentneedtobe
considered108.
NationalClinicalGuidelineCentre,2013.
15
StrokeRehabilitation
Developmentoftheguideline
2 Developmentoftheguideline
2.1 WhatisaNICEclinicalguideline?
NICEclinicalguidelinesarerecommendationsforthecareofindividualsinspecificclinicalconditions
orcircumstanceswithintheNHSfrompreventionandselfcarethroughprimaryandsecondary
caretomorespecialisedservices.Webaseourclinicalguidelinesonthebestavailableresearch
evidence,withtheaimofimprovingthequalityofhealthcare.Weusepredeterminedand
systematicmethodstoidentifyandevaluatetheevidencerelatingtospecificreviewquestions.
NICEclinicalguidelinescan:
providerecommendationsforthetreatmentandcareofpeoplebyhealthprofessionals
beusedtodevelopstandardstoassesstheclinicalpracticeofindividualhealthprofessionals
beusedintheeducationandtrainingofhealthprofessionals
helppatientstomakeinformeddecisions
improvecommunicationbetweenpatientandhealthprofessional
Whileguidelinesassistthepracticeofhealthcareprofessionals,theydonotreplacetheirknowledge
andskills.
Weproduceourguidelinesusingthefollowingsteps:
GuidelinetopicisreferredtoNICEfromtheDepartmentofHealth
Stakeholdersregisteraninterestintheguidelineandareconsultedthroughoutthedevelopment
process
ThescopeispreparedbytheNationalClinicalGuidelineCentre(NCGC)
TheNCGCestablishesaguidelinedevelopmentgroup
Adraftguidelineisproducedafterthegroupassessestheavailableevidenceandmakes
recommendations
Thereisaconsultationonthedraftguideline
Thefinalguidelineisproduced
TheNCGCandNICEproduceanumberofversionsofthisguideline:
thefullguidelinecontainsalltherecommendations,plusdetailsofthemethodsusedandthe
underpinningevidence
theNICEguidelineliststherecommendations
theNICEPathwayisanonlinetoolforhealthprofessionalsthatbringstogetherthe
recommendationsfromthisguidanceandallrelatedNICEguidance.
informationforthepublic(understandingNICEguidanceorUNG)iswrittenusingsuitable
languageforpeoplewithoutspecialistmedicalknowledge
Thisversionisthefullversion.TheotherversionscanbedownloadedfromNICEatwww.nice.org.uk
2.2 Remit
NICEreceivedtheremitforthisguidelinefromtheDepartmentofHealth.Theycommissionedthe
NCGCtoproducetheguideline.
Theremitforthisguidelineis:toproduceajointclinicalandsocialcareguidelineonthelongterm
rehabilitationandsupportofstrokepatients.
NationalClinicalGuidelineCentre,2013.
16
StrokeRehabilitation
Developmentoftheguideline
2.3 Whodevelopedthisguideline?
AmultidisciplinaryGuidelineDevelopmentGroup(GDG)comprisingprofessionalgroupmembersand
consumerrepresentativesofthemainstakeholdersdevelopedthisguideline(seesectionon
GuidelineDevelopmentGroupMembershipandacknowledgements).
TheNationalInstituteforHealthandClinicalExcellencefundstheNationalClinicalGuidelineCentre
(NCGC)andthussupportedthedevelopmentofthisguideline.TheGDGwasconvenedbytheNCGC
andchairedbyDrDianePlayfordinaccordancewithguidancefromtheNationalInstituteforHealth
andClinicalExcellence(NICE).
Thegroupmetapproximatelyevery5weeksduringthedevelopmentoftheguideline.Atthestartof
theguidelinedevelopmentprocessallGDGmembersdeclaredinterestsincludingconsultancies,fee
paidwork,shareholdings,fellowshipsandsupportfromthehealthcareindustry.Atallsubsequent
GDGmeetings,membersdeclaredarisingconflictsofinterest,whichwerealsorecorded(Appendix
[C]).
Memberswereeitherrequiredtowithdrawcompletelyorforpartofthediscussioniftheirdeclared
interestmadeitappropriate.Thedetailsofdeclaredinterestsandtheactionstakenareshownin
Appendix[C].
StafffromtheNCGCprovidedmethodologicalsupportandguidanceforthedevelopmentprocess.
Theteamworkingontheguidelineincludedaprojectmanager,systematicreviewers,health
economistsandinformationscientists.Theyundertooksystematicsearchesoftheliterature,
appraisedtheevidence,conductedmetaanalysisandcosteffectivenessanalysiswhereappropriate
anddraftedtheguidelineincollaborationwiththeGDG.
2.4 Whatthisguidelinecovers
Theguidelinecoversadultsandyoungpeople16orolderwhohavehadastrokeandhavecontinuing
impairment(2weeksormorepoststroke),limitedactivityorparticipationrestriction.
Theclinicalareascoveredincluded:therapiestoimprovephysical,cognitiveandspeechfunctions,
activitiesofdailylivingandvocationalrehabilitation,interventionstoaddressdysphagiaandvisual
fieldloss,informationandsupportforpatientsandcarers,earlysupporteddischargeandintensityof
rehabilitationtherapy.Theinterventionsconsideredandthesubsequentrecommendationsmade
arenotsettingspecificandincludehealthorsocialcareservices.
ForfurtherdetailspleaserefertothescopeinAppendixAandreviewquestionsinAppendixE.
2.5 Whatthisguidelinedoesnotcover
Childrenunder16yearsandpeoplewhohadhadatransientischaemicattackwerenotincluded.The
guidelinedidnotconsiderprimaryorsecondarypreventionofstroke,acutestrokeorassessmentfor
rehabilitation.
2.6 RelationshipsbetweentheguidelineandotherNICEguidance
RelatedNICEInterventionalProcedures:
Electricalstimulationfordropfootofcentralneurologicalorigin.NICEinterventionalprocedure
guidance278(2009).Availablefromwww.nice.org.uk/guidance/IPG278
NationalClinicalGuidelineCentre,2013.
17
StrokeRehabilitation
Developmentoftheguideline
RelatedNICEClinicalGuidelines:
Depressioninadults(update).NICEclinicalguidelineCG90(2009).Availablefrom:
http://publications.nice.org.uk/depressioninadultscg90.
Depressioninadultswithachronicphysicalhealthproblem:Treatmentandmanagement.NICE
clinicalguidelineCG91(2009).Availablefrom:http://publications.nice.org.uk/depressioninadults
withachronicphysicalhealthproblemcg91.
Faecalincontinence:ThemanagementoffaecalincontinenceinadultsNICEclinicalguidelineCG49
(2007).Availablefrom:http://publications.nice.org.uk/faecalincontinencecg49.
Falls:theassessmentandpreventionoffallsinolderpeople.NICEclinicalguidelineCG21(2004)
http://publications.nice.org.uk/fallscg21.
Generalisedanxietydisorderandpanicdisorder(withorwithoutagoraphobia)inadults:
Managementinprimary,secondaryandcommunitycare.NICEclinicalguidelineCG113(2011).
Availablefrom:http://publications.nice.org.uk/generalisedanxietydisorderandpanicdisorder
withorwithoutagoraphobiainadultscg113.
Neuropathicpain:Thepharmacologicalmanagementofneuropathicpaininadultsinnonspecialist
settingsNICEclinicalguidelineCG96(2010).http://publications.nice.org.uk/neuropathicpaincg96.
Nutritionsupportinadults:Oralnutritionsupport,enteraltubefeedingandparenteralnutrition.
NICEclinicalguidelineCG32(2006).Availablefrom:http://publications.nice.org.uk/nutrition
supportinadultscg32.
PatientexperienceinadultNHSservices:improvingtheexperienceofcareforpeopleusingadult
NHSservices.NICEclinicalguidelineCG138(2012)http://publications.nice.org.uk/patient
experienceinadultnhsservicesimprovingtheexperienceofcareforpeopleusingadultcg138.
Stroke:Diagnosisandinitialmanagementofacutestrokeandtransientischaemicattack(TIA).NICE
clinicalguidelineCG68(2008).Availablefrom:http://publications.nice.org.uk/strokecg68.
Urinaryincontinenceinneurologicaldisease:managementoflowerurinarytractdysfunctionin
neurologicaldisease.NICEclinicalguidelineCG148(2012).Availablefrom:
http://guidance.nice.org.uk/CG148.
Medicinesadherence:involvingpatientsindecisionsaboutprescribedmedicinesandsupporting
adherence.NICEclinicalguidelineCG76(2009).Availablefrom:http://www.nice.org.uk/CG76
Lipidmodification:Cardiovascularriskassessmentandthemodificationofbloodlipidsforthe
primaryandsecondarypreventionofcardiovasculardisease.NICEclinicalguidelineCG67(2008).
Availablefrom:http://www.nice.org.uk/CG67.
Hypertension:clinicalmanagementofprimaryhypertensioninadults.NICEclinicalguidelineCG127
(2011):Availablefrom:http://guidance.nice.org.uk/CG127.
Type2Diabetes:themanagementoftype2diabetes(update).NICEclinicalguidelineCG87(2009):
Availablefrom:http://www.nice.org.uk/CG87.
Atrialfibrillation.NICEclinicalguidelineCG36(2006):Availablefrom:http://www.nice.org.uk/CG36
RelatedNICEPublicHealthGuidance:
Managementoflongtermsicknessandincapacityforwork:Guidanceforprimarycareand
employersonthemanagementoflongtermsicknessandincapacity.NICEpublichealthguidance19
(2009).Availablefrom:www.nice.org.uk/guidance/PH19.
NationalClinicalGuidelineCentre,2013.
18
StrokeRehabilitation
Developmentoftheguideline
NICERelatedGuidancecurrentlyindevelopment:
Falls(update)NICEclinicalguideline(publicationexpectedJune2013).
Lipidmodification(update).NICEclinicalguideline(publicationTBC).
Neuropathicpain:pharmacologicalmanagementinadultsinnonspecialistsettings.NICEclinical
guideline(publicationexpectedAugust2013).
Type2diabetesNICEclinicalguideline(publicationTBC).
Oralhealth:innursingandresidentialcareNICEpublichealthguidance(publicationTBC).
Workplacehealth:employeeswithchronicdiseasesandlongtermconditionsNICEpublichealth
guidance(publicationTBC).
NationalClinicalGuidelineCentre,2013.
19
StrokeRehabilitation
Guidelinesummary
3 Guidelinesummary
3.1 Keyprioritiesforimplementation
TheGDGidentifiedkeyprioritiesforimplementation.Theyselectedrecommendationsthatwould:
Haveahighimpactonoutcomesthatareimportanttopatients
Haveahighimpactonreducingvariationincareandoutcomes
LeadtoamoreefficientuseofNHSresources
Promotepatientchoice
IndoingthistheGDGalsoconsideredwhichrecommendationswereparticularlylikelytobenefit
fromimplementationsupport.Theconsideredwhetherarecommendation:
Requireschangesinservicedelivery
Requiresretrainingofprofessionalsorthedevelopmentofnewskillsandcompetencies
Affectsandneedstobeimplementedacrossvariousagenciesorsettings
Maybeviewedaspotentiallycontentiousordifficulttoimplementforotherreasons
Thefollowingrecommendationshavebeenidentifiedasprioritiesforimplementation.
3.1.1
Strokeunits
1. Peoplewithdisabilityafterstrokeshouldreceiverehabilitationinadedicatedstrokeinpatient
unitandsubsequentlyfromaspecialiststroketeamwithinthecommunity.
3.1.2
Thecoremultidisciplinarystroketeam
2. Acoremultidisciplinarystrokerehabilitationteamshouldcomprisethefollowingprofessionals
withexpertiseinstrokerehabilitation:
o consultantphysicians
o nurses
o physiotherapists
o occupationaltherapists
o speechandlanguagetherapists
o clinicalpsychologists
o rehabilitationassistants
o socialworkers.
3.1.3
Healthandsocialcareinterface
3. Healthandsocialcareprofessionalsshouldworkcollaborativelytoensureasocialcare
assessmentiscarriedoutpromptly,whereneeded,beforethepersonwithstrokeistransferred
fromhospitaltothecommunity.Theassessmentshould:
o identifyanyongoingneedsofthepersonandtheirfamilyorcarer,forexample,accessto
benefits,careneeds,housing,communityparticipation,returntowork,transportandaccess
tovoluntaryservices
o bedocumentedandallneedsrecordedinthepersonshealthandsocialcareplan,withacopy
providedtothepersonwithstroke.
3.1.4
Transferofcarefromhospitaltocommunity
4. Offerearlysupporteddischargetopeoplewithstrokewhoareabletotransferfrombedtochair
independentlyorwithassistance,aslongasasafeandsecureenvironmentcanbeprovided.
NationalClinicalGuidelineCentre,2013.
20
StrokeRehabilitation
Guidelinesummary
3.1.5
Settinggoalsforrehabilitation
5. Ensurethatgoalsettingmeetingsduringstrokerehabilitation:
o aretimetabledintotheworkingweek
o involvethepersonwithstrokeand,whereappropriate,theirfamilyorcarerinthediscussion.
3.1.6
Intensityofstrokerehabilitation
6. Offerinitiallyatleast45minutesofeachrelevantstrokerehabilitationtherapyforaminimumof
5daysperweektopeoplewhohavetheabilitytoparticipate,andwherefunctionalgoalscanbe
achieved.Ifmorerehabilitationisneededatalaterstage,tailortheintensitytothepersons
needsatthattimea.
3.1.7
Cognitivefunctioning
7. Screenpeopleafterstrokeforcognitivedeficits.Whereacognitivedeficitisidentified,carryouta
detailedassessmentusingvalid,reliableandresponsivetoolsbeforedesigningatreatment
programme.
3.1.8
Emotionalfunctioning
8. Assessemotionalfunctioninginthecontextofcognitivedifficultiesinpeopleafterstroke.Any
interventionchosenshouldtakeintoconsiderationthetypeorcomplexityofthepersons
neuropsychologicalpresentationandrelevantpersonalhistory.
3.1.9
Swallowing
9. Offerswallowingtherapyatleast3timesaweektopeoplewithdysphagiaafterstrokewhoare
abletoparticipate,foraslongastheycontinuetomakefunctionalgains.Swallowingtherapy
couldincludecompensatorystrategies,exercisesandposturaladvice.
3.1.10
Returntowork
10.Returntoworkissuesshouldbeidentifiedassoonaspossibleafterthepersonsstroke,reviewed
regularlyandmanagedactively.Activemanagementshouldinclude:
o identifyingthephysical,cognitive,communicationandpsychologicaldemandsofthejob(for
example,multitaskingbyansweringemailsandtelephonecallsinabusyoffice)
o identifyinganyimpairmentsonworkperformance(forexample,physicallimitations,anxiety,
fatiguepreventingattendanceforafulldayatwork,cognitiveimpairmentspreventingmulti
tasking,andcommunicationdeficits)
o tailoringanintervention(forexample,teachingstrategiestosupportmultitaskingormemory
difficulties,teachingtheuseofvoiceactivatedsoftwareforpeoplewithdifficultytyping,and
deliveryofworksimulations)
o educatingabouttheEqualityAct2010bandsupportavailable(forexample,anaccesstowork
scheme)
o workplacevisitsandliaisonwithemployerstoestablishreasonableaccommodations,suchas
provisionofequipmentandgradedreturntowork.
Intensityoftherapyfordysphagia,providedaspartofspeechandlanguagetherapyisaddressedin
recommendation58.
HMGovernment(2010)EqualityAct[online]
NationalClinicalGuidelineCentre,2013.
21
StrokeRehabilitation
Guidelinesummary
3.1.11
Longtermhealthandsocialsupport
o Reviewthehealthandsocialcareneedsofpeopleafterstrokeandtheneedsoftheircarersat
6monthsandannuallythereafter.Thesereviewsshouldcoverparticipationandcommunity
rolestoensurethatpeoplesgoalsareaddressed.
3.2 Fulllistofrecommendations
1.
Peoplewithdisabilityafterstrokeshouldreceiverehabilitationinadedicated
strokeinpatientunitandsubsequentlyfromaspecialiststroketeamwithin
thecommunity.
2.
Aninpatientstrokerehabilitationserviceshouldconsistofthefollowing:
3.
4.
adedicatedstrokerehabilitationenvironment
acoremultidisciplinaryteam(seerecommendation3)whohavethe
knowledge,skillsandbehaviourstoworkinpartnershipwithpeople
withstrokeandtheirfamiliesandcarerstomanagethechanges
experiencedasaresultofastroke.
accesstootherservicesthatmaybeneeded,forexample:
continenceadvice
dietetics
electronicaids(forexample,remotecontrolsfordoors,lightsand
heating,andcommunicationaids)
liaisonpsychiatry
orthoptics
orthotics
pharmacy
podiatry
wheelchairservices
amultidisciplinaryeducationprogramme.
Acoremultidisciplinarystrokerehabilitationteamshouldcomprisethe
followingprofessionalswithexpertiseinstrokerehabilitation:
consultantphysicians
nurses
physiotherapists
occupationaltherapists
speechandlanguagetherapists
clinicalpsychologists
rehabilitationassistants
socialworkers.
Throughoutthecarepathway,therolesandresponsibilitiesofthecore
multidisciplinarystrokerehabilitationteamshouldbeclearlydocumented
andcommunicatedtothepersonandtheirfamilyorcarer.
NationalClinicalGuidelineCentre,2013.
22
StrokeRehabilitation
Guidelinesummary
5.
Membersofthecoremultidisciplinarystroketeamshouldscreentheperson
withstrokeforarangeofimpairmentsanddisabilities,inordertoinformand
directfurtherassessmentandtreatment.
6.
Healthandsocialcareprofessionalsshouldworkcollaborativelytoensurea
socialcareassessmentiscarriedoutpromptly,whereneeded,beforethe
personwithstrokeistransferredfromhospitaltothecommunity.The
assessmentshould:
7.
identifyanyongoingneedsofthepersonandtheirfamilyorcarer,for
example,accesstobenefits,careneeds,housing,community
participation,returntowork,transportandaccesstovoluntary
services.
bedocumentedandallneedsrecordedinthepersonshealthandsocial
careplan,withacopyprovidedtothepersonwithstroke.
Offertrainingincare(forexample,inmovingandhandlingandhelpingwith
dressing)tofamilymembersorcarerswhoarewillingandabletobeinvolved
insupportingthepersonaftertheirstroke.
Reviewfamilymembersandcarerstrainingandsupportneedsregularly
(asaminimumatthepersons6monthandannualreviews),
acknowledgingthattheseneedsmaychangeovertime.
8.
Offerearlysupporteddischargetopeoplewithstrokewhoareableto
transferfrombedtochairindependentlyorwithassistance,aslongasasafe
andsecureenvironmentcanbeprovided.
9.
Earlysupporteddischargeshouldbepartofaskilledstrokerehabilitation
serviceandshouldconsistofthesameintensityoftherapyandrangeof
multidisciplinaryskillsavailableinhospital.Itshouldnotresultinadelayin
deliveryofcare.
10.
Hospitalsshouldhavesystemsinplacetoensurethat:
peopleafterstrokeandtheirfamiliesandcarers(asappropriate)are
involvedinplanningfortransferofcare,andcarersreceivetrainingin
care(forexample,inmovingandhandlingandhelpingwithdressing)
peopleafterstrokeandtheirfamiliesandcarersfeeladequately
informed,preparedandsupported
GPsandotherappropriatepeopleareinformedbeforetransferofcare
anagreedhealthandsocialcareplanisinplace,andthepersonknows
whomtocontactifdifficultiesarise
appropriateequipment(includingspecialistseatingandawheelchairif
needed)isinplaceatthepersonsresidence,regardlessofsetting.
11.
Beforetransferfromhospitaltohomeortoacaresetting,discussandagree
ahealthandsocialcareplanwiththepersonwithstrokeandtheirfamilyor
carer(asappropriate),andprovidethistoallrelevanthealthandsocialcare
providers.
12.
Beforetransferofcarefromhospitaltohomeforpeoplewithstroke:
establishthattheyhaveasafeandenablinghomeenvironment,for
example,checkthatappropriateequipmentandadaptationshave
beenprovidedandthatcarersaresupportedtofacilitate
independence,and
NationalClinicalGuidelineCentre,2013.
23
StrokeRehabilitation
Guidelinesummary
13.
undertakeahomevisitwiththemunlesstheirabilitiesandneedscanbe
identifiedinotherways,forexample,bydemonstrating
independenceinallselfcareactivities,includingmealpreparation,
whileintherehabilitationunit.
Ontransferofcarefromhospitaltothecommunity,provideinformationto
allrelevanthealthandsocialcareprofessionalsandthepersonwithstroke.
Thisshouldinclude:
asummaryofrehabilitationprogressandcurrentgoals
diagnosisandhealthstatus
functionalabilities(includingcommunicationneeds)
careneeds,includingwashing,dressing,helpwithgoingtothetoiletand
eating
psychological(cognitiveandemotional)needs
medicationneeds(includingthepersonsabilitytomanagetheir
prescribedmedicationsandanysupporttheyneedtodoso)
socialcircumstances,includingcarersneeds
mentalcapacityregardingthetransferdecision
managementofrisk,includingtheneedsofvulnerableadults
plansforfollowup,rehabilitationandaccesstohealthandsocialcare
andvoluntarysectorservices.
14.
Ensurethatpeoplewithstrokewhoaretransferredfromhospitaltocare
homesreceiveassessmentandtreatmentfromstrokerehabilitationand
socialcareservicestothesamestandardsastheywouldreceiveintheirown
homes.
15.
Localhealthandsocialcareprovidersshouldhavestandardoperating
procedurestoensurethesafetransferandlongtermcareofpeopleafter
stroke,includingthoseincarehomes.Thisshouldincludetimelyexchangeof
informationbetweendifferentprovidersusinglocalprotocols.
16.
Aftertransferofcarefromhospital,peoplewithdisabilitiesafterstroke
(includingpeopleincarehomes)shouldbefollowedupwithin72hoursby
thespecialiststrokerehabilitationteamforassessmentofpatientidentified
needsandthedevelopmentofsharedmanagementplans.
17.
Provideadviceonprescribedmedicationsforpeopleafterstrokeinlinewith
recommendationsinMedicinesadherence(NICEclinicalguideline76).
18.
Onadmissiontohospital,toensuretheimmediatesafetyandcomfortofthe
personwithstroke,screenthemforthefollowingand,ifproblemsare
identified,startmanagementassoonaspossible:
orientation
positioning,movingandhandling
swallowing
transfers(forexample,frombedtochair)
pressurearearisk
continence
NationalClinicalGuidelineCentre,2013.
24
StrokeRehabilitation
Guidelinesummary
communication,includingtheabilitytounderstandandfollow
instructionsandtoconveyneedsandwishes
nutritionalstatusandhydration(followtherecommendationsinStroke
[NICEclinicalguideline68]andNutritionsupportinadults[NICE
clinicalguideline32]).
19.
Performafullmedicalassessmentofthepersonwithstroke,including
cognition(attention,memory,spatialawareness,apraxia,perception),vision,
hearing,tone,strength,sensationandbalance.
20.
Acomprehensiveassessmentofapersonwithstrokeshouldtakeinto
account:
21.
theirpreviousfunctionalabilities
impairmentofpsychologicalfunctioning(cognitive,emotionaland
communication)
impairmentofbodyfunctions,includingpain
activitylimitationsandparticipationrestrictions
environmentalfactors(social,physicalandcultural).
Informationcollectedroutinelyfrompeoplewithstrokeusingvalid,reliable
andresponsivetoolsshouldincludethefollowingonadmissionand
discharge:
NationalInstitutesofHealthStrokeScale
BarthelIndex.
22.
Informationcollectedfrompeoplewithstrokeusingvalid,reliableand
responsivetoolsshouldbefedbacktothemultidisciplinaryteamregularly.
23.
Takeintoconsiderationtheimpactofthestrokeonthepersonsfamily,
friendsand/orcarersand,ifappropriate,identifysourcesofsupport.
24.
Informthefamilymembersandcarersofpeoplewithstrokeabouttheirright
tohaveacarersneedsassessment.
25.
Ensurethatpeoplewithstrokehavegoalsfortheirrehabilitationthat:
26.
27.
aremeaningfulandrelevanttothem
focusonactivityandparticipation
arechallengingbutachievable
includebothshorttermandlongtermelements.
Ensurethatgoalsettingmeetingsduringstrokerehabilitation:
aretimetabledintotheworkingweek
involvethepersonwithstrokeand,whereappropriate,theirfamilyor
carerinthediscussion.
Ensurethatduringgoalsettingmeetings,peoplewithstrokeareprovided
with:
anexplanationofthegoalsettingprocess
theinformationtheyneedinaformatthatisaccessibletothem
thesupporttheyneedtomakedecisionsandtakeanactivepartin
settinggoals.
NationalClinicalGuidelineCentre,2013.
25
StrokeRehabilitation
Guidelinesummary
28.
Givepeoplecopiesoftheiragreedgoalsforstrokerehabilitationaftereach
goalsettingmeeting.
29.
Reviewpeoplesgoalsatregularintervalsduringtheirstrokerehabilitation.
30.
Provideinformationandsupporttoenablethepersonwithstrokeandtheir
familyorcarer(asappropriate)toactivelyparticipateinthedevelopmentof
theirstrokerehabilitationplan.
31.
Strokerehabilitationplansshouldbereviewedregularlybythe
multidisciplinaryteam.Timethesereviewsaccordingtothestageof
rehabilitationandthepersonsneeds.
32.
Documentationaboutthepersonsstrokerehabilitationshouldbe
individualised,andshouldincludethefollowinginformationasaminimum:
basicdemographics,includingcontactdetailsandnextofkin
diagnosisandrelevantmedicalinformation
listofcurrentmedications,includingallergies
standardisedscreeningassessments(seerecommendation18)
thepersonsrehabilitationgoals
multidisciplinaryprogressnotes
akeycontactfromthestrokerehabilitationteam(includingtheircontact
details)tocoordinatethepersonshealthandsocialcareneeds
dischargeplanninginformation(includingaccommodationneeds,aids
andadaptations)
jointhealthandsocialcareplans,ifdeveloped
followupappointments.
33.
Offerinitiallyatleast45minutesofeachrelevantstrokerehabilitation
therapyforaminimumof5daysperweektopeoplewhohavetheabilityto
participate,andwherefunctionalgoalscanbeachieved.Ifmore
rehabilitationisneededatalaterstage,tailortheintensitytothepersons
needsatthattimec.
34.
Considermorethan45minutesofeachrelevantstrokerehabilitationtherapy
5daysperweekforpeoplewhohavetheabilitytoparticipateandcontinue
tomakefunctionalgains,andwherefunctionalgoalscanbeachieved.
35.
Ifpeoplewithstrokeareunabletoparticipatein45minutesofeach
rehabilitationtherapy,ensurethattherapyisstilloffered5daysperweekfor
ashortertimeatanintensitythatallowsthemtoactivelyparticipate.
36.
Workingwiththepersonwithstrokeandtheirfamilyorcarer,identifytheir
informationneedsandhowtodeliverthem,takingintoaccountspecific
impairmentssuchasaphasiaandcognitiveimpairments.Pacethe
informationtothepersonsemotionaladjustment.
37.
Provideinformationaboutlocalresources(forexample,leisure,housing,
socialservicesandthevoluntarysector)thatcanhelptosupporttheneeds
andprioritiesofthepersonwithstrokeandtheirfamilyorcarer.
Intensityoftherapyfordysphagia,providedaspartofspeechandlanguagetherapyisaddressedin
recommendation58.
NationalClinicalGuidelineCentre,2013.
26
StrokeRehabilitation
Guidelinesummary
38.
Reviewinformationneedsatthepersons6monthandannualstroke
reviewsandatthestartandcompletionofanyinterventionperiod.
39.
NICEhasproducedguidanceonthecomponentsofgoodpatientexperience
inadultNHSservices.FollowtherecommendationsinPatientexperiencein
adultNHSservices(NICEclinicalguideline138)d.
40.
Screenpeopleafterstrokeforcognitivedeficits.Whereacognitivedeficitis
identified,carryoutadetailedassessmentusingvalid,reliableand
responsivetoolsbeforedesigningatreatmentprogramme.
41.
Provideeducationandsupportforpeoplewithstrokeandtheirfamiliesand
carerstohelpthemunderstandtheextentandimpactofcognitivedeficits
afterstroke,recognisingthatthesemayvaryovertimeandindifferent
settings.
42.
Assesstheeffectofvisualneglectafterstrokeonfunctionaltaskssuchas
mobility,dressing,eatingandusingawheelchair,usingstandardised
assessmentsandbehaviouralobservation.
43.
Useinterventionsforvisualneglectafterstrokethatfocusontherelevant
functionaltasks,takingintoaccounttheunderlyingimpairment.Forexample:
interventionstohelppeoplescantotheneglectedside,suchasbrightly
colouredlinesorhighlighterontheedgeofthepage
alertingtechniquessuchasauditorycues
repetitivetaskperformancesuchasdressing
alteringtheperceptualinputusingprismglasses.
44.
Assessmemoryandotherrelevantdomainsofcognitivefunctioning(suchas
executivefunctions)inpeopleafterstroke,particularlywhereimpairmentsin
memoryaffecteverydayactivity.
45.
Useinterventionsformemoryandcognitivefunctionsafterstrokethatfocus
ontherelevantfunctionaltasks,takingintoaccounttheunderlying
impairment.Interventionscouldinclude:
increasingawarenessofthememorydeficit
enhancinglearningusingerrorlesslearningandelaborativetechniques
(makingassociations,useofmnemonics,internalstrategiesrelated
toencodinginformationsuchaspreview,question,read,state,test)
externalaids(forexample,diaries,lists,calendarsandalarms)
environmentalstrategies(routinesandenvironmentalprompts).
46.
Assessattentionandcognitivefunctionsinpeopleafterstrokeusing
standardisedassessments.Usebehaviouralobservationtoevaluatethe
impactoftheimpairmentonfunctionaltasks.
47.
Considerattentiontrainingforpeoplewithattentiondeficitsafterstroke.
48.
Useinterventionsforattentionandcognitivefunctionsafterstrokethatfocus
ontherelevantfunctionaltasks.Forexample,usegenerictechniquessuchas
managingtheenvironmentandprovidingpromptsrelevanttothefunctional
task.
Forrecommendationsoncontinuityofcareandrelationshipsseesection1.4andforrecommendationson
enablingpatientstoactivelyparticipateintheircareseesection1.5.
NationalClinicalGuidelineCentre,2013.
27
StrokeRehabilitation
Guidelinesummary
49.
Assessemotionalfunctioninginthecontextofcognitivedifficultiesinpeople
afterstroke.Anyinterventionchosenshouldtakeintoconsiderationthetype
orcomplexityofthepersonsneuropsychologicalpresentationandrelevant
personalhistory.
50.
Supportandeducatepeopleafterstrokeandtheirfamiliesandcarers,in
relationtoemotionaladjustmenttostroke,recognisingthatpsychological
needsmaychangeovertimeandindifferentsettings.
51.
Whenneworpersistingemotionaldifficultiesareidentifiedatthepersons6
monthorannualstrokereviews,referthemtoappropriateservicesfor
detailedassessmentandtreatment.
52.
Managedepressionoranxietyinpeopleafterstrokewhohavenocognitive
impairmentinlinewithrecommendationsinDepressioninadultswitha
chronicphysicalhealthproblem(NICEclinicalguideline91)andGeneralised
anxietydisorder(NICEclinicalguideline113).
53.
Screenpeopleafterstrokeforvisualdifficulties.
54.
Offereyemovementtherapytopeoplewhohavepersistinghemianopiaafter
strokeandwhoareawareofthecondition.
55.
Whenadvisingpeoplewithvisualproblemsafterstrokeaboutdriving,
consulttheDriverandVehicleLicensingAgency(DVLA)regulations.
56.
Referpeoplewithpersistingdoublevisionafterstrokeforformalorthoptic
assessment.
57.
Assessswallowinginpeopleafterstrokeinlinewithrecommendationsin
Stroke(NICEclinicalguideline68).
58.
Offerswallowingtherapyatleast3timesaweektopeoplewithdysphagia
afterstrokewhoareabletoparticipate,foraslongastheycontinuetomake
functionalgains.Swallowingtherapycouldincludecompensatorystrategies,
exercisesandposturaladvice.
59.
Ensurethateffectivemouthcareisgiventopeoplewithdifficultyswallowing
afterstroke,inordertodecreasetheriskofaspirationpneumonia.
60.
Healthcareprofessionalswithrelevantskillsandtraininginthediagnosis,
assessmentandmanagementofswallowingdisordersshouldregularly
monitorandreassesspeoplewithdysphagiaafterstrokewhoarehaving
modifiedfoodandliquiduntiltheyarestable(thisrecommendationisfrom
Nutritionsupportinadults[NICEclinicalguideline32]).
61.
Providenutritionsupporttopeoplewithdysphagiainlinewith
recommendationsinNutritionsupportinadults(NICEclinicalguideline32)
andStroke(NICEclinicalguideline68).
62.
Screenpeopleafterstrokeforcommunicationdifficultieswithin72hoursof
onsetofstrokesymptoms.
63.
Eachstrokerehabilitationserviceshoulddeviseastandardisedprotocolfor
screeningforcommunicationdifficultiesinpeopleafterstroke.
64.
Provideappropriateinformation,educationandtrainingtothe
multidisciplinarystroketeamtoenablethemtosupportandcommunicate
effectivelywiththepersonwithcommunicationdifficultiesandtheirfamily
orcarer.
65.
Speechandlanguagetherapyforpeoplewithstrokeshouldbeledand
supervisedbyaspecialistspeechandlanguagetherapistworking
NationalClinicalGuidelineCentre,2013.
28
StrokeRehabilitation
Guidelinesummary
collaborativelywithotherappropriatelytrainedpeopleforexample,speech
andlanguagetherapyassistants,carersanfriends,andmembersofthe
voluntarysector.
66.
Provideopportunitiesforpeoplewithcommunicationdifficultiesafterstroke
tohaveconversationandsocialenrichmentwithpeoplewhohavethe
training,knowledge,skillsandbehaviourstosupportcommunication.This
shouldbeinadditiontotheopportunitiesprovidedbyfamilies,carersand
friends.
67.
Speechandlanguagetherapistsshouldassesspeoplewithlimitedfunctional
communicationafterstrokefortheirpotentialtobenefitfromusinga
communicationaidorothertechnologies(forexample,homebased
computertherapiesorsmartphoneapplications).
68.
Providecommunicationaidsforthosepeopleafterstrokewhohavethe
potentialtobenefit,andoffertraininginhowtousethem.
69.
Tellthepersonwithcommunicationdifficultiesafterstrokeabout
communitybasedcommunicationandsupportgroups(suchasthose
providedbythevoluntarysector)andencouragethemtoparticipate.
70.
Whenpersistingcommunicationdifficultiesareidentifiedatthepersons6
monthorannualstrokereviews,referthembacktoaspeechandlanguage
therapistfordetailedassessment,andoffertreatmentifthereispotentialfor
functionalimprovement.
71.
Makesurethatallwritteninformation(includingthatrelatingtomedical
conditionsandtreatment)isadaptedforpeoplewithaphasiaafterstroke.
Thisshouldinclude,forexample,appointmentletters,rehabilitation
timetablesandmenus.
72.
Helpandenablepeoplewithcommunicationdifficultiesafterstroketo
communicatetheireverydayneedsandwishes,andsupportthemto
understandandparticipateinbotheverydayandmajorlifedecisions.
73.
Ensurethatenvironmentalbarrierstocommunicationareminimisedfor
peopleafterstroke.Forexample,makesuresignageisclearandbackground
noiseisminimised.
74.
Referpeoplewithsuspectedcommunicationdifficultiesafterstroketoa
speechandlanguagetherapistfordetailedanalysisofspeechandlanguage
impairmentsandassessmentoftheirimpact.
75.
Speechandlanguagetherapistsshould:
providedirectimpairmentbasedtherapyforcommunication
impairments(forexample,aphasiaordysarthria)
helpthepersonwithstroketouseandenhancetheirremaining
languageandcommunicationabilities
teachothermethodsofcommunicating,suchasgestures,writingand
usingcommunicationprops
coachpeoplearoundthepersonwithstroke(includingfamilymembers,
carersandhealthandsocialcarestaff)todevelopsupportive
communicationskillstomaximisethepersonscommunication
potential
helpthepersonwithaphasiaordysarthriaandtheirfamilyorcarerto
adjusttoacommunicationimpairment
NationalClinicalGuidelineCentre,2013.
29
StrokeRehabilitation
Guidelinesummary
supportthepersonwithcommunicationdifficultiestorebuildtheir
identity
supportthepersontoaccessinformationthatenablesdecisionmaking.
76.
Offertrainingincommunicationskills(suchasslowingdown,not
interrupting,usingcommunicationprops,gestures,drawing)tothe
conversationpartnersofpeoplewithaphasiaafterstroke.
77.
Providephysiotherapyforpeoplewhohaveweaknessintheirtrunkorupper
orlowerlimb,sensorydisturbanceorbalancedifficultiesafterstrokethat
haveaneffectonfunction.
78.
Peoplewithmovementdifficultiesafterstrokeshouldbetreatedby
physiotherapistswhohavetherelevantskillsandtraininginthediagnosis,
assessmentandmanagementofmovementinpeoplewithstroke.
79.
Treatmentforpeoplewithmovementdifficultiesafterstrokeshould
continueuntilthepersonisabletomaintainorprogressfunctioneither
independentlyorwithassistancefromothers(forexample,rehabilitation
assistants,familymembers,carersorfitnessinstructors).
80.
Considerstrengthtrainingforpeoplewithmuscleweaknessafterstroke.This
couldincludeprogressivestrengthbuildingthroughincreasingrepetitionsof
bodyweightactivities(forexample,sittostandrepetitions),weights(for
example,progressiveresistanceexercise),orresistanceexerciseonmachines
suchasstationarycycles.
81.
Encouragepeopletoparticipateinphysicalactivityafterstroke.
82.
Assesspeoplewhoareabletowalkandaremedicallystableaftertheir
strokeforcardiorespiratoryandresistancetrainingappropriatetotheir
individualgoals.
83.
Cardiorespiratoryandresistancetrainingforpeoplewithstrokeshouldbe
startedbyaphysiotherapistwiththeaimthatthepersoncontinuesthe
programmeindependentlybasedonthephysiotherapistsinstructions(see
recommendation84).
84.
Forpeoplewithstrokewhoarecontinuinganexerciseprogramme
independently,physiotherapistsshouldsupplyanynecessaryinformation
aboutinterventionsandadaptationssothatwherethepersonisusingan
exerciseprovider,theprovidercanensuretheirprogrammeissafeand
tailoredtotheirneedsandgoals.Thisinformationmaytaketheformof
writteninstructions,telephoneconversationsorajointvisitwiththeprovider
andthepersonwithstroke,dependingontheneedsandabilitiesofthe
exerciseproviderandthepersonwithstroke.
85.
Tellpeoplewhoareparticipatinginfitnessactivitiesafterstrokeabout
commonpotentialproblems,suchasshoulderpain,andadvisethemtoseek
advicefromtheirGPortherapistiftheseoccur.
86.
Donotroutinelyofferwristandhandsplintstopeoplewithupperlimb
weaknessafterstroke.
87.
Considerwristandhandsplintsinpeopleatriskafterstroke(forexample,
peoplewhohaveimmobilehandsduetoweakness,andpeoplewithhigh
tone),to:
maintainjointrange,softtissuelengthandalignment
increasesofttissuelengthandpassiverangeofmovement
NationalClinicalGuidelineCentre,2013.
30
StrokeRehabilitation
Guidelinesummary
facilitatefunction(forexample,ahandsplinttoassistgriporfunction)
aidcareorhygiene(forexample,byenablingaccesstothepalm)
increasecomfort(forexample,usingasheepskinpalmprotectortokeep
fingernailsawayfromthepalmofthehand).
88.
Wherewristandhandsplintsareusedinpeopleafterstroke,theyshouldbe
assessedandfittedbyappropriatelytrainedhealthcareprofessionalsanda
reviewplanshouldbeestablished.
89.
Teachthepersonwithstrokeandtheirfamilyorcarerhowtoputthesplint
onandtakeitoff,careforthesplintandmonitorforsignsofrednessand
skinbreakdown.Provideapointofcontactforthepersonifconcerned.
90.
Donotroutinelyofferpeoplewithstrokeelectricalstimulationfortheirhand
andarm.
91.
Consideratrialofelectricalstimulationinpeoplewhohaveevidenceof
musclecontractionafterstrokebutcannotmovetheirarmagainst
resistance.
92.
Ifatrialoftreatmentisconsideredappropriate,ensurethatelectrical
stimulationtherapyisguidedbyaqualifiedrehabilitationprofessional.
93.
Theaimofelectricalstimulationshouldbetoimprovestrengthwhile
practisingfunctionaltasksinthecontextofacomprehensivestroke
rehabilitationprogramme.
94.
Continueelectricalstimulationifprogresstowardsclearfunctionalgoalshas
beendemonstrated(forexample,maintainingrangeofmovement,or
improvinggraspandrelease).
95.
Considerconstraintinducedmovementtherapyforpeoplewithstrokewho
havemovementof20degreesofwristextensionand10degreesoffinger
extension.Beawareofpotentialadverseevents(suchasfalls,lowmoodand
fatigue).
96.
Provideinformationforpeoplewithstrokeandtheirfamiliesandcarerson
howtopreventpainortraumatotheshoulderiftheyareatriskof
developingshoulderpain(forexample,iftheyhaveupperlimbweaknessand
spasticity).
97.
Manageshoulderpainafterstrokeusingappropriatepositioningandother
treatmentsaccordingtoeachpersonsneed.
98.
ForguidanceonmanagingneuropathicpainfollowNeuropathicpain(NICE
clinicalguideline96).
99.
Offerpeoplerepetitivetasktrainingafterstrokeonarangeoftasksforupper
limbweakness(suchasreaching,grasping,pointing,movingand
manipulatingobjectsinfunctionaltasks)andlowerlimbweakness(suchas
sittostandtransfers,walkingandusingstairs).
100.
Offerwalkingtrainingtopeopleafterstrokewhoareabletowalk,withor
withoutassistance,tohelpthembuildenduranceandmovemorequickly.
101.
Considertreadmilltraining,withorwithoutbodyweightsupport,asone
optionofwalkingtrainingforpeopleafterstrokewhoareabletowalkwith
orwithoutassistance.
102.
Offerelectromechanicalgaittrainingtopeopleafterstrokeonlyinthe
contextofaresearchstudy.
NationalClinicalGuidelineCentre,2013.
31
StrokeRehabilitation
Guidelinesummary
103.
Consideranklefootorthosesforpeoplewhohavedifficultywithswing
phasefootclearanceafterstroke(forexample,trippingandfalling)and/or
stancephasecontrol(forexample,kneeandanklecollapseorkneehyper
extensions)thataffectswalking.
104.
Assesstheabilityofthepersonwithstroketoputontheanklefootorthosis
orensuretheyhavethesupportneededtodoso.
105.
Assesstheeffectivenessoftheanklefootorthosisforthepersonwith
stroke,intermsofcomfort,speedandeaseofwalking.
106.
Assessmentforandtreatmentwithanklefootorthosesshouldonlybe
carriedoutaspartofastrokerehabilitationprogrammeandperformedby
qualifiedprofessionals.
107.
Forguidanceonfunctionalelectricalstimulationforthelowerlimbsee
Functionalelectricalstimulationfordropfootofcentralneurologicalorigin
(NICEinterventionalprocedureguidance278).
108.
Provideoccupationaltherapyforpeopleafterstrokewhoarelikelyto
benefit,toaddressdifficultieswithpersonalactivitiesofdailyliving.Therapy
mayconsistofrestorativeorcompensatorystrategies.
Restorativestrategiesmayinclude:
encouragingpeoplewithneglecttoattendtotheneglectedside
encouragingpeoplewitharmweaknesstoincorporatebotharms
establishingadressingroutineforpeoplewithdifficultiessuchaspoor
concentration,neglectordyspraxiawhichmakedressing
problematic.
Compensatorystrategiesmayinclude:
teachingpeopletodressonehanded
teachingpeopletousedevicessuchasbathinganddressingaids.
109.
Peoplewhohavedifficultiesinactivitiesofdailylivingafterstrokeshould
haveregularmonitoringandtreatmentbyoccupationaltherapistswithcore
skillsandtrainingintheanalysisandmanagementofactivitiesofdailyliving.
Treatmentshouldcontinueuntilthepersonisstableorabletoprogress
independently.
110.
Assesspeopleafterstrokefortheirequipmentneedsandwhethertheir
familyorcarersneedtrainingtousetheequipment.Thisassessmentshould
becarriedoutbyanappropriatelyqualifiedprofessional.Equipmentmay
includehoists,chairraisersandsmallaidssuchaslonghandledsponges.
111.
Ensurethatappropriateequipmentisprovidedandavailableforuseby
peopleafterstrokewhentheyaretransferredfromhospital,whateverthe
setting(includingcarehomes).
112.
Returntoworkissuesshouldbeidentifiedassoonaspossibleafterthe
personsstroke,reviewedregularlyandmanagedactively.Active
managementshouldinclude:
identifyingthephysical,cognitive,communicationandpsychological
demandsofthejob(forexample,multitaskingbyansweringemails
andtelephonecallsinabusyoffice)
identifyinganyimpairmentsonworkperformance(forexample,physical
limitations,anxiety,fatiguepreventingattendanceforafulldayat
NationalClinicalGuidelineCentre,2013.
32
StrokeRehabilitation
Guidelinesummary
work,cognitiveimpairmentspreventingmultitasking,and
communicationdeficits)
tailoringanintervention(forexample,teachingstrategiestosupport
multitaskingormemorydifficulties,teachingtheuseofvoice
activatedsoftwareforpeoplewithdifficultytyping,anddeliveryof
worksimulations)
educatingabouttheEqualityAct2010eandsupportavailable(for
example,anaccesstoworkscheme)
workplacevisitsandliaisonwithemployerstoestablishreasonable
accommodations,suchasprovisionofequipmentandgradedreturn
towork.
113.
Managereturntoworkorlongtermabsencefromworkforpeopleafter
strokeinlinewithrecommendationsinManaginglongtermsicknessand
incapacityforwork(NICEpublichealthguidance19).
114.
Informpeopleafterstrokethattheycanselfrefer,usuallywiththesupport
ofaGPornamedcontact,iftheyneedfurtherstrokerehabilitationservices.
115.
Provideinformationsothatpeopleafterstrokeareabletorecognisethe
developmentofcomplicationsofstroke,includingfrequentfalls,spasticity,
shoulderpainandincontinence.
116.
Encouragepeopletofocusonlifeafterstrokeandhelpthemtoachievetheir
goals.Thismayinclude:
facilitatingtheirparticipationincommunityactivities,suchasshopping,
civicengagement,sportsandleisurepursuits,visitingtheirplaceof
worshipandstrokesupportgroups
supportingtheirsocialroles,forexample,work,education,volunteering,
leisure,familyandsexualrelationships
providinginformationabouttransportanddriving(includingDVLA
requirements;seewww.dft.gov.uk/dvla/medical/aag).
117.
ManageincontinenceafterstrokeinlinewithrecommendationsinUrinary
incontinenceinneurologicaldisease(NICEclinicalguideline148)andFaecal
incontinence(NICEclinicalguideline49).
118.
Reviewthehealthandsocialcareneedsofpeopleafterstrokeandtheneeds
oftheircarersat6monthsandannuallythereafter.Thesereviewsshould
coverparticipationandcommunityrolestoensurethatpeoplesgoalsare
addressed.
119.
Forguidanceonsecondarypreventionofstroke,followrecommendationsin
Lipidmodification(NICEclinicalguideline67),Hypertension(NICEclinical
guideline127),Type2diabetes(NICEclinicalguideline87)andAtrial
fibrillation(NICEclinicalguideline36).
120.
Provideadviceonprescribedmedicationsinlinewithrecommendationsin
Medicinesadherence(NICEclinicalguideline76).
NationalClinicalGuidelineCentre,2013.
33
StrokeRehabilitation
Guidelinesummary
3.3 Keyresearchrecommendations
3.3.1
Upperlimbelectricalstimulation(ES)
What is the clinical and cost effectiveness of electrical stimulation (ES) as an adjunct
to rehabilitation to improve hand and arm function in people after stroke, from early
rehabilitation through to use in the community?
3.3.2
Intensiverehabilitationafterstroke
In people after stroke what is the clinical and cost effectiveness of intensive
rehabilitation (6 hours per day) versus moderate rehabilitation (2 hours per day) on
activity, participation and quality of life outcomes?
3.3.3
Neuropsychologicaltherapies
Which cognitive and which emotional interventions provide better outcomes for
identified subgroups of people with stroke and their families and carers at different
stages of the stroke pathway?
3.3.4
Shoulderpain
Which people with a weak arm after stroke are at risk of developing shoulder pain?
What management strategies are effective in the prevention or management of
shoulder pain of different aetiologies?
ForfurtherdetailspleaserefertoAppendixL.
NationalClinicalGuidelineCentre,2013.
34
StrokeRehabilitation
Methods
4 Methods
Thischaptersetsoutindetailthemethodsusedtogeneratetherecommendationsthatare
presentedinsubsequentchapters.Thisguidancewasdevelopedinaccordancewiththemethods
outlinedintheNICEGuidelinesManual2009187.
4.1 Developingthereviewquestionsandoutcomes
ReviewquestionsweredevelopedinaPICOframework(patient,intervention,comparisonand
outcome)forinterventionreviews.Thiswastoguidetheliteraturesearchingprocess,appraisal,and
synthesisofevidenceandtofacilitatethedevelopmentofrecommendationsbytheguideline
developmentgroup(GDG).TheyweredraftedbytheNCGCtechnicalteamandrefinedandvalidated
bytheGDG.Thequestionswerebasedonthekeyclinicalareasidentifiedinthescope(AppendixA).
Atotalof22reviewquestionswereidentified.
Fullliteraturesearches,criticalappraisalsandevidencereviewswerecompletedforallthespecified
clinicalquestions.
Chapter
Reviewquestions
Outcomes
Structureand
settings:strokeunits
Inpeopleafterstroke,does
organisedrehabilitationcare
(comprehensive,rehabilitationand
mixedrehabilitationstrokeunits)
improveoutcome(mortality,
dependency,requirementfor
institutionalcareandlengthof
hospitalstay)?
Structureand
settings:early
supporteddischarge
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
earlysupporteddischargeversus
usualcare?
Servicedelivery:goal
setting
Doestheapplicationofpatientgoal
settingaspartofplanningstroke
rehabilitationactivitiesleadtoan
improvementinpsychological
wellbeing,functioningandactivity?
Psychologicalwellbeing
viewsaboutthequalityofthegoalsetting
process
satisfactionwithoutcome
healthrelatedqualityoflife
physicalfunction
ActivitiesofDailyLiving(ADL)
Servicedelivery:
intensityof
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
Lengthofstay
FunctionalIndependenceMeasure(FIM)
NationalClinicalGuidelineCentre,2013.
35
Death
Deathordependency
Deathorinstitutionalcare
Durationofstayinhospitalorinstitutionor
both
Qualityoflife
Patientandcarersatisfaction
BarthelIndex
Lengthofhospitalstay
FunctionalIndependenceMeasure(FIM)
Caregiverstrainindex
Falls
Readmissionstohospital
HospitalAnxietyandDepressionScale
(HADS)
Mortality
QualityOfLife
NottinghamExtendedActivitiesofDaily
Living
StrokeRehabilitation
Methods
Chapter
rehabilitation
Reviewquestions
intensiverehabilitationversus
standardrehabilitation?
Outcomes
Supportand
information:
supported
informationprovision
Whatistheclinicalandcost
effectivenessofsupported
informationprovisionversus
unsupportedinformationprovision
onmoodanddepressioninpeople
withstroke?
Impactonmood/depression:
HospitalAnxietyandDepressionScale
(HADS)
GeneralHealthQuestionnaire
VisualAnalogueMoodScale
StrokeAphasicDepressionQuestionnaire
(SADQ)
GeriatricDepressionScale
BeckDepressionInventory
Selfefficacy
GeneralSelfefficacyScale
StrokeSelfefficacyQuestionnaire
LocusofControlScale
Extendedactivitiesofdailyliving(EADL)
NottinghamextendedADL
FrenchayActivitiesIndex
Yalemoodquestion
Cognitivefunctions:
visualneglect
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
cognitiverehabilitationversus
usualcaretoimprovespatial
awarenessand/orvisualneglect?
Cognitivefunctions:
memoryfunctions
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
memorystrategiesversususual
caretoimprovememory?
WechslerMemoryScale,
Rivermeadbehaviouralmemory
assessment,
Minimentalstateexamination(MMSE),
AddenbrooksCognitiveExamination
Revised,
AbbreviatedMentalTestScore.
Cognitivefunctions:
attentionfunction
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
sustainedattentiontrainingversus
usualcaretoimproveattention?
Minimentalstateexamination,Behavioural
inattentiontest,drawingtests,line
bisectiontest,cancellationtests,sentence
reading,targetscreenexaminations,
RivermeadPerceptualAssessmentBattery
Emotional
functioning
Inpeopleafterstrokewhatisthe
QualityofLife(forbothcarerandpatient)
clinicalandcosteffectivenessof
AnyQOLanddepressionoutcomes
psychologicaltherapiesprovidedto
NationalClinicalGuidelineCentre,2013.
36
BarthelIndex
QualityofLife(anymeasure)
NottinghamActivitiesofDailyLiving
Rankin
Rivermeadmobilityindex
FrenchayActivitiesIndex
Minimentalstateexamination(MMSE),
BehaviouralInattentionTest(BIT),
Drawingtests(forexample:clockdrawing),
LineBisectiontests,
Allcancellationtests(including:line
cancellation,bellcancellation),
Sentencereading,
Targetscreenexaminations(lumptogether
allcancellationtestsanddrawingtests),
RivermeadPerceptualAssessmentBattery
(RPAB)
StrokeRehabilitation
Methods
Chapter
Reviewquestions
thefamily(includingthepatients)?
Outcomes
includingthefollowing:strokeimpactscale,
EuroQoL,caregiverburdenscale,caregiver
strainindex,carerstrainindex,burdenof
strokescale,Strokeandaphasiaqualityof
lifescale,ASCOTscale.
Occurrenceofdepression/anxiety/moodin
carers
BeckDepressionInventory,Beck
DepressionInventory2,Geriatric
DepressionScale,neuropsychiatric
inventory,HospitalAnxietyandDepression
Scale(HADS),Generalhealthquestionnaire,
VisualAnalogueMoodScale,SADQ.
Vision:eye
movementtherapy
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
eyemovementtherapyforvisual
fieldlossversususualcare?
Digestivesystems:
swallowing
Inpeopleafterstrokewhatisthe
Occurrenceofaspirationpneumonia
clinicalandcosteffectivenessof
Occurrenceofchestinfections
interventionsforswallowingversus
Reductioninhospitalstay
alternativeinterventions
Reductioninreadmission
Returntonormaldiet
Communication:
Aphasia
Inpeopleafterstrokeisspeechand Functionalcommunication(languageor
languagetherapycomparedtono
communicationskillssufficienttopermit
speechandlanguagetherapyor
thetransmissionofmessageviaspoken,
placebo(socialsupportand
writtenornonverbalmodalities,ora
stimulation)effectiveinimproving
combinationofthesechannels)
language/communicationabilities
Formalmeasuresofreceptivelanguage
and/orpsychologicalwellbeing?
skills(languageunderstanding)
Formalmeasuresofexpressivelanguage
skills(languageproduction)
Overalllevelofseverityofaphasiaas
measuredbyspecialisttestbatteries(may
includeWesternAphasiaBatteryorPorch
IndexofCommunicativeAbilities)
Psychologicalorsocialwellbeingincluding
depression,anxietyanddistress
Patientsatisfaction/carerandfamilyviews
Compliance/dropout
Communication:
Dysarthria
Inpeopleafterstrokeisspeechand Measuresoffunctionalcommunication
languagetherapycomparedto
Formalmeasuresofreceptivelanguage
socialsupportandstimulation
skills(languageunderstanding)
effectiveinimprovingdysarthria?
Formalmeasuresofexpressivelanguage
skills(languageproduction)
Psychologicalorsocialwellbeingincluding
depression,anxietyanddistress
FrenchayDysarthriaAssessment.
NationalClinicalGuidelineCentre,2013.
37
Reading(speedandaccuracy)
Eyemovementtasks
Scanning
LetterCancellationTest
StrokeRehabilitation
Methods
Chapter
Reviewquestions
Outcomes
Measuresofarticulation(range,speed,
strength,andcoordination)
Perceptualmeasuresofvoiceandprosody
(forexample,VocalProfileAnalysis)
Acousticmeasures(forexample,
fundamentalfrequency,pitchperturbation
(jitter),amplitudeperturbation(shimmer),
asmeasuredby,computerisedsoundor
spectrography)
Communication:
Inpeopleafterstrokewith
intensityofspeech
communicationdifficultieswhatis
andlanguagetherapy theclinicalandcosteffectiveness
ofintensivespeechtherapyversus
standardspeechtherapy?
Anyoutcomereportedinthepapers.
Examplesinclude:
FunctionalAssessmentofCommunication
SkillsforAdults(ASHAFACS)
BostonNamingTest
WesternAphasiaBattery
StrokeDysphasiaIndex
McKennaGradedNamingTest
Communication:
Listeneradvice
Whatlisteneradvice
skills/informationwouldhelp
familymembers/carersimprove
communicationinpeoplewith
aphasiaafterstroke?
Anyoutcome
Qualityoflife
Movementstrength
training
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
strengthtrainingversususualcare
onimprovingfunctionand
reducingdisability?
Movement:fitness
training
Inpeopleafterstroke,does
cardiorespiratoryorresistance
fitnesstrainingimproveoutcome
(fitness,function,qualityoflife,
andmood)andreducedisability?
NationalClinicalGuidelineCentre,2013.
38
UpperLimb
MRCScale
NewtonMetres
Fuglmeyer
ActionResearchArmTest(ARAT)
FunctionalIndependenceMeasurement
(FIM)
BarthelIndex
Adverseeventspainorspasticity
LowerLimb/Trunk
TimedUpandGoTest
Anytimedwalk
Walkingdistance
Functional;IndependenceMeasure(FIM)
BarthelIndex
Adverseeventsfalls,painorspasticity
Mortalityrate
Dependenceorlevelofdisability
Physicalfitness
Mobility
Physicalfunction
Qualityoflife
Mood
StrokeRehabilitation
Methods
Chapter
Reviewquestions
Outcomes
Movement:hand
andarm:orthoses
upperlimb
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
orthosesforpreventionoflossof
rangeoftheupperlimbversus
usualcare?
Rangeofmovementassessedby
goniometry
Movement:handand Inpeopleafterstrokewhatisthe
arm:electrical
clinicalandcosteffectivenessof
stimulation
ElectricalStimulationforhand
functionversususualcare?
Anyoutcomereportedinthepaper.
UpperLimboutcomesincluding:
o ActionResearchArmTest(ARAT)
o FuglMeyerAssessment(FMA)
o 9holepegtest
o gripstrength.
Movement:Hand
andarm:constraint
inducedmovement
therapy
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
constraintinducedtherapyversus
usualcareonimprovingfunction
andreducingdisability?
Movement:
Repetitivetask
training
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
repetitivetasktrainingversususual
careonimprovingfunctionand
reducingdisability?
Lowerlimb
Anytimedwalk,6m,5m,10mwalk
Changeinwalkingdistance
Rivermeadmobilityindex
Upperlimb
Arm:
FuglMeyerAssessment,
ActionResearchArmTest(ARAT)
Hand:
Anypegholetest,
FrenchayArmTest,
MotorAssessmentScale(MAS)
Movement:walking
therapy:treadmill
training
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessofall
treadmillversususualcareon
improvingwalking?
Inpeopleafterstrokewhocan
walk,whatistheclinicalandcost
effectivenessoftreadmillplusbody
supportversustreadmillonlyon
improvingwalking?
Movement:walking
therapy:
electromechanical
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
electromechanicalgaittraining
Walkingspeeds(5metres/10metres/30
metres)
Anytimedwalk
NationalClinicalGuidelineCentre,2013.
39
FunctionalIndependenceMeasure(FIM)
BarthelIndex
FuglMeyerAssessment
ActionResearchArmTest(ARAT)
WolfMotorFunctionTest(WMFT)
9holepegtest
Anyadverseevent
Walkingspeeds(5m/10m/30m)
Timedwalk
Walkingendurance
FunctionalIndependenceMeasure(FIM)
BarthelIndex
RivermeadMobilityIndex
StrokeRehabilitation
Methods
Chapter
gaittraining
Reviewquestions
versususualcareonimproving
functionandreducingdisability?
Outcomes
Walkingendurance
FunctionalIndependenceMeasure(FIM)
BarthelIndex
RivermeadMobilityIndex
Movement:walking
therapy:orthoses
anklefoot
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
anklefootorthosesofalltypesto
improvewalkingfunctionversus
usualcare?
Gaitspeed:6minwalk,10mtimedwalk
LowerlimbMAS(stairs)
Timedwalk
Walkingendurance
FunctionalIndependence
Measure(FIM)/BarthelIndex
RivermeadMobilityIndex
Cadence
Gaitsymmetry(stancetime,steplength)
QualityofLifeoutcomes
Selfcare
Inpeopleafterstrokewhatisthe
clinicalandcosteffectivenessof
intensiveoccupationaltherapy
focusedspecificallyonpersonal
activitiesofdailylivingversususual
care?
Longtermhealthand Inpeopleafterstrokewhatisthe
socialsupport
clinicalandcosteffectivenessof
interventionstoaidreturntowork
versususualcare?
NottinghamExtendedActivitiesofDaily
Living(NEADL)
ExtendedActivitiesofDailyLiving(EADL)
FunctionalIndependenceMeasure(FIM)
BarthelIndex
NottinghamStrokeDressingAssessment
NorthwickParkNursingDependencyScale
RivermeadMobilityIndex
Samejobsameemployer
Samejobdifferentemployer
Differentjobsameemployer
Differentjobdifferentemployer
Unemployment
Retiredduetoillhealth
Voluntarywork
Benefitclaims
Duringthedevelopmentofquestionsconcerningemploymentandreturntowork,provisionof
information,deliveryofpsychologicaltherapiesandearlysupporteddischarge,theGDGtookthe
followingissuesintoconsideration:
WhentheGDGformulatedthequestionaboutaidstoreturntowork,theyacknowledgedthe
universalconsensusintheliteratureaboutthepredictivefactorsrestrictingpeopleafterstroketo
returntowork.Forthisreason,theybelievedthatthereviewofobservationalorcohortstudies
investigatingthisissuewouldnotprovideanyaddedvalueintheformulationof
recommendationsforthisguideline.TheGDGbelievedthatrandomisedtrialsinvestigatingthe
impactofanytypeofinterventionthatcouldfacilitatepeopletoreturntoemployment(either
formerornewemployment)wasahigherpriorityforthepurposesofthisguideline.Inaddition,
theGDGnotedthatthenatureofvocationalinterventionswouldbeverydiverseandtailoredto
individualcircumstances(typeofdisability,natureofemployment).
NationalClinicalGuidelineCentre,2013.
40
StrokeRehabilitation
Methods
Duringtheformulationofaquestionrelatedtoprovisionofinformationforpeopleafterstroke
andtheircarers,theGDGhadafulldiscussionwithregardtothelargeandheterogeneousareaof
informationprovision.Wewereclearlyunabletoaddressallinformationaspectswithinthe
timelineavailable.TheGDGagreedthatpeopleafterstrokeliveinarichinformation
environment,althoughitisnotalwaystailoredtothepatientsneeds.TheGDGfeltitwas
particularlyimportanttolookattheevidencepertainingtotheprovisionofsupported
information(informationgivenwithadditionalsupportofsomekindsuchastheactiveprovision
ofinformation,theencouragementoffeedback,availabilityofpeersupportoruseofinteractive
computerprogrammeasopposedtotheprovisionofleaflets/bookletsinisolation)inorderto
investigateitsimpactonmoodanddepressioninpeopleafterstrokeandpotentiallydirectthe
developmentofrecommendationsinthisarea.
Forthepsychologicalsupportquestion,theGDGthoughtthatthisshouldinvestigatethe
effectivenessofthepsychologicaltherapiessuchasfamilytherapy,cognitivebehaviourtherapy
andrelationshipcounsellingprovidedtothefamily(includingthepersonwithstroke)onthe
qualityoflifeofpeopleswithstrokeandtheircarers.Thegroupacknowledgedthatitwasnot
usualtohaveapsychologicaltherapyinisolationandthereforeallofthesetherapiesmayalso
includesomeformofeducationincombination.InlightofthepublicationofthePatient
experienceinadultNHSservices(NICEclinicalguideline138)theGDGagreedthatthisguidance
couldbecrossreferencedwhereappropriate
Whenformulatingthequestiononearlysupporteddischarge,theGDGagreedtoinvestigatethe
effectivenessofearlysupporteddischargeonimprovingspecificpatientandhospitalrelated
outcomes(suchasmortality,qualityoflife,readmissionsandlengthofstayinthehospital).The
GDGdidnotconsiderthatpatientswouldhaveanydifferentinformationneedsafterearly
supporteddischargetootherpatientsbeingdischargedfromhospital.
Duringthedevelopmentofquestionsforthisguidelinescopingsearchesforcohortstudieswere
undertakenandweconsultedwiththeGDGonwhethertheywereawareofanylargecohortstudies
intheseareasthatwouldjustifyincludingstudiesotherthanrandomisedtrials.Nonewere
identified.
4.2 Searchingforevidence
4.2.1
Clinicalliteraturesearch
Theaimoftheliteraturereviewwastoidentifyallavailable,relevantpublishedevidenceinrelation
tothekeyclinicalquestionsgeneratedbytheGDG.Systematicliteraturesearcheswereundertaken
toidentifyevidencewithinpublishedliteratureinordertoanswerthereviewquestionsasperThe
GuidelinesManual[2009]187.Clinicaldatabasesweresearchedusingrelevantmedicalsubject
headings,freetexttermsandstudytypefilterswhereappropriate.Studiespublishedinlanguages
otherthanEnglishwerenotreviewed.Wherepossible,searcheswererestrictedtoarticlespublished
inEnglishlanguage.Allsearcheswereconductedoncoredatabases,MEDLINE,Embase,Cinahland
TheCochraneLibrary.Additionalsubjectspecificdatabaseswereusedforsomequestions:PsycInfo
forpatientviews,allsearcheswereupdatedon5thOct2012.Nopapersafterthisdatewere
considered.
Searchstrategieswerecheckedbylookingatreferencelistsofrelevantkeypapers,checkingsearch
strategiesinothersystematicreviewsandaskingtheGDGforknownstudiesinaspecificarea.The
questions,thestudytypesapplied,thedatabasessearchedandtheyearscoveredcanbefoundin
Appendix[D].
Duringthescopingstage,asearchwasconductedforguidelinesandreportsonthewebsiteslisted
belowandonorganisationsrelevanttothetopic.Searchingforgreyliteratureorunpublished
literaturewasnotundertaken.Allreferencessentbystakeholderswereconsidered.
NationalClinicalGuidelineCentre,2013.
41
StrokeRehabilitation
Methods
GuidelinesInternationalNetworkdatabase(www.gin.net)
NationalGuidelineClearingHouse(www.guideline.gov/)
NationalInstituteforHealthandClinicalExcellence(NICE)(www.nice.org.uk)
NationalInstitutesofHealthConsensusDevelopmentProgram(consensus.nih.gov/)
HealthInformationResources,NHSEvidence(www.library.nhs.uk/)
ThetitlesandabstractsofrecordsretrievedbythesearcheswerescannedforrelevancetotheGDGs
clinicalquestions.Anypotentiallyrelevantpublicationswereobtainedinfulltext.Thesewere
assessedagainsttheinclusioncriteriaandthereferencelistswerescannedforanyarticlesnot
previouslyidentified.FurtherreferenceswerealsosuggestedbytheGDG.
4.2.2
Healtheconomicliteraturesearch
Systematicliteraturesearcheswerealsoundertakentoidentifyhealtheconomicevidencewithin
publishedliteraturerelevanttothereviewquestions.Theevidencewasidentifiedbyconductinga
broadsearchrelatingtotheguidelinepopulationintheNHSeconomicevaluationdatabase(NHS
EED),theHealthEconomicEvaluationsDatabase(HEED)andhealthtechnologyassessment(HTA)
databaseswithnodaterestrictions.Additionally,thesearchwasrunonMEDLINEandEmbase,witha
specificeconomicfilter,toensurerecentpublicationsthathadnotyetbeenindexedbythese
databaseswereidentified.StudiespublishedinlanguagesotherthanEnglishwerenotreviewed.
Wherepossible,searcheswererestrictedtoarticlespublishedinEnglishlanguage.
ThesearchstrategiesforhealtheconomicsareincludedinAppendix[D].Allsearcheswereupdated
on5thOct2012.Nopaperspublishedafterthisdatewereconsidered.
4.3 Evidenceofeffectiveness
TheResearchFellow:
Identifiedpotentiallyrelevantstudiesforeachreviewquestionfromtherelevantsearchresults
byreviewingtitlesandabstracts.Twentypercentofthesiftandselectionofpaperswasquality
assuredbyasecondreviewertoeliminateanypotentialofselectionbiasorerror.Fullpapers
werethenobtained.
Reviewedfullpapersagainstprespecifiedinclusion/exclusioncriteriatoidentifystudiesthat
addressedthereviewquestionintheappropriatepopulationandreportedonoutcomesof
interest(reviewprotocolsareincludedinAppendix[D]).
CriticallyappraisedrelevantstudiesusingtheappropriatechecklistasspecifiedinTheGuidelines
Manual187
Extractedkeyinformationaboutthestudysmethodsandresultsintoevidencetables(evidence
tablesareincludedinAppendix[H]).
Generatedsummariesoftheevidencebyoutcome(includedintherelevantchapterwriteups):
o Randomisedstudies:metaanalysed,whereappropriateandreportedinGRADEprofiles(for
clinicalstudies)seebelowfordetails.
4.3.1
Inclusion/exclusioncriteria
Theinclusion/exclusionofstudieswasbasedonthereviewprotocols.TheGDGwereconsulted
aboutanyuncertaintyregardinginclusion/exclusionofselectedstudies.Minimumsamplesizeand
theproportionofparticipantswithstrokewereamongtheinclusion/exclusioncriteriausedforthe
selectionofstudiesintheevidencereviews.TheGDGagreedthat(withtheexceptionofreview
questionsoncognitivefunctionsandFunctionalElectricalStimulation)thesamplesizeof20
participants(10ineacharm)wouldbetheminimumrequirementforastudytobeincluded.Forthe
reviewquestionsoncognitivefunctions,theminimumsamplesizewouldbesetat10participantsin
NationalClinicalGuidelineCentre,2013.
42
StrokeRehabilitation
Methods
totalduetothenatureofinterventionsandtheavailabilityofstudiesintheliterature.Thisdecision
onstudiessamplesizecutoffpointswasmadeforpragmaticreasons.
Wehaveincludedanystudyonstrokepopulationatleast2weekspoststroke.Wedidntapplyany
restrictiononselectionofstudieswithpopulationsonlongtermrehabilitation.
Duetothenatureofinterventionsinvestigatedinthefollowingevidencereviews;memory
strategies,eyemovementtherapy,swallowing,constraintinducedmovementtherapy,treadmill,
electromechanicalgaittraining,anklefoot,aidstoreturntowork,whichaimedultimatelytoreduce
disabilityandwouldbeapplicabletootherpopulations(whohavenotexperiencedstroke),theGDG
decidedthatwecouldusemixedpopulationsforreviewingthesequestions,aslongastheminimum
proportionofparticipantswithstrokeinthesestudieswassetat50%.Seethereviewprotocolsin
AppendixEandexcludedstudiesbythereviewquestions(withtheirexclusionreasons)inAppendix
Mforfulldetails.
4.3.2
Methodsofcombiningclinicalstudies
Datasynthesisforinterventionreviews
Wherepossible,metaanalyseswereconductedtocombinetheresultsofstudiesforeachreview
questionusingCochraneReviewManager(RevMan5)software.Fixedeffects(MantelHaenszel)
techniqueswereusedtocalculateriskratios(relativerisk)forthebinaryoutcomes.Theoutcome(s)
was(were)analysedusinganinversevariancemethodforpoolingweightedmeandifferencesand
wherethestudieshaddifferentscales,standardisedmeandifferenceswereused.
Statisticalheterogeneitywasassessedbyconsideringthechisquaredtestforsignificanceatp<0.1or
anIsquaredinconsistencystatisticof>50%toindicatesignificantheterogeneity.Wheresignificant
heterogeneitywaspresent,wecarriedoutasensitivityanalysiswithparticularattentionpaidto
allocationconcealment,blindingandlosstofollowup(missingdata).Incaseswheretherewas
inadequateallocationconcealment,unclearblindingordifferentialmissingdatamorethan20%in
thetwogroups,thiswasexaminedinasensitivityanalysis.Forthelatter,thedurationoffollowup
wasalsotakenintoconsiderationpriortoincludinginasensitivityanalysis.Nosubgroupanalyses
werepredefinedwiththeexceptionoftheclinicalquestionforconstraintinducedtherapyforwhich
asubgroupanalysisondurationofintervention(moreorlessthan5hours)wasprespecified(see
AppendixEforfurtherdetails).
Ifnosensitivityanalysiswasfoundtocompletelyresolvestatisticalheterogeneitythenarandom
effects(DerSimonianandLaird)modelwasemployedtoprovideamoreconservativeestimateofthe
effect.
Forcontinuousoutcomes,themeansandstandarddeviationswererequiredformetaanalysis.
However,incaseswherestandarddeviationswerenotreported,thestandarderrorwascalculatedif
thepvaluesor95%confidenceintervalswerereportedandmetaanalysiswasundertakenwiththe
meanandstandarderrorusingthegenericinversevariancemethodinCochraneReviewManager
(RevMan5)software.Whentheonlyevidencewasbasedonstudiessummarisedresultsbyonly
presentingmedians(andinterquartilerange),oronlypvaluesthisinformationwasincludedinthe
GRADEtableswithoutcalculatingtherelativeandabsoluteeffect.Consequently,imprecisionof
effectcouldnotbeassessedwhenresultswerenotpresentedinthestudiesbymeansandstandard
deviations.
Forbinaryoutcomes,absoluteeventrateswerealsocalculatedusingtheGRADEprosoftwareusing
eventrateinthecontrolarmofthepooledresults.
NationalClinicalGuidelineCentre,2013.
43
StrokeRehabilitation
Methods
Theresultsfromcrossoverstudieswerecombinedinametaanalysiswiththosefromparallel
randomisedtrials,onlyaftercorrectionshavebeenmadetothestandarderrorforthecrossover
trials.
4.3.3
Typeofstudies
Systematicreviews,doubleblinded,singleblindedandunblindedparallelrandomisedcontrolled
trials(RCTs)andcrossoverrandomizedstudieswereincludedintheevidencereviewsforthis
guideline.
Weincludedrandomisedtrials,astheyareconsideredthemostrobusttypeofstudydesignthat
couldproduceanunbiasedestimateoftheinterventioneffects.TheGDGbelievedthatthereason
whynolargetrialswerefoundforthispopulationwaslargelybecausestrokeunitsarerelativelynew
andpriortotheirformationithasnotbeenpossibletoconductlargemulticentreRCTs.
Wealsosearchedforsystematicreviewsofcohortstudies,howevernonewasfoundinanyreview
question.TheGDGdecidednottoincludeindividualcohortstudies.Cohortstudieshavebeenbased
inrehabilitationunitswheretherearemixedpopulationgroupsandextractingstrokedatafrom
thosemixedpopulationswouldbechallenging.Preliminarysearchesundertakendidnotfindany
largecohortstudies;thereforetheGDGagreedthatindividualcohortstudieswouldnotprovideany
addedvaluetothereviewsofindividualinterventions.
Formostofthereviewsthecontentofinterventionsandthereferredpopulationswithinthe
includedstudieswasfoundtobeverydiverse,makingtheextractionofrelevantdatachallengingand
timeconsuming.Inaddition,theGDGhaddifficultiesindrawingoverallconclusionsonthebodyof
evidencepresentedanditwasoftennotpossibletomakerecommendationsspecifyingwhat
interventionsshouldcompriseof.Intheseinstances,theGDGdecidedthattheresultsofeach
outcomeshouldbepresentedseparatelyforeachstudyandametaanalysiscouldnotbeconducted.
Duetothediversityofinterventions,itwasdecidedtoincludeasummarytableofstudiesincluded
withindividualcharacteristics(population,intervention,control,outcomes)atthebeginningofeach
evidencereview.
4.3.4
Typeofanalysis
EstimatesofeffectfromindividualstudieswerebasedonIntentionToTreat(ITT)analysiswiththe
exceptionoftheoutcomeofexperienceofadverseeventswhereasweusedAvailableCaseAnalysis
(ACA).ITTanalysisiswhereallparticipantsincludedintherandomisationprocesswereconsideredin
thefinalanalysisbasedontheinterventionandcontrolgroupstowhichtheywereoriginally
assigned.Weassumedthatparticipantsinthetrialslosttofollowupdidnotexperiencetheoutcome
ofinterest(forcategoricaloutcomes)andtheywouldnotconsiderablychangetheaveragescoresof
theirassignedgroups(forcontinuousoutcomes).
ItisimportanttonotethatITTanalysestendtobiastheresultstowardsnodifference.ITTanalysisis
aconservativeapproachtoanalysethedata,andthereforetheeffectmaybesmallerthaninreality.
However,themajorityofoutcomesselectedtobereviewedwerecontinuousoutcomes,veryfew
peopledroppedoutandmostofthestudiesreporteddataonanITTbasis.
4.3.5
Appraisingthequalityofevidencebyoutcomes
TheevidenceforoutcomesfromtheincludedRCTswasevaluatedandpresentedusinganadaptation
oftheGradingofRecommendationsAssessment,DevelopmentandEvaluation(GRADE)toolbox
developedbytheinternationalGRADEworkinggroup(http://www.gradeworkinggroup.org/).The
software(GRADEpro)developedbytheGRADEworkinggroupwasusedtoassessthequalityofeach
outcome,takingintoaccountindividualstudyqualityandthemetaanalysisresults.Thesummaryof
NationalClinicalGuidelineCentre,2013.
44
StrokeRehabilitation
Methods
studiescharacteristicsandfindingswaspresentedinonetableinthisguideline.The
Clinical/EconomicStudyCharacteristicstableincludesdetailsofthequalityassessmentwhilethe
Clinical/EconomicSummaryofFindingstableincludespooledoutcomedataandwhere
appropriate,anabsolutemeasureofinterventioneffectandthesummaryofqualityofevidencefor
thatoutcome.Inthistable,thecolumnsforinterventionandcontrolindicatesummariesofthesum
ofthesamplesizeforcontinuousoutcomes.Forbinaryoutcomessuchasnumberofpatientswithan
adverseevent,theeventrates(n/N:numberofpatientswitheventsdividedbysumofnumberof
patients)areshownwithpercentages.Reportingorpublicationbiaswasonlytakeninto
considerationinthequalityassessmentandincludedintheClinicalStudyCharacteristicstableifit
wasapparent.
EachoutcomewasexaminedseparatelyforthequalityelementslistedanddefinedinTable1and
eachgradedusingthequalitylevelslistedinTable2.Themaincriteriaconsideredintheratingof
theseelementsarediscussedbelow(seesection4.3.6GradingofEvidence).Footnoteswereusedto
describereasonsforgradingaqualityelementashavingseriousorveryseriousproblems.The
ratingsforeachcomponentweresummedtoobtainanoverallassessmentforeachoutcome.
Table3:TheGRADEtoolboxiscurrentlydesignedonlyforrandomisedtrialsandobservational
studies
Table1:DescriptionsofqualityelementsinGRADEforinterventionstudies
Table1:
DescriptionofqualityelementsinGRADEforinterventionstudies
Qualityelement
Description
Limitations
Limitationsinthestudydesignandimplementationmaybiastheestimatesofthe
treatmenteffect.Majorlimitationsinstudiesdecreasetheconfidenceintheestimate
oftheeffect.
Inconsistency
Inconsistencyreferstoanunexplainedheterogeneityofresults.
Indirectness
Indirectnessreferstodifferencesinstudypopulation,intervention,comparatorand
outcomesbetweentheavailableevidenceandthereviewquestion,or
recommendationmade.
Imprecision
Resultsareimprecisewhenstudiesincluderelativelyfewpatientsandfeweventsand
thushavewideconfidenceintervalsaroundtheestimateoftheeffectrelativetothe
clinicallyimportantthreshold.
Publicationbias
Publicationbiasisasystematicunderestimateoranoverestimateoftheunderlying
beneficialorharmfuleffectduetotheselectivepublicationofstudies.
Table2:
LevelsofqualityelementsinGRADE
Level
Description
None
Therearenoseriousissueswiththeevidence
Serious
Theissuesareseriousenoughtodowngradetheoutcomeevidencebyonelevel
Veryserious
Theissuesareseriousenoughtodowngradetheoutcomeevidencebytwolevels
Table3:
OverallqualityofoutcomeevidenceinGRADE
Level
Description
High
Furtherresearchisveryunlikelytochangeourconfidenceintheestimateofeffect
Moderate
Furtherresearchislikelytohaveanimportantimpactonourconfidenceintheestimate
ofeffectandmaychangetheestimate
Low
Furtherresearchisverylikelytohaveanimportantimpactonourconfidenceinthe
estimateofeffectandislikelytochangetheestimate
NationalClinicalGuidelineCentre,2013.
45
StrokeRehabilitation
Methods
Level
Description
Verylow
Anyestimateofeffectisveryuncertain
4.3.6
Gradingthequalityofclinicalevidence
Afterresultswerepooled,theoverallqualityofevidenceforeachoutcomewasconsidered.The
followingprocedurewasadoptedwhenusingGRADE:
11.Aqualityratingwasassigned,basedonthestudydesign.RCTsstartHIGHandobservational
studiesasLOW,uncontrolledcaseseriesasLOWorVERYLOW.
12.Theratingwasthendowngradedforthespecifiedcriteria:Studylimitations,inconsistency,
indirectness,imprecisionandreportingbias.Thesecriteriaaredetailedbelow.Observational
studieswereupgradediftherewasalargemagnitudeofeffect,doseresponsegradient,andifall
plausibleconfoundingwouldreduceademonstratedeffectorsuggestaspuriouseffectwhen
resultsshowednoeffect.Eachqualityelementconsideredtohaveseriousorveryseriousrisk
ofbiaswasrateddown1or2pointsrespectively.
13.Thedowngraded/upgradedmarkswerethensummedandtheoverallqualityratingwasrevised.
Forexample,allRCTsstartedasHIGHandtheoverallqualitybecameMODERATE,LOWorVERY
LOWif1,2or3pointsweredeductedrespectively.
14.Thereasonsorcriteriausedfordowngradingwerespecifiedinthefootnotes.
Thedetailsofcriteriausedforeachofthemainqualityelementarediscussedfurtherinthefollowing
sections4.3.7to4.3.10.
4.3.7
Studylimitations
ThemainlimitationsforrandomisedcontrolledtrialsarelistedinTable4.
Outcomesfromstudieswhichwerenotdoubleblindedweredowngradedonstudylimitationsdueto
thehigherriskofbias.However,theGDGexpressedtheirconcernthatconductingdoubleblinded
trialsinstrokerehabilitationwasnotpracticalasitwouldbeimpossibletoblindthetrialparticipant
duetothenatureoftheinterventionsdeliveredinstrokerehabilitation.However,singleblindedand
unblindedtrialsweredowngradedtomaintainaconsistentapproachinqualityratingacrossthe
guidelinefollowingtheapplicationofGRADEsystem,recognisingthatadoubleblindedtrialwould
providetheleastbiasedoutcomesinaclinicalsetting.Table4listedthelimitationsconsideredfor
randomisedcontrolledtrials.
Table4:
Studylimitationsofrandomisedcontrolledtrials
Limitation
Explanation
Allocation
concealment
Thoseenrollingpatientsareawareofthegrouptowhichthenextenrolledpatient
willbeallocated(majorprobleminpseudoorquasirandomisedtrialswith
allocationbydayofweek,birthdate,chartnumber,etc.)
Lackofblinding
Patient,caregivers,thoserecordingoutcomes,thoseadjudicatingoutcomes,ordata
analystsareawareofthearmtowhichpatientsareallocated.Baselinedifferences
arealsoassessedinthiscategory.
Incomplete
accountingof
patientsand
outcomeevents
Losstofollowupnotaccountedandfailuretoadheretotheintentiontotreat
principlewhenindicated
Selectiveoutcome
reporting
Reportingofsomeoutcomesandnotothersonthebasisoftheresults
Otherlimitations
Forexample:
NationalClinicalGuidelineCentre,2013.
46
StrokeRehabilitation
Methods
Limitation
Explanation
Stoppingearlyforbenefitobservedinrandomisedtrials,inparticularintheabsence
ofadequatestoppingrules
Useofinvalidatedpatientreportedoutcomes
Carryovereffectsincrossovertrials
Recruitmentbiasinrandomisedtrials
4.3.8
Inconsistency
Inconsistencyreferstoanunexplainedheterogeneityofresults.Whenestimatesofthetreatment
effectacrossstudiesdifferwidely(i.e.heterogeneityorvariabilityinresults),thissuggeststrue
differencesinunderlyingtreatmenteffect.Whenheterogeneityexists(Chisquarep<0.1orIsquared
inconsistencystatisticof>50%),butnoplausibleexplanationcanbefound(forexampleacuteor
chronicstrokepopulations,durationofintervention,differentfollowupperiods),thequalityof
evidencewasdowngradedbyoneortwolevels,dependingontheextentofuncertaintytothe
resultscontributedbytheinconsistencyintheresults.Duetothediversityofinterventionsusedin
theincludedtrialsforthisguideline,therewerecaseswheretheGDGbelievedthepresentationof
evidenceshouldbekeptseparateandexplanatoryfootnotesweregiveninGRADEtableswhere
appropriate.InadditiontotheIsquareandChisquarevalues,thedecisionfordowngradingwas
alsodependentonfactorssuchaswhethertheinterventionisassociatedwithbenefitinallother
outcomesorwhethertheuncertaintyaboutthemagnitudeofbenefit(orharm)oftheoutcome
showingheterogeneitywouldinfluencetheoveralljudgmentaboutnetbenefitorharm(acrossall
outcomes).
Ifinconsistencycouldbeexplainedbasedonprespecifiedsubgroupanalysis,theGDGtookthisinto
accountandconsideredwhethertomakeseparaterecommendationsbasedontheidentified
explanatoryfactors,i.e.populationandintervention.Wheresubgroupanalysisgivesaplausible
explanationofheterogeneity,thequalityofevidencewouldnotbedowngraded.Themostcommon
factorofsubgroupanalysiswasthetimesincestrokeeventandtheGDGconsideredtheevidenceof
someoutcomesseparatelyforacuteandchronicstrokepatients.
4.3.9
Indirectness
Directnessreferstotheextenttowhichthepopulations,intervention,comparisonsandoutcome
measuresaresimilartothosedefinedintheinclusioncriteriaforthereviews.Indirectnessis
importantwhenthesedifferencesareexpectedtocontributetoadifferenceineffectsize,ormay
affectthebalanceofharmsandbenefitsconsideredforanintervention.TheGDGdecidedthatfor
specificquestions(forexamplethereviewofinterventionstoassessclinicalandcosteffectivenessof
interventionstoaidreturntowork)thereviewofevidencecouldincludemixedpopulationswithat
least50%strokepatients.
4.3.10
Imprecision
Thesamplesize,eventrates,theresultingwidthofconfidenceintervalsandtheminimalimportant
differenceintheoutcomebetweenthetwogroupswerethemaincriteriaconsidered.
Thethresholdsofimportantbenefitsorharms,ortheMID(minimalimportantdifference)foran
outcomeareimportantconsiderationsfordeterminingwhetherthereisaclinicallyimportant
differencebetweeninterventionandcontrolgroupsandinassessingimprecision.Forcontinuous
outcomes,theMIDisdefinedasthesmallestdifferenceinscoreintheoutcomeofinterestthat
informedpatientsorinformedproxiesperceiveasimportant,etherbeneficialorharmful,andthat
wouldleadthepatientorcliniciantoconsiderachangeinthemanagement(98124,231,232).Aneffect
NationalClinicalGuidelineCentre,2013.
47
StrokeRehabilitation
Methods
estimatelargerthantheMIDisconsideredtobeclinicallyimportant.Fordichotomousoutcomes,
theMIDisconsideredintermsofchangesofabsoluterisk.
Thedifferencebetweentwointerventions,asobservedinthestudies,wascomparedagainstthe
MIDwhenconsideringwhetherthefindingswereofclinicalimportance;thisisusefultoguide
decisions.Forexample,iftheeffectwassmall(lessthantheMID),thisfindingsuggeststhatthere
maynotbeenoughdifferencetostronglyrecommendoneinterventionovertheotherbasedonthat
outcome.
Wesearchedtheliteratureforpublishedstudieswhichgaveaminimalimportantdifferencepoint
estimatefortheoutcomesspecifiedintheprotocolandagreementwasobtainedfromtheGDGfor
theiruseinassessingimprecisionthroughoutthereviewsintheguideline.Table5presentstheMID
thresholdsusedforthespecifiedoutcomesandthesourceofbaseevidence.Wherenopublished
studieswerefoundonMIDsforoutcomes,thedefaultGRADEproMIDswasused.Forcategorical
data,wecheckedwhethertheconfidenceintervaloftheeffectcrossedoneortwoendsoftherange
of0.751.25.Forquantitativeoutcomestwoapproacheswereused.Whenonlyonetrialwas
includedastheevidencebaseforanoutcome,themeandifferencewasconvertedtothe
standardizedmeandifference(SMD)andcheckedtoseeiftheconfidenceintervalcrossed0.5.
However,themeandifference(95%confidenceinterval)wasstillpresentedintheGradetables.If
twoormoreincludedtrialsreportedaquantitativeoutcomethenthedefaultapproachof
multiplying0.5bystandarddeviation(takenasthemedianofthestandarddeviationsacrossthe
metaanalysedstudies)wasemployed.WhenthedefaultMIDswereused,theGDGwouldassessthe
estimateofeffectwithrespectstotheMID,andthentheimprecisionmaybereconsidered.
Theconfidenceintervalforthepooledorbestestimateofeffectwasconsideredinrelationtothe
MID,asillustratedinFigure1.Essentially,iftheconfidenceintervalcrossedtheMIDthreshold,there
wasuncertaintyintheeffectestimateinsupportingourrecommendation(becausetheCIwas
consistentwithtwodecisions)andtheeffectestimatewasratedasimprecise.
Table5:
AgreedMIDsfromtheliterature
Outcomes
AgreedMID
Evidencebase
Otherconsiderations
BarthelIndex
1.85points(SE1.45)
Hsieh,Wang,Wu,Chen,Sheu,
Hsieh2007.116
Taiwansetting(n=43)
Papersaimtoestimate
MID
ActionResearch 12and17pointsfor
ArmTest(ARAT) theaffecteddominant
andnondominant
sidesrespectively
Lang,Edwards,Birkenmeier,
Dromerick2008.141
Inpatientrehabilitation
hospitalsettingearly
afterstrokepatientswith
hemiparesis(N=52)
Papersaimtoestimate
MID.
FuglMeyer
Assessment
(FMA)
VanderLee,Beckerman,
LankhorstandBouter2001.269
Paperassessedsensitivityof
theresearcharmtestin22
chronicstrokepatients
Differenceby10%of
thetotalscale
WolfMotor
FunctionTest
(WMFT)
Animprovementof19 Lang,Edwards,Birkenmeier,
secondsonthe
Dromerick2008141
affecteddominantside
(16%ofthe120second
limit)
Inpatientrehabilitation
hospitalsettingearly
afterstrokepatientswith
hemiparesis(N=52)
Papersaimtoestimate
MID.
MotorActivity
Log(MAL)
Atleast1.0and1.1
points(1718%ofthe
scale)fortheaffected
dominantandnon
Inpatientrehabilitation
hospitalsettingearly
afterstrokepatientswith
hemiparesis(N=52)
Lang,Edwards,Birkenmeier,
Dromerick2008141
NationalClinicalGuidelineCentre,2013.
48
StrokeRehabilitation
Methods
Outcomes
AgreedMID
dominantsides
respectively
Evidencebase
Otherconsiderations
Functional
Independence
Measure(FIM)
22pointsforthetotal
FIM,17points(onthe
105pointscale16%)
forthemotorFIMand
3pointsforthe
cognitiveFIM.
Beninato,GillBody,Salles,
Stark,BlackSchaffer,Stein.
2006.24
Patientswithstrokein
longtermacutehospital.
(N=113)
Papersaimtoestimate
MID
Walkingspeed
(forchronic
strokepatients)
20cm/sec
PerryJ,GarrettM,GronleyJK,
MulroySJ.Classificationof
walkinghandicapinstroke
population.Stroke1995;26:
98289.202
chronicstrokepatients
(over3monthspoststroke)
Walkingspeed
(foracute
strokepatients)
16cm/sec
TilsonJK,SullivanK,CenSY,
RoseD.K,CH.Koradia,SP.
Azen,PW.Duncan2010.258
Firsttimestrokepatients
(2060dayspoststroke)
withseveregait
impairments(N=283)
Papersaimtoestimate
MIDforgaitspeed
TimedUpand
Go
10sec
PerryJ,GarrettM,GronleyJK,
MulroySJ.Classificationof
walkinghandicapinstroke
population.Stroke1995;26:
98289.202
StairsTest
15sec
PodsiadloD,RichardsonS.The
timedUp&Go:atestofbasic
functionalmobilityforfrail
elderlypersons.JAmGeriatr
Soc1991;39:14248.207
6minutewalk
test
28m
DeanCM,RichardsCL,
MalouinF2000.58
Rangeof
movement
(wrist
extensibility)
5ochange(SD4.1o)
LanninNA,CusickA,McCluskey MIDtakenfromsamplesize
A,HerbertRD2007.144
calculation(N=63)
Papersaimtoestimate
MID.
NationalClinicalGuidelineCentre,2013.
49
StrokeRehabilitation
Methods
Figure1: Illustrationofpreciseandimprecisionoutcomesbasedontheconfidenceintervalof
outcomesinaForrestplot
Source: FigureadaptedfromGRADEProsoftware.
MID=minimalimportantdifferencedeterminedforeachoutcome.TheMIDsarethethresholdfor
appreciablebenefitsandharms.Theconfidenceintervalsofthetopthreepointsofthediagramwere
consideredprecisebecausetheupperandlowerlimitsdidnotcrosstheMID.Conversely,thebottom
threepointsofthediagramwereconsideredimprecisebecauseallofthemcrossedtheMIDand
reducedourcertaintyoftheresults.
4.4 Evidenceofcosteffectiveness
TheGuidelineDevelopmentGroup(GDG)isrequiredtomakedecisionsbasedonthebestavailable
evidenceofbothclinicalandcosteffectiveness.Guidelinerecommendationsshouldbebasedonthe
estimatedcostsofthetreatmentoptionsinrelationtotheirexpectedhealthbenefits(thatis,their
costeffectiveness),ratherthanonthetotalcostorresourceimpactofimplementingthem.Thus,if
theevidencesuggeststhataninterventionprovidessignificanthealthbenefitsatanacceptablecost
perpatienttreated,itshouldberecommendedevenifitwouldbeexpensivetoimplementacross
thewholepopulation.
Evidenceoncosteffectivenessrelatedtothekeyclinicalissuesbeingaddressedintheguidelinewas
sought.Thehealtheconomistundertook:
Asystematicreviewofthepublishedeconomicliterature.
Newcosteffectivenessanalysisinpriorityareas.
Whennorelevantpublishedstudieswerefound,andanewanalysiswasnotprioritised,theGDG
madeaqualitativejudgementaboutcosteffectivenessbyconsideringexpecteddifferencesin
resourceusebetweencomparatorsandrelevantUKNHSunitcostsalongsidetheresultsofthe
clinicalreviewofeffectivenessevidence.Whereconsidereduseful,thisincludedcalculationof
expectedcostdifferencesandconsiderationoftheQALYgainthatwouldberequiredtojustifythe
expectedadditionalcostoftheinterventionbeingconsidered.Unitcostswerebasedonpublished
nationalsourcewhereavailable.Staffcostsarereportedusingthetypicalsalarybandofsomeone
deliveringtheinterventionasidentifiedbyclinicalGDGmembers.Itshouldbenotedhoweverthatin
NationalClinicalGuidelineCentre,2013.
50
StrokeRehabilitation
Methods
practicestaffbandswillvaryduetotheneedforaskillmixacrossteams.Inputstocalculations
shouldnotbeinterpretedasrecommendationsaboutwhoshoulddelivercare.
4.4.1
Literaturereview
Thehealtheconomist:
Identifiedpotentiallyrelevantstudiesforeachreviewquestionfromtheeconomicsearchresults
byreviewingtitlesandabstractsfullpaperswerethenobtained.
Reviewedfullpapersagainstprespecifiedinclusion/exclusioncriteriatoidentifyrelevantstudies
(seebelowfordetails).
CriticallyappraisedrelevantstudiesusingtheeconomicevaluationschecklistasspecifiedinThe
GuidelinesManual187.
Extractedkeyinformationaboutthestudysmethodsandresultsintoevidencetables(evidence
tablesareincludedinAppendixH).
GeneratedsummariesoftheevidenceinNICEeconomicevidenceprofiles(includedinthe
relevantchapterwriteups)seebelowfordetails.
4.4.1.1
Inclusion/exclusion
Fulleconomicevaluations(studiescomparingcostsandhealthconsequencesofalternativecourses
ofaction:costutility,costeffectiveness,costbenefitandcostconsequenceanalyses)and
comparativecostingstudiesthataddressedthereviewquestionintherelevantpopulationwere
consideredpotentiallyapplicableaseconomicevidence.
Studiesthatonlyreportedcostperhospital(notperpatient),oronlyreportedaveragecost
effectiveness,withoutdisaggregatedcostsandeffects,wereexcluded.Abstracts,posters,reviews,
letters/editorials,foreignlanguagepublicationsandunpublishedstudieswereexcluded.Studies
judgedtohaveanapplicabilityratingofnotapplicablewereexcluded(thisincludedstudiesthat
tooktheperspectiveofanonOECDcountry).
Remainingstudieswereprioritisedforinclusionbasedontheirrelativeapplicabilitytothe
developmentofthisguidelineandthestudylimitations.Forexample,ifahighquality,directly
applicableUKanalysiswasavailableotherlessrelevantstudiesmaynothavebeenincluded.Where
exclusionsoccurredonthisbasis,thisisnotedintherelevantsection.
Formoredetailsabouttheassessmentofapplicabilityandmethodologicalqualityseetheeconomic
evaluationchecklist(TheGuidelinesManual,AppendixH187)andthehealtheconomicsresearch
protocolinAppendixE.
4.4.1.2
NICEeconomicevidenceprofiles
TheNICEeconomicevidenceprofilehasbeenusedtosummarisecostandcosteffectiveness
estimates.Theeconomicevidenceprofileshows,foreacheconomicstudy,anassessmentof
applicabilityandmethodologicalquality,withfootnotesindicatingthereasonsfortheassessment.
Theseassessmentsweremadebythehealtheconomistusingtheeconomicevaluationchecklistfrom
TheGuidelinesManual,AppendixH187.Italsoshowsincrementalcosts,incrementaleffects(for
example,QALYs)andtheincrementalcosteffectivenessratiofromtheprimaryanalysis,aswellas
informationabouttheassessmentofuncertaintyintheanalysis.SeeTable6formoredetails.
IfanonUKstudywasincludedintheprofile,theresultswereconvertedintopoundssterlingusing
theappropriatepurchasingpowerparity194.
NationalClinicalGuidelineCentre,2013.
51
StrokeRehabilitation
Methods
Table6:
ContentofNICEeconomicprofile
Item
Description
Study
Firstauthorname,reference,dateofstudypublicationandcountryperspective.
Limitations
Anassessmentofmethodologicalqualityofthestudy(a):
Minorlimitationsthestudymeetsallqualitycriteria,orthestudyfailstomeetone
ormorequalitycriteria,butthisisunlikelytochangetheconclusionsaboutcost
effectiveness.
Potentiallyseriouslimitationsthestudyfailstomeetoneormorequalitycriteria,
andthiscouldchangetheconclusionaboutcosteffectiveness
Veryseriouslimitationsthestudyfailstomeetoneormorequalitycriteriaand
thisisverylikelytochangetheconclusionsaboutcosteffectiveness.Studieswith
veryseriouslimitationswouldusuallybeexcludedfromtheeconomicprofiletable.
Applicability
Anassessmentofapplicabilityofthestudytotheclinicalguideline,thecurrentNHS
situationandNICEdecisionmaking(a):
Directlyapplicabletheapplicabilitycriteriaaremet,oroneormorecriteriaare
notmetbutthisisnotlikelytochangetheconclusionsaboutcosteffectiveness.
Partiallyapplicableoneormoreoftheapplicabilitycriteriaarenotmet,andthis
mightpossiblychangetheconclusionsaboutcosteffectiveness.
Notapplicableoneormoreoftheapplicabilitycriteriaarenotmet,andthisis
likelytochangetheconclusionsaboutcosteffectiveness.
Othercomments
Particularissuesthatshouldbeconsideredwheninterpretingthestudy.
Incrementalcost
Themeancostassociatedwithonestrategyminusthemeancostofacomparator
strategy.
Incrementaleffects
ThemeanQALYs(orotherselectedmeasureofhealthoutcome)associatedwith
onestrategyminusthemeanQALYsofacomparatorstrategy.
ICER
Incrementalcosteffectivenessratio:theincrementalcostdividedbytherespective
QALYsgained.
Uncertainty
AsummaryoftheextentofuncertaintyabouttheICERreflectingtheresultsof
deterministicorprobabilisticsensitivityanalyses,orstochasticanalysesoftrialdata,
asappropriate.
(a) LimitationsandapplicabilitywereassessedusingtheeconomicevaluationchecklistfromTheGuidelinesManual,
AppendixH187
4.4.2
Undertakingnewhealtheconomicanalysis
Aswellasreviewingthepublishedeconomicliteratureforeachreviewquestion,asdescribedabove,
neweconomicanalysiswasundertakenbythehealtheconomistinselectedareas.Priorityareasfor
newhealtheconomicanalysiswereagreedbytheGDGafterformationofthereviewquestionsand
considerationoftheavailablehealtheconomicevidence.
TheGDGidentifiedintensityofrehabilitationasthehighestpriorityareaforanoriginaleconomic
model.Thisissueimpactsthelargestgroupofpeopleintheguidelineasitrelatestothewhole
populationratherthanaspecificsubset.Inaddition,theGDGconsideredthattheintensityof
rehabilitationprovidedcurrentlyvariesconsiderablyfromservicetoserviceintermsofhoursperday
anddurationoftherapy,anditisgenerallylowerthanthatcurrentlyrecommendedintheNICE
qualitystandardforongoingrehabilitation.Thereforerecommendationsinthisareawereconsidered
likelytohavethebiggestimpactonNHSresourcesandpatientoutcomes.
Thefollowinggeneralprincipleswereadheredtoindevelopingthecosteffectivenessanalysis:
MethodswereconsistentwiththeNICEreferencecase185.
TheGDGwasconsultedduringtheconstructionandinterpretationofthemodel.
NationalClinicalGuidelineCentre,2013.
52
StrokeRehabilitation
Methods
Modelinputswerebasedonthesystematicreviewoftheclinicalliteraturesupplementedwith
otherpublisheddatasourceswherepossible.
Whenpublisheddatawasnotavailableexpertopinionwasusedtopopulatethemodel.
Modelinputsandassumptionswerereportedfullyandtransparently.
Theresultsweresubjecttosensitivityanalysisandlimitationswerediscussed.
ThemodelwaspeerreviewedbyanotherhealtheconomistattheNCGC.
FullmethodsfortheintensityofrehabilitationcosteffectivenessanalysisaredescribedinAppendix
K.
4.4.3
Costeffectivenesscriteria
NICEsreportSocialvaluejudgements:principlesforthedevelopmentofNICEguidancesetsoutthe
principlesthatGDGsshouldconsiderwhenjudgingwhetheraninterventionoffersgoodvaluefor
money186,187.
Ingeneral,aninterventionwasconsideredtobecosteffectiveifeitherofthefollowingcriteria
applied(giventhattheestimatewasconsideredplausible):
a. Theinterventiondominatedotherrelevantstrategies(thatis,itwasbothlesscostlyintermsof
resourceuseandmoreclinicallyeffectivecomparedwithalltheotherrelevantalternative
strategies),or
b. Theinterventioncostlessthan20,000perqualityadjustedlifeyear(QALY)gainedcompared
withthenextbeststrategy.
IftheGDGrecommendedaninterventionthatwasestimatedtocostmorethan20,000perQALY
gained,ordidnotrecommendonethatwasestimatedtocostlessthan20,000perQALYgained,
thereasonsforthisdecisionarediscussedexplicitlyinthefromevidencetorecommendations
sectionoftherelevantchapterwithreferencetoissuesregardingtheplausibilityoftheestimateor
tothefactorssetoutintheSocialvaluejudgements:principlesforthedevelopmentofNICE
guidance186.
IfastudyreportedthecostperlifeyeargainedbutnotQALYs,thecostperQALYgainedwas
estimatedbymultiplyingbyanappropriateutilityestimatetoaidinterpretation.Theestimatedcost
perQALYgainedisreportedintheeconomicevidenceprofilewithafootnotedetailingthelifeyears
gainedandtheutilityvalueused.WhenQALYsorlifeyearsgainedarenotusedintheanalysis,
resultsaredifficulttointerpretunlessonestrategydominatestheotherswithrespecttoevery
relevanthealthoutcomeandcost.
4.5 PostconsultationprotocolincludingmodifiedDelphimethodology
Duringconsultation,substantialstakeholdercommentswerereceivedwhichhighlightedanumberof
significantissuesinrelationtotheguidelinescopeandrecommendationsdevelopedintheguideline.
Stakeholdersraisedconcernsthattheguidelinewasincompletebecauseofthenumberofareasin
therehabilitationpatientcarepathwaythattheguidelinehadnotcovered,andthismayresultin
therapiesandservicesforthestrokepopulationbeingreducedorevenwithdrawn.Theareas
identifiedintheconsultationperiodincluded:
servicedelivery,rolesandresponsibilityofthemultidisciplinaryteam/strokerehabilitation
services
holisticassessment,careplanning,goalsetting,ongoingreviewandmonitoring
transferofcare/dischargeplanningandinterfacewithsocialcare
longtermhealthandsocialsupportforpeopleafterstrokeandpatientinformationneeds
NationalClinicalGuidelineCentre,2013.
53
StrokeRehabilitation
Methods
Stakeholdersalsoconsideredthatsometopicsincludedinthescopehadnotbeenaddressed
adequately,includingmooddisorders(depressionandanxiety),physicalfitnessandexercise,other
speechandlanguagetherapiesanddiplopia.
Thefocusoftheoutcomesfortheinterventionsincludedintheguidelinehasbeenonfunctionand
mobilityasthesewereconsideredbytheGuidelineDevelopmentGroup(GDG)tohavethebiggest
impactonpatientslives.Howevermanystakeholdersconsideredthatthepatientexperienceand
holisticapproachestocarehadbeenneglectedandrepresentedamajorgapintheguidance.Inlight
ofthecommentsreceivedfromstakeholders,theGDGagreedthatadditionalworkshouldbecarried
outforsomeoftheseareasorreferencemadetootherNICEguidance,inordertoproduceamore
completepieceofguidancethatwouldbeusefultohealthprofessionalsdeliveringrehabilitationtoa
strokepopulation.ThecurrentguidancehasfollowedstandardNICEmethodologyandtheGDGwere
inagreementthatforthoseareaswhereeitherweakornoevidencewasavailablearobustprocess
neededtobefollowed.
InconsultationwithNICEandtheGDGtheNCGCtechnicalteamconductedadditionalworkto
addresstheareasidentifiedbystakeholdersandnotcoveredintheoriginalscope.Comprehensive
searchesofdatabaseswithtermsdesignedtoidentifyevidencerelatedtothetopicsoutlinedabove
wereundertakenfollowingtheNICEprocessbutrestrictedtoretrieveotherguidelinesand
systematicreviewsonly.Inadditionasimilarscopingsearchwasdoneforeconomicevidence
relatingtothesameareas.Thesearchstrategywaslimitedtocaptureonlyeconomicevaluations.A
firstsiftwasundertakentoidentifypotentiallyrelevanteconomicpapersrelatedtothetopicslisted
above.
ReviewsoftheclinicalandeconomicliteraturewereundertakenfollowingtheusualNICEprocess
andpresentedtotheGDGwhousedthisevidenceasabasistomakefurtherrecommendations.
WheretherewererecommendationsinotherNICEguidancerelevanttothestrokepopulationand
addressedcommentshighlightedbystakeholders,crossreferencetothesewasmaderatherthan
undertakingfurtheroriginalwork.
RelevantguidelinesidentifiedfromthecomprehensivesearchwerequalityassessedusingtheAGREE
IItoolchecklist.Thoseofsufficientqualitywerereviewedforrecommendationsrelatingtothetopics
identifiedinthestakeholderconsultation.
ThefullprotocolcanbefoundinAppendixB.
ModifiedDelphiconsensusmethodology
Astheevidencebasewasweakorabsentformanyoftheareasstakeholderswishedtheguidelineto
includeadifferentmethodology.Thiswasseenasnecessarysinceitwouldprovidearobustprocess
toenabletheGDGtomakefurtherrecommendations.Wheretherewasalackofpublishedevidence
theNCGCtechnicalteamusedamodifiedDelphimethod(anonymous,multiround,consensus
buildingtechnique)basedonotheravailableguidelinesorexpertopinion.Thistypeofsurveyhas
beenusedsuccessfullyforgenerating,analysingandsynthesisingexpertviewtoreachagroup
consensusposition.Thetechniqueusessequentialquestionnairestosolicitindividualresponses,
withthepotentialthreatofpeerpressureremoved95.Thisisanimportantconsiderationandisakey
strengthofthetechnique.StraussandZieglers249(1975)seminalworkonthetechniquehighlights
thefeaturesofthetechnique:
Enablestheeffectiveuseofapanelofexperts
Dataisgeneratedthroughsequentialquestioning
Highlightsconsensusanddivergentopinion
Anonymityisguaranteed
Ithandlesjudgementaldataeffectively
NationalClinicalGuidelineCentre,2013.
54
StrokeRehabilitation
Methods
InNICEprocesses,littleornoevidenceforreviewsisanexceptionalcircumstancewhenformal
consensustechniques(suchastheDelphimethod)canbeadopted187.Themethodsandprocess
proposedwasdiscussedwithmethodologicaladviserswithinNICEandtheprotocolwasagreedand
signedoffbythempriortoworkbeingcarriedout.
Delphistatementsweredistilledfromthecontentofexistingnationalandinternationalstroke
rehabilitationguidelines.Theidentifiedguidelineswerequalityassuredbytworesearchfellowsusing
theAppraisalofGuidelinesforResearchandEvaluation(AGREEII)instrumentasdescribedinthe
AppendixFTherelevantsectionsoftheguidelinesweresummarised(andnotedwhetherthe
recommendationswerebasedonconsensusorevidence)andthesesummarieswereusedasthe
basisfordraftstatements.Statementswerethendiscussedandrevisedwithtwoexternalexperts
recruitedtoactasconsultantsinthedevelopmentofthesurveystatements.Atablewiththe
relevantguidelinesectionsandfirstdraftstatementcanbefoundinAppendixF.
TheDelphipanelcomprisedofstrokerehabilitationcliniciansandotherprofessionalswithsignificant
experienceinstrokerehabilitation(referredtoastheDelphipanel)coveringawiderangeof
disciplinesinvolvedinstrokecare.Membersofthepanelwereidentifiedbymeansofnominationby
theGDG,andthesewerethencollatedandreviewedbythechairoftheGDGandtheRCP
IntercollegiateStrokeWorkingPartyand,afterremovalofduplicates,inspectedfor
representativeness.Inthefirstinstance164expertswerecontactedandinvitedtoparticipate.The
professionscomprisedof:geriatricians,neurologists,nurses,occupationaltherapists,peoplefrom
patientrepresentation/organisations,physiotherapists,psychologists,research/policymakers,
socialworkers,speechandlanguagetherapists,strokephysiciansandotherhealthcare
professionals(forexampleorthoptists,dieticians,GPsandpharmacists).
Asurvey,consistingof68statementsplus3demographicquestions(profession,setting,and
geographicarea),wasthencirculatedtotheDelphipanel.Freetextboxeswereavailableforpanel
comments,thesewerethenevaluatedandusedtoreviseandrefinestatementsifnecessary.This
processwascarriedoutinconjunctionwiththeconsultantexpertsaswellastheChairofthe
guideline.Theresultsfromeachroundwassummarisedandthencommunicatedtoparticipants.
Fourroundsofthesurveywereundertakenintotal.Forthemajorityofstatements(plus
demographics),aLikertscalewasappliedtoindicatethelevelofagreement.Somestatements
employedmultiplechoiceoptions.AfouroptionLikertscalewasused:stronglydisagree,disagree,
agreeandstronglyagree.ThepurposeofusingafourpointscalewastobeconsistentforDelphi
panelmemberswhomayhavebeenfamiliarwithboththesizeofscaleandtermsusedtosupport
DelphiprocessesfrompreviousconsensusworkinStrokeCare.InpublishedliteratureaboutDelphi
methodologytherehasbeenmuchdebateaboutwhatpercentageofagreementamongDelphipanel
membersconstitutesconsensus(seeMurphyetals1998HealthTechnologyAssessment)181onthis
subject).Whilethereisnouniversalagreementorguidelinesonthelevelofconsensus,Keeneyetal.
(2011)135suggestedthatresearchersshoulddecideontheconsensuslevelbeforecommencingthe
studyandconsiderusingahighlevelofconsensus,suchas70%.
InlinewithKeeneyetal(2011)135alevelof70%orhigherofparticipantsstronglyagreeingwasset
forrounds1and2,withthisthresholdforconsensusbeingreviewedinrounds3and4.Inanalysing
thedata,andinunderstandingthedifficultyofreachingconsensusinthelatterroundswhere
iterationhadfeatured,adecisionwasreachedbythetechnicalteamtolowerthethreshold
marginallyto67%stronglyagreeaslongasthemajorityofotherparticipantresponseswere
agree.Theanalysisofthisineveryitemadoptingthisapproachinthelatterroundswasthatthe
combinedDelphipanelresponsewasinexcessof90%ofparticipantseitherrespondingstrongly
agree(atleast67%oftotalparticipantresponse)oragree.Thiswasapragmaticresponsebythe
technicalteamandmeetspublishedcriteriathatconsensusisachievedwhen66.6%ofaDelphipanel
agrees.Statementsthatreachedtheselevelswouldnotfeatureinthenextround.Statementsthat
didnotreachthislevelwerereviewedbythetechnicalteamwiththeGDGchairandexpert
consultantsandwereamendedbasedonthepanelscommentsinthesurvey.Whentherewerelow
NationalClinicalGuidelineCentre,2013.
55
StrokeRehabilitation
Methods
levelsofdisagreement,somestatementswerenoteditedandreincludedinthenextround.With
alreadylowlevelsofdisagreementitwasfeltthatreinclusionofthesestatementswouldencourage
panelmemberswhoagreedtoshifttoastronglyagreeresponse.Thisprocedureofreevaluation
continueduntileithertheconsensusratewasachievedoruntiltheDelphipanelmembersnolonger
modifiedtheirpreviousestimates/responses(orcomments).Insummary,whenboththelevelof
agreementandthetypeofcommentsnolongerchangeditwasagreedthatafurtherroundwould
notachieveconsensus.Thecommentsthatillustratedthesedifferencesinopinionsorcomments
thatshowedagreementbutnolongerchangedwerethenhighlightedinthefinalDelphireport.
ThereisnocompleteagreementaboutwhentoterminateaDelphisurvey,andoneresearcherhas
statedifnoconsensusemerges,atleastacrystallizingofthedisparatepositionsusuallybecomes
apparent(Gordon,1971)97.
SincetherewasanoverrepresentationofphysiotherapistsintheDelphipanelresponseswere
inspectedbyprofessionintheanalysis.Therewerenosystematicdifferencesinphysiotherapists
responsescomparedtothoseofotherprofessions.Hencefurtherdetailsofresponsesperprofession
werenotincludedinthereport.However,intheGDGmeetinginwhichrecommendationswere
draftedfromtheDelphistatementsGDGmemberswereinformedabouttheDelphicompositionand
askedtoconsiderthisintheirdiscussionofthestatements.
ThefullreportwascirculatedtotheGDG.Theconsensusstatementsemergingfromtheiterative
modifiedDelphitechniquewerepresentedtotheGDGandformedthebasisofdiscussion.The
economicsearchresultswererecheckedtoseeiftherewereanyeconomicanalysesrelatingtoareas
wherenewrecommendationshadbeenmade.Sincenoeconomicevaluationswasfoundonthenew
areasoftheguideline,theGDGmadeaqualitativejudgementaboutthecosteffectivenessofthe
interventionstheywantedtorecommendbasedontheDelphistatements.Economicconsiderations
weredraftedforallthosenewrecommendationswhereeconomicimplicationsweredeemed
important.
Asummaryoftheareasthatareaddressedinthepostconsultationprocessandthetypeofevidence
identifiedisprovidedinTable7below.
Table7:
Summaryofpostconsultationtopicsandlevelofevidenceidentified(consensusrefers
tothoseareasthatwillbecoveredbythemodifiedDelphi.
Areastoaddress
Evidence
servicedelivery
multidisciplinaryteams
strokeunits
consensus
systematicreviewidentified
assessmentforrehab
careplans
goalsetting
ongoingmonitoring
consensus
consensus
systematicreviewidentified
consensus
dischargeplanning/transferofcare
interfacewithsocialcare
consensus
consensus
longtermhealthandsocialsupport
consensus
visualimpairment(diplopia)
consensus
physicalfitness
systematicreviewidentified
speechandlanguagetherapies
aphasia
systematicreviewidentified
NationalClinicalGuidelineCentre,2013.
56
StrokeRehabilitation
Methods
Areastoaddress
apraxia
dysarthria
Evidence
consensus
consensus
shoulderpain
consensus
patientinformation
crossrefertoNICEguidance
consensus
TheGDGformulatednewrecommendationsbasedontheconsensusstatements.ThefullDelphi
reportisinAppendixF
4.6 Developingrecommendations
Overthecourseoftheguidelinedevelopmentprocess,theGDGwaspresentedwith:
Evidencetablesoftheclinicalandeconomicevidencereviewedfromtheliterature.Allevidence
tablesareinAppendicesHandI.
Summaryofclinicalandeconomicevidenceandquality(aspresentedinchapters717).
Forestplots(AppendixJ).
Adescriptionofthemethodsandresultsofthecosteffectivenessanalysisundertakenforthe
guideline(AppendixK).
RecommendationsweredraftedonthebasisoftheGDGinterpretationoftheavailableevidence,
takingintoaccountthebalanceofbenefits,harmsandcosts.Whenclinicalandeconomicevidence
wasofpoorquality,conflictingorabsent,theGDGdraftedrecommendationsbasedontheirexpert
opinion.Theconsiderationsformakinginformalconsensusbasedrecommendationsincludethe
balancebetweenpotentialharmsandbenefits,economicorimplicationscomparedtothebenefits,
currentpractices,recommendationsmadeinotherrelevantguidelines,patientpreferencesand
equalityissues.Theinformalconsensusrecommendationsweredonethroughdiscussionsinthe
GDG.TheGDGmayalsoconsiderwhethertheuncertaintyissufficienttojustifydelayingmakinga
recommendationtoawaitfurtherresearch,takingintoaccountthepotentialharmoffailingtomake
aclearrecommendation(SeeAppendixL).
ThemainconsiderationsspecifictoeachrecommendationareoutlinedintheRecommendations
andlinktoevidencesectionswithineachchapter.
4.6.1
Researchrecommendations
Whenareaswereidentifiedforwhichgoodevidencewaslacking,theguidelinedevelopmentgroup
consideredmakingrecommendationsforfutureresearch.Decisionsaboutinclusionwerebasedon
factorssuchas:
theimportancetopatientsorthepopulation
nationalpriorities
potentialimpactontheNHSandfutureNICEguidance
ethicalandtechnicalfeasibility
4.6.2
Validationprocess
Theguidanceissubjecttoaneightweekpublicconsultationandfeedbackaspartofthequality
assuranceandpeerreviewthedocument.Allcommentsreceivedfromregisteredstakeholdersare
respondedtointurnandpostedontheNICEwebsitewhentheprepublicationcheckofthefull
NationalClinicalGuidelineCentre,2013.
57
StrokeRehabilitation
Methods
guidelineoccurs.Basedoncommentsfromthestakeholdersduringthisconsultationfurtherareas
wereidentifiedwhereguidanceneededinordertoaddressthepatientpathwaymore
comprehensively.ForthisreasonapostconsultationprotocolwasdrawnupandagreedwithNICE
(seesection4.5).Asecondconsultationwasthenheldafterthisextendeddevelopmentperiod.
4.6.3
Updatingtheguideline
Followingpublication,andinaccordancewiththeNICEguidelinesmanual,NICEwillaskaNational
CollaboratingCentreortheNationalClinicalGuidelineCentretoadviseNICEsGuidanceexecutiveon
whethertheevidencebasehasprogressedsignificantlytoaltertheguidelinerecommendationsand
warrantanupdate.
4.6.4
Disclaimer
Healthcareprovidersneedtouseclinicaljudgement,knowledgeandexpertisewhendeciding
whetheritisappropriatetoapplyguidelines.Therecommendationscitedhereareaguideandmay
notbeappropriateforuseinallsituations.Thedecisiontoadoptanyoftherecommendationscited
heremustbemadebythepractitionersinlightofindividualpatientcircumstances,thewishesofthe
patient,clinicalexpertiseandresources.
TheNationalClinicalGuidelineCentredisclaimsanyresponsibilityfordamagesarisingoutoftheuse
ornonuseoftheseguidelinesandtheliteratureusedinsupportoftheseguidelines.
4.6.5
Funding
TheNationalClinicalGuidelineCentrewascommissionedbytheNationalInstituteforHealthand
ClinicalExcellencetoundertaketheworkonthisguideline.
NationalClinicalGuidelineCentre,2013.
58
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5 Organisinghealthandsocialcareforpeople
needingrehabilitationafterstroke
Rehabilitationmaytakeplaceinavarietyofsettings,bothinhospitalandinthecommunity,inout
patientsandintheindividualsownhome.Whatiscriticalisthatwhateverthesetting,peoplewith
strokegetaccesstothelevelofrehabilitationthatmeetstheirneeds.Thischapterconsidersthe
evidenceforthestructureofmultidisciplinarystroketeams,rehabilitationunits,earlysupported
dischargeandtheintensityorrehabilitation.
Asearchforsystematicreviewswascarriedoutforstrokerehabilitationunits,dischargeplanning,
interfacewithsocialcareandmultidisciplinaryteamworking.AnupdateofaCochranesystematic
review251formsthebasisoftherecommendationsregardingstrokerehabilitationservices.Therewas
alackofdirectevidenceformultidisciplinaryteamwork,interfacewithsocialcareanddischarge
planning(seesections5.2,5.3,5.4.4).Thereforerecommendationsinthesesectionswerebasedon
modifiedDelphiconsensusstatementsthatweredrawnupfromexistingnationalandinternational
publishedguidelines.InthesesectionswewillprovidetablesofDelphistatementsthatreached
consensusandstatementsthatdidnotreachconsensusandgiveasummaryofhowtheywereused
todrawuptherecommendations.Fordetailsontheprocessandmethodologyusedforthemodified
DelphisurveyseeAppendixF.
5.1 Strokeunits
5.1.1
EvidenceReview:Inpeopleafterstroke,doesorganisedrehabilitationcare
(comprehensive,rehabilitationandmixedrehabilitationstrokeunits)improveoutcome
(mortality,dependency,requirementforinstitutionalcareandlengthofhospitalstay)?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention
Organisedstrokeunitssuchas:
Strokeward(includingamultidisciplinaryteaminadiscretearea
caringexclusivelyforstrokepatients).Subdividedinto:
o Rehabilitationstrokeunits(acceptingpatientsafteracute
management)
o Comprehensivestrokeunits(combinedacuteaswellas
rehabilitation)
Mixedrehabilitationward(amultidisciplinaryteamincluding
specialistnursingstaffprovidingrehabilitationservices)
Comparison
Generalmedicalward:careinanacutemedicalorneurologyward
withoutroutinemultidisciplinaryinput.
Outcomes
Death
Deathordependency
Deathorinstitutionalcare
Durationofstayinhospitalorinstitutionorboth
Qualityoflife
Patientandcarersatisfaction
NationalClinicalGuidelineCentre,2013.
59
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.1.1.1
ClinicalEvidenceReview
Asearchwasconductedforsystematicreviewscomparingtheclinicaleffectivenessoforganised
strokeunits(comprehensivestrokeunits,rehabilitationstrokeunits,andmixedrehabilitationward)
withgeneralmedicalwardstoimprovehealthoutcomesforadultsandyoungpeople16orolder
whohavehadastroke.
OneCochranesystematicreview251wasidentified.TheCochranerevieworiginallyincluded31trials
(RCTs).Fromthesetrials,weexcludedthosethataddressedanacutepopulation(2weekspost
stroke)andthatcomparedmobilestroketeamtogeneralmedicalwardleaving20trialsthat
matchedourprotocol.These(20)trialswereincludedforthisreview.
AfurthersystematicsearchwasconductedforanytrialpublishedsinceApril2006whichwasthe
searchcutoffdateoftheincludedCochranereview,butnostudieswereidentified.
IntheCochranesystematicreviewthefollowingstrategyofanalysiswasadopted:
Differenttypesoforganisedstrokeunitswerecomparedtogeneralmedicalwards.Thesewere:
o Comprehensivestrokeward
o rehabilitationstrokeward
o mixedrehabilitationstrokeward
Subgroupanalyseswerecarriedoutcomparingcomprehensive,rehabilitation,andmixed
rehabilitationstrokewardstogeneralmedicalwardsfordeath,deathordependency,deathor
institutionalcare(median12months;range6to12months)anddurationofstayinhospitalor
institutionorboth(Table10)
Sensitivityanalysiswasconductedbyexcludingtrialswithahighriskofbias.Thisdidnotaffect
theestimateofeffect
Lengthofstaywascalculatedindifferentways(forexampleacutehospitalstay,totalstayin
hospitalorinstitution).Thesecalculationsweresubjecttomethodologicallimitations
Twotrials126120extendedfollowuptofiveandtenyearspoststroke(Table11)
Patientcarersatisfactionandqualityoflifeoutcomeswereintendedassecondaryoutcomesbuta
metaanalysiswasnotreported
Totalmortalityanddurationofstayinhospitalorinstitutionacrossalltrialsaswellaswithinthe
differentsettingsoforganisedstrokeunitswereanalysed.Forthisreason,intheGRADEtableswe
haveonerowforthetotaleffectaswellasthreeotherrowsforthesubgroups(differentsettingsof
organisedstrokeunit).
Theevidencestatementsalsoreflectthetotaleffectsaswellasthesubgroupanalysis.
PleaseseeAppendixMforexcludedtrials.
Table8:
OverviewofstrokeunitscomparedintheCochranereview
STUDIES
162
Beijing ;
Edinburgh90;
GoteborgOstra253;
GoteborgSahlgren78;
Joinville35;Perth103;
Stockholm273;
Svendborg148;
Trondheim120;Umea
248
NUMBEROF
PARTICIPANTS
2574participants
INTERVENTION
COMPARISON
OUTCOMES
Comprehensive
strokeward
Generalmedical
ward
Death(median
followupof12
months;range
from6weeksto
12months)
*Deathor
dependency
**Deathor
NationalClinicalGuidelineCentre,2013.
60
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
NUMBEROF
PARTICIPANTS
STUDIES
INTERVENTION
COMPARISON
Rehabilitation
strokeward
Generalmedical
ward
Birmingham201;
Helsinki132;Illinois96;
Kuopio239;;NewYork
81
;Newcastle4
Mixed
rehabilitation
ward
247
Table9:
630participants
Generalmedical
ward
OUTCOMES
institutional
care
Durationof
stayinhospital
orinstitutionor
both
Death;deathordependency;deathorinstitutionalcareatfiveand10yearfollowup
NUMBEROF
PARTICIPANTS
STUDIES
126
Nottingham ;
Trondheim120
535participants
INTERVENTION
COMPARISON
OUTCOMES
Rehabilitation
strokeward
Generalmedical
ward
Death
Comprehensive
strokeward
*Deathor
dependency
**Deathor
institutional
care
Note.GMW=GeneralMedicalWard;MRW=MixedRehabilitationWard;inbothTable8andTable9*Dependencyis
definedasarequirementforphysicalattentionsuchasassistancefortransfers,mobility,dressing,feedingor
toileting(andwherecriteriaforindependencewereapproximatelyequivalenttoamodifiedRankinscoreof0
to2,aBarthelIndexofmorethan18outof20oranActivityIndex(AI)ofmorethan83);**Requirementfor
longterminstitutionalcareistakentomeancareinaresidentialhome,nursinghome,orhospitalattheendof
scheduledfollowup.
NationalClinicalGuidelineCentre,2013.
61
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Comparison:Organisedstrokeunitcareversusgeneralmedicalward(medianfollowup12months)
Table10: Organisedstrokeunitcare(comprehensivestrokeward,rehabilitationstrokewardand,mixedrehabilitationward)versusgeneralmedical
wardStudyreferencesandsummaryoffindings
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Deathbytheendofscheduledfollowup
20
Seesub
group
below
(next6
rows)
RCTsingle
blinded
Serious
limitation(a)
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(95%CI)
(%)
Absolute
effect(95%
CI)
Confidenc
e(in
effect)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
374/1932
(19.40%)
410/1807
(24.10%)
RR0.9(0.79 23fewer
to1.01)
per1000
(from48
fewerto2
more)
Moderat
e
291/1259
(23.10%)
RR0.92(0.8 18fewer
to1.06)
per1000
(from46
fewerto14
more)
Moderat
e
DeathbytheendofscheduledfollowupComprehensivestrokewardversusgeneralmedicalward
10
Beijing162;
Edinburgh
90
;
Goteborg
Ostra253;
Goteberg
Sahlgren
78
;Joinville
35
;Perth
RCTsingle
blinded
Serious
Noserious
limitation(a) inconsistency
Noserious
indirectness
Noserious
imprecision
267/1315
(20.30%)
NationalClinicalGuidelineCentre,2013.
62
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
103
;
Stockholm
273
;
Svendborg
148
;
Trondheim
120
;Umea
248
Design
Limitations
Inconsistency
Indirectness
Imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
Absolute
effect(95%
CI)
Confidenc
e(in
effect)
DeathbytheendofscheduledfollowupRehabilitationstrokewardversusgeneralmedicalward
4
Dover
GMW247;
Nottingha
mGMW
126
;
Orpington
1993
GMW128;
Orpington
1995129
RCTsingle
blinded
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
58/285
(20.40%)
68/250
(27.20%)
RR0.77
(0.57to
1.03)
63fewer
per1000
(from117
fewerto8
more)
Low
51/298
(17.10%)
RR0.93
(0.66to
12fewer
per1000
(from58
Low
DeathbytheendofscheduledfollowupMixedrehabilitationwardversusgeneralmedicalward
6
Birmingha
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(f
49/332
(14.80%)
NationalClinicalGuidelineCentre,2013.
63
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
201
m ;
Helsinki132;
Illinois96;
Kuopio239;
NewYork
81
;
Newcastle
4
Limitations
Inconsistency
Indirectness
Imprecision
)
Noserious
indirectness
Noserious
imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
1.31)
Absolute
effect(95%
CI)
fewerto53
more)
Confidenc
e(in
effect)
50fewer
per1000
(from21
fewerto79
fewer)
Moderate
RR0.9(0.82 41fewer
to0.99)
per1000
(from4
fewerto73
fewer)
Moderate
Deathorinstitutionalcarebytheendofscheduledfollowup
19
Seesub
group
below
(next6
rows)
RCTsingle
blinded
Serious
limitation(a)
Noserious
inconsistency
695/1901
(36.60%)
746/1784
(41.80%)
RR0.88
(0.81to
0.95)
DeathorinstitutionalcarebytheendofscheduledfollowupComprehensivestrokewardversusgeneralmedicalward
10
Beijing162;
Edinburgh
90
;
Goteborg
Ostra253;
Goteberg
RCTsingle
blinded
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
477/1315
(36.30%)
511/1259
(40.60%)
NationalClinicalGuidelineCentre,2013.
64
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Sahlgren
78
;Joinville
35
;Perth
103
;
Stockholm
273
;
Svendborg
148
;
Trondheim
120
;Umea
248
Design
Limitations
Inconsistency
Indirectness
Imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
Absolute
effect(95%
CI)
Confidenc
e(in
effect)
62fewer
per1000
(from129
fewerto22
more)
Low
DeathorinstitutionalcarebytheendofscheduledfollowupRehabilitationstrokewardversusgeneralmedicalward
4
Dover
GMW247;
Nottingha
mGMW
126
;
Orpington
1993
GMW128;
Orpington
1995129
RCTsingle
blinded
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
105/283
(37.10%)
111/250
(44.40%)
DeathorinstitutionalcarebytheendofscheduledfollowupMixedrehabilitationwardversusgeneralmedicalward
NationalClinicalGuidelineCentre,2013.
65
RR0.86
(0.71to
1.05)
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
5
RCTsingle
132
Helsinki ; blinded
Illinois96;
Kuopio239;
NewYork
81
;
Newcastle
4
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
Absolute
effect(95%
CI)
Confidenc
e(in
effect)
Limitations
Inconsistency
Indirectness
Imprecision
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
113/303
(37.30%)
124/275
(45.10%)
RR0.82
(0.68to
0.99)
81fewer
per1000
(from5
fewerto
144fewer)
Moderat
e
Noserious
indirectness
Noserious
imprecision
792/1415
(56%)
836/1346
(62.10%)
RR0.89
(0.84to
0.95)
68fewer
per1000
(from31
fewerto99
fewer)
Moderat
e
RR0.89
(0.82to
0.97)
67fewer
per1000
(from18
fewerto
110fewer)
Moderate
Deathordependencybytheendofscheduledfollowup
17
Seesub
group
below
(next6
rows)
RCTsingle
blinded
Serious
limitation(a)
Noserious
inconsistency
DeathordependencybytheendofscheduledfollowupComprehensivestrokewardversusgeneralmedicalward
7
Beijing162;
Edinburgh
90
;
Goteberg
Sahlgren
78
;Joinville
RCTsingle
blinded
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
448/800
(56%)
487/798
(61%)
NationalClinicalGuidelineCentre,2013.
66
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
35
;Perth
103
;
Trondheim
120
;Umea
248
Design
Limitations
Inconsistency
Indirectness
Imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
Absolute
effect(95%
CI)
Confidenc
e(in
effect)
RR0.95
(0.85to
1.06)
36fewer
per1000
(from107
fewerto43
more)
Moderat
e
RR0.83
(0.71to
0.96)
98fewer
per1000
(from23
fewerto
Moderat
e
DeathordependencybytheendofscheduledfollowupRehabilitationstrokewardversusgeneralmedicalward
4
RCTsingle
blinded
Dover
GMW247;
Nottingha
mGMW
126
;
Orpington
1993
GMW128;
Orpington
1995(Kalra
1995)129
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
189/283
(66.80%)
178/250
(71.20%)
DeathordependencybytheendofscheduledfollowupMixedrehabilitationwardversusgeneralmedicalward
6
RCTsingle
blinded
Birmingha
201
m ;
Helsinki132;
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
155/332
(46.70%)
171/298
(57.40%)
NationalClinicalGuidelineCentre,2013.
67
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Illinois96;
Kuopio239;
NewYork
81
;
Newcastle
4
Design
Limitations
Inconsistency
Indirectness
Imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
SeeForest
plotsfor
study
meansand
SDs
SeeForest
plotsfor
study
meansand
SDs
Absolute
effect(95%
CI)
166fewer)
Confidenc
e(in
effect)
Lengthofstay(days)inahospitalorinstitution(Betterindicatedbylowervalues)
16
Seesub
group
below
(next6
rows)
RCTsingle
blinded
Serious
limitation(a)
Serious
inconsistency(c
)
Noserious
indirectness
Noserious
imprecision
0.09(
0.24,0.05)
SMD0.09
Low
lower(0.24
lowerto
0.05higher)
Lengthofstay(days)inahospitalorinstitutionComprehensivestrokewardversusgeneralmedicalward(Betterindicatedbylowervalues)
10
Beijing162;
Edinburgh
90
;
Goteborg
Ostra253;
Goteberg
Sahlgren
78
;Joinville
RCTsingle
blinded
Serious
limitation(a)
Serious
inconsistency(d
)
Noserious
indirectness
Noserious
imprecision
SeeForest
plotsfor
study
meansand
SDs
SeeForest
plotsfor
study
meansand
SDs
NationalClinicalGuidelineCentre,2013.
68
0.19(
0.35,0.02)
SMD0.19
lower(0.35
to0.02
lower)
Low
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
35
;Perth
103
;
Stockholm
273
;
Svendborg
148
;
Trondheim
120
;Umea
248
Design
Limitations
Inconsistency
Indirectness
Imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
Absolute
effect(95%
CI)
Confidenc
e(in
effect)
Lengthofstay(days)inahospitalorinstitutionRehabilitationstrokewardversusgeneralmedicalward(Betterindicatedbylowervalues)
3
Dover
GMW247;
Nottingha
mGMW
126
;
Orpington
1993
GMW128
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(e
)
Dover
GMW:0(0)
Nottingham
GMW:
76.72
(39.37)
Orpington
1993GMW:
0(0)
Dover
GMW:0(0)
Nottingham
GMW:
60.38
(48.91)
Orpington
1993
GMW:0(0)
0.37(0.07,
0.67)
SMD0.37
Moderat
higher(0.07 e
to0.67
higher)
Lengthofstay(days)inahospitalorinstitutionMixedrehabilitationwardversusgeneralward(Betterindicatedbylowervalues)
3
RCTsingle
132
Helsinki ; blinded
Kuopio239;
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Helsinki:
23.6(38.8)
Kuopio:
Helsinki:
30.5(70.6)
Kuopio:
NationalClinicalGuidelineCentre,2013.
69
0.08(0.21,
0.37)
SMD0.08
High
higher(0.21
lowerto
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Newcastle
4
Design
Limitations
Inconsistency
Indirectness
Imprecision
Organised
strokeunit
care
Mean(SD)/
Frequency
(%)
162.5(125)
Newcastle:
52(45)
Risk
ratio(RR)/
General
Standardise
medical
dMean
wards
Difference
Mean(SD)/ (SMD)/
Frequency Relative
(%)
(95%CI)
129.5(119)
Newcastle:
41(34)
Absolute
effect(95%
CI)
0.37higher)
Confidenc
e(in
effect)
(a)
(b)
(c)
(d)
(e)
(f)
Unclearrandomisation;unclearallocationconcealment.Limitationswereconsideredbystudyweightsinthemetaanalysis
ConfidenceintervalcrossesoneendofdefaultMID(0.75)
Heterogeneity;I2=73%
Heterogeneity;I2=74%
ConfidenceintervalcrossesoneendofdefaultMID(0.5)
ConfidenceintervalcrossesbothendsofdefaultMID(0.75;1.25)
Comparison:Comprehensive/rehabilitationstrokeunitversusgeneralmedicalward(longtermfollowup)
Table11: Comprehensive/rehabilitationstrokeunitversusgeneralmedicalwardClinicalstudycharacteristicsandclinicalsummaryoffindings
Qualityassessment
Noof
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Organised
General
NationalClinicalGuidelineCentre,2013.
70
Effect
Confidence
(ineffect)
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
studies
strokeunit
Mean(SD)/
Frequency
(%)
medical
ward
Mean
(SD)/
Frequency
(%)
Riskratio(RR)/
Standardised
Mean
Difference
(SMD)/
(95%CI)
Absolute
effect/
Standardis
edMean
Difference
(SMD)
(95%CI)
Deathatfiveyearfollowup
2
Nottingh
am126;
Trondhei
m120
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
a)
144/286
(50.30%)
155/249
(62.20%)
RR0.82(0.71
to0.95)
112fewer Moderate
per1000
(from31
fewerto
181fewer)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
172/286
(60.10%)
178/249
(71.50%)
RR0.85(0.75
to0.96)
107fewer High
per1000
(from29
fewerto
179fewer)
Noserious
limitation
Serious
inconsistency(b
)
Noserious
indirectness
Noserious
imprecision
223/286
(78%)
214/249
(85.90%)
RR0.91(0.84
to0.99)
77fewer
Moderate
per1000
(from9
fewerto
138fewer)
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
205/286
(71.70%)
207/249
(83.10%)
RR0.87(0.79
to0.95)
108fewer High
per1000
(from42
fewerto
175fewer)
Deathorinstitutionalcareatfiveyearfollowup
2
Nottingh
am126;
Trondhei
m120
RCTsingle
blinded
Noserious
limitation
Deathordependencyatfiveyearfollowup
2
Nottingh
am126;
Trondhei
m120
RCTsingle
blinded
Deathat10yearfollowup
2
Nottingh
am126;
Trondhei
m120
RCTsingle
blinded
Deathorinstitutionalcareat10yearfollowup
NationalClinicalGuidelineCentre,2013.
71
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Qualityassessment
Summaryoffindings
Effect
Noof
studies
2
Nottingh
am126;
Trondhei
m120
Organised
strokeunit
Mean(SD)/
Frequency
(%)
General
medical
ward
Mean
(SD)/
Frequency
(%)
Riskratio(RR)/
Standardised
Mean
Difference
(SMD)/
(95%CI)
Absolute
effect/
Standardis
edMean
Difference
(SMD)
(95%CI)
Confidence
(ineffect)
Design
Limitations
Inconsistency
Indirectness
Imprecision
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
220/286
(76.90%)
214/249
(85.90%)
RR0.9(0.83to
0.98)
86fewer
High
per1000
(from17
fewerto
146fewer)
Serious
inconsistency(c
)
Noserious
indirectness
Noserious
imprecision
249/286
(87.10%)
224/249
(90%)
RR0.97(0.91
to1.03)
27fewer
per1000
(from81
fewerto
27more)
Deathordependencyat10yearfollowup
2
Nottingh
am126;
Trondhei
m120
RCTsingle
blinded
Noserious
limitation
(a) ConfidenceintervalcrossesoneendofdefaultMID(0.75)
(b) Heterogeneity;I2=64%
(c) Heterogeneity;I2=51%
NationalClinicalGuidelineCentre,2013.
72
Moderate
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.1.1.2
Economicevidence
Twostudiesthatincludedtherelevantcomparisonarereviewed44,176.Thesearesummarisedinthe
economicevidenceprofilebelow(Table12andQALYsnotused
(a)
(b)
(c)
(d)
(e)
(f)
SomeuncertaintyaboutapplicabilityofnonUKresourceuseandunitcosts
Someuncertaintyaboutapplicabilityofresourceuseandunitcostsfromover10yearsago
Someuncertaintyininterpretingtheresultsoftheanalysisintermsofthehealthoutcomes
Nosensitivityanalysis
Costingisbasedonthepracticeofonehospitalsouncertaintyastowhetheritreflectsnationalcosts
Someuncertaintyaboutthecomparabilityofthehealthoutcomesintheanalysistothosespecifiedinthereview
protocol
Table13).SeealsothefullstudyevidencetablesinAppendixI.
Onestudy(Major,1996165)thatmettheinclusioncriteriawasselectivelyexcludeddueto
methodologicallimitations.
Table12: StrokeunitsversusgeneralmedicalwardcareEconomicstudycharacteristics
Study
44
Claesson2000
(Sweden)
Applicability Limitations
Othercomments
Partially
applicable
(a)(b)(c)
Costconsequenceanalysis(varioushealthoutcomes)
Acutestrokeunitswerelinkedtoageriatricwardfor
longertermrehabilitation
Withintrialanalysis,clinicaleffectivenessdatareported
separatelyinFagerberg200078(includedinclinical
review)
Potentially
serious
limitations
(e)(f)
Moodie2006176
(Australia)
Partially
applicable
(a)(b)(d)
Veryserious
limitations
(e)(g)
Costeffectivenessanalysis(healthoutcomes=
thoroughadherencetodefinedprocessofcare
measuresandratesofseveremedicalcomplications)
Strokecareunitvs.generalmedicalward
Withintrialanalysis
(g) QALYsnotused
(h) SomeuncertaintyaboutapplicabilityofnonUKresourceuseandunitcosts
(i) Someuncertaintyaboutapplicabilityofresourceuseandunitcostsfromover10yearsago
(j) Someuncertaintyininterpretingtheresultsoftheanalysisintermsofthehealthoutcomes
(k) Nosensitivityanalysis
(l) Costingisbasedonthepracticeofonehospitalsouncertaintyastowhetheritreflectsnationalcosts
(m) Someuncertaintyaboutthecomparabilityofthehealthoutcomesintheanalysistothosespecifiedinthereview
protocol
Table13: StrokeunitsversusgeneralmedicalwardcareEconomicsummaryoffindings
Incremental
cost
Study
44
Incrementaleffects
Costeffectiveness
Uncertainty
Claesson2000
(Sweden)
Saves845
(a)
Nosignificantdifference
N/A
NR
Moodie2006176
(Australia)
1553
(b)
Higheradherencetoprocess
indicatorsandreducedrate
ofseveremedical
complicationswasobserved
onstrokeunits
4891perpatientwith
thoroughadherence
gained
8116perpatientwith
severecomplications
avoided
NR
N/A=notapplicable;NR=notreported
(a) ConvertedtoUKpoundsusingexchangeratequotedinthestudy
NationalClinicalGuidelineCentre,2013.
73
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
(b) ConvertedtoUKpoundsusingrelevantpurchasingpowerparities194
5.1.1.3
Evidencestatements
Clinicalevidencestatements
Deathbytheendofscheduledfollowup
Twentystudiescomprising3739participantsfoundnosignificantdifferenceinrateofmortality
betweenorganisedstrokeunits(comprehensive,rehabilitationand,mixedrehabilitationwards)and
generalmedicalwardbytheendofscheduledfollowup(MODERATECONFIDENCEINEFFECT).
Tenstudies162902537835103273148120248comprising2574participantsfoundnosignificant
differenceinrateofmortalitybetweencomprehensivestrokewardandgeneralmedical
wardbytheendofscheduledfollowup(MODERATECONFIDENCEINEFFECT).
Fourstudies247126128129comprising535participantsfoundnosignificantdifferenceinrate
ofmortalitybetweenrehabilitationstrokewardandgeneralmedicalwardbytheendof
scheduledfollowup(LOWCONFIDENCEINEFFECT).
Sixstudies20113296239814comprising630participantsfoundnosignificantdifferenceinrate
ofmortalitybetweenmixedrehabilitationwardandgeneralmedicalwardbytheendof
scheduledfollowup(LOWCONFIDENCEINEFFECT).
Deathorinstitutionalcarebytheendofscheduledfollowup
Nineteenstudiescomprising3685participantsfoundthatsignificantlyfewerpeopleintheorganised
strokeunit(comprehensive,rehabilitationand,mixedrehabilitationwards)diedorrequired
institutionalcarebytheendofscheduledfollowupcomparedtogeneralmedicalward(MODERATE
CONFIDENCEINEFFECT).
Tenstudies162902537835103273148120248comprising2574participantsfoundthatsignificantly
fewerpeopleincomprehensivestrokewarddiedorrequiredinstitutionalcarebytheendof
scheduledfollowupcomparedtogeneralmedicalward(MODERATECONFIDENCEIN
EFFECT).
Fourstudies247126128129comprising533participantsfoundnosignificantdifferenceinrate
ofmortalityorinstitutionalcarebetweenrehabilitationstrokewardandgeneralmedical
wardbytheendofscheduledfollowup(LOWCONFIDENCEINEFFECT).
Fivestudies13296239814comprising578participantsfoundthatsignificantlyfewerpeoplein
themixedrehabilitationwarddiedorrequiredinstitutionalcarebytheendofscheduled
followupcomparedtogeneralmedicalward(MODERATECONFIDENCEINEFFECT).
Deathordependencybytheendofscheduledfollowup
Seventeenstudiescomprising2763participantsfoundthatsignificantlyfewerpeopleinorganised
strokeunit(comprehensive,rehabilitationand,mixedrehabilitationwards)diedorweredependent
bytheendofscheduledfollowupcomparedtogeneralmedicalward(MODERATECONFIDENCEIN
EFFECT).
Sevenstudies162907835103120248comprising1598participantsfoundthatsignificantlyfewer
peopleincomprehensivestrokewarddiedorweredependentbytheendofscheduled
followupcomparedtogeneralmedicalward(MODERATECONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
74
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Fourstudies247126128129comprising535participantsfoundnosignificantdifferenceinrate
ofmortalityordependencybetweentherehabilitationstrokewardandgeneralmedical
wardbytheendofscheduledfollowup(MODERATECONFIDENCEINEFFECT).
Sixstudies20113296239814comprising630participantsfoundthatsignificantlyfewerpeoplein
themixedrehabilitationwarddiedorweredependentbytheendofscheduledfollowup
comparedtogeneralmedicalward(MODERATECONFIDENCEINEFFECT).
Lengthofstay(days)inhospitalorinstitution
Sixteenstudiescomprising3121participantsfoundnosignificantdifferenceinlengthofstay(days)in
hospitalorinstitutionorbothbetweenorganisedstrokeunits(comprehensive,rehabilitation,and
mixedrehabilitationstrokewards)andgeneralmedicalwards(LOWCONFIDENCEINEFFECT).
Tenstudies162902537835103273148120248comprising2556participantsfoundastatistically
significantdifferenceinlengthofstay(days)inhospitalorinstitutioninfavourof
comprehensivestrokewardcomparedtogeneralmedicalward(LOWCONFIDENCEIN
EFFECT).
Threestudies247126128comprising178participantsfoundastatisticallysignificantdifference
inlengthofstay(days)inahospitalorinstitutioninfavourofgeneralmedicalward
comparedtorehabilitationstrokeward(MODERATECONFIDENCEINEFFECT).
Threestudies1324,239comprising387participantsfoundnosignificantdifferenceinlengthof
stay(days)inhospitalorinstitutionbetweenmixedrehabilitationwardandgeneralmedical
ward(HIGHCONFIDENCEINEFFECT).
Deathatfiveyearfollowup
Twostudies126120comprising535participantsfoundthatsignificantlyfewerpeopleintheorganised
strokeunit(comprehensiveandrehabilitationstrokewards)diedatfiveyearfollowupcomparedto
generalmedicalward(MODERATECONFIDENCEINEFFECT).
Deathorinstitutionalcareatfiveyearfollowup
Twostudies126120comprising535participantsfoundthatsignificantlyfewerpeopleintheorganised
strokeunit(comprehensiveandrehabilitationstrokeward)diedorrequiredinstitutionalcareatfive
yearfollowupcomparedtogeneralmedicalward(HIGHCONFIDENCEINEFFECT).
Deathordependencyatfiveyearfollowup
Twostudies126120comprising535participantsfoundthatsignificantlyfewerpeopleintheorganised
strokeunit(comprehensiveandrehabilitationstrokeward)diedorweredependentatfiveyear
followupcomparedtogeneralmedicalward(MODERATECONFIDENCEINEFFECT).
Deathat10yearfollowup
Twostudies126120comprising535participantsfoundthatsignificantlyfewerpeopleintheorganised
strokeunit(comprehensiveandrehabilitationstrokeward)diedat10yearfollowupcomparedto
generalmedicalward(HIGHCONFIDENCEINEFFECT).
Deathorinstitutionalcareat10yearfollowup
Twostudies126120comprising535participantsfoundthatsignificantlyfewerpeopleintheorganised
strokeunit(comprehensiveandrehabilitationstrokeward)diedorrequiredinstitutionalcareat10
yearfollowupcomparedtogeneralmedicalward(HIGHCONFIDENCEINEFFECT).
Deathordependencyat10yearfollowup
NationalClinicalGuidelineCentre,2013.
75
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Twostudies126120comprising535participantsfoundnosignificantdifferenceinrateofmortalityor
dependencybetweentheorganisedstrokeunit(comprehensiveandrehabilitationstrokeward)and
generalmedicalwardat10yearfollowup(MODERATECONFIDENCEINEFFECT).
Economicevidencestatements
Onepartiallyapplicablestudywithpotentiallyseriouslimitationsshowedthatthecostsper
patientinastrokeunitwaslowercomparedtoageneralmedicalwardwithnosignificant
differenceintermsofhealthoutcomes.
Onepartiallyapplicablestudywithveryseriouslimitationsshowedthatcareonstokeunitscost
morethancareongeneralmedicalwards.However,thequalityofcaredeliveredonstrokeunits
wasmuchhigher.
NationalClinicalGuidelineCentre,2013.
76
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.1.2
Recommendationsandlinkstoevidence
1. Peoplewithdisabilityafterstrokeshouldreceive
rehabilitationinadedicatedstrokeinpatientunitand
subsequentlyfromaspecialiststroketeamwithinthe
community.
2. Aninpatientstrokerehabilitationserviceshouldconsistof
thefollowing:
adedicatedstrokerehabilitationenvironment
acoremultidisciplinaryteam(seerecommendation3)
whohavetheknowledge,skillsandbehaviourstoworkin
partnershipwithpeoplewithstrokeandtheirfamiliesand
carerstomanagethechangesexperiencedasaresultofa
stroke.
accesstootherservicesthatmaybeneeded,forexample:
- continenceadvice
- dietetics
- electronicaids(forexample,remotecontrolsfor
doors,lightsandheating,andcommunicationaids)
- liaisonpsychiatry
- orthoptics
- orthotics
- pharmacy
- podiatry
- wheelchairservices
amultidisciplinaryeducationprogramme.
Recommendation
Relativevaluesofdifferent
outcomes
DeathordependencyorinstitutionalcarewereconsideredbytheGDG
tobethemostcriticaloutcomesqualityoflifeandpatientandcarer
satisfactionwerealsoimportantoutcomes.Durationofstayinhospital
orinstitutionorboth,wasseenaslessimportantoutcomessincesuch
measuresareoftenveryvariableandoftenaffectedbyoutliers.The
Cochranereviewreporteddeath,admittancetoinstitutionalcareand
lengthofhospitalstayasoutcomes.
Tradeoffbetweenclinical
benefitsandharms
TheGDGagreedthatthereisclearevidencethatoutcomesforpatients
withresidualdisabilityarebetterwhenmanagedinadedicatedstroke
rehabilitationunitattheposttwoweekperiodafterstroke.Thishas
beendemonstratedbothinthepapersconsideredbutalsofrom
experienceinclinicalpractice.TheGDGacknowledgedthatfromthe
rehabilitationunitpeoplewouldbeassessedforsuitabilityforearly
supporteddischargeortoremainonthestrokerehabilitationunit.No
harmswereassociatedwithcareintheseunits.
Economicconsiderations
TheGDGrecognisedthattheavailabilityofstrokeunitsisstandard.
Strokeunitsareexpectedtobemoreexpensivethangeneralmedical
wardduetoprovisionofmorespecialisedservicesandincreased
NationalClinicalGuidelineCentre,2013.
77
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
resourceuseforexample theuseofmorespecialisedstaff.
Aneconomicstudyshowedthatthecostsperpatientinastrokeunit
waslowercomparedtoageneralmedicalwardwithnosignificant
differenceintermsofhealthoutcomes,whileanothereconomicstudy
showedthatcareonstokeunitscostmorethancareongeneralmedical
wardsbutthequalityofcaredeliveredonstrokeunitswasmuchhigher.
Theeconomicstudiesincludedintheeconomicliteraturerevieware
basedonsingletrials,whereastheNCGCclinicalreviewpoolstheoverall
effectivenessofstrokeunitsfromseveralRCTs.Thepotentialbenefits
(decreasedmortality,decreaseddependencyandneedfor
institutionalisedcare)ofdedicatedstrokeunitsarethoughttobelikely
tooffsetthecosts.
Qualityofevidence
Theveryacutestrokepopulation(2weekspoststoke)wasexcluded
fromthisreviewbecausethispopulationhasalreadybeenaddressedin
theStrokeguideline(CG68).Thosestudiesthataddressedmobilestroke
unitswerealsoexcludedastheGDGagreedtreatmentwouldnotbe
providedviathismeansanymore.
TheincludedstudiesintheCochranereviewhadlargenumbersof
participants.Theconfidenceintheeffectofspecifiedoutcomesranged
fromlowtohighwiththemajoritybeingmoderate.
Organisedstrokeunitsshowedasignificantreductionindeathor
institutionalcareanddeathordependencyattheendofscheduled
followup.
Oftheorganisedstrokeunits,thecomprehensiveandrehabilitation
strokewardshowedasignificantreductionindeath;deathor
institutionalcareatfiveandtenyearfollowup;andareductionindeath
ordependencyatfiveyearfollowup.
Theevidencewasfoundtobeveryrobustforstrokerehabilitationunits
andmustremainamajorcomponentforstrokecarepathway.
Otherconsiderations
Strokerehabilitationunitsprovideanenvironmentforappropriate
assessmentforongoingcareandsupportforpeopleafterstroke.
Thedefinitionofwhataspecialiststrokerehabilitationserviceshould
consistofwastakenfromtheStrokeUnitTrialistsCollaboration
outlinedintheCochranereview251.TheGDGagreedthatthiswas
universallyacceptedandalthoughtheevidencecomesfrominpatient
strokeunitsitisequallyappropriateforearlysupporteddischarge
communityteams.TheGDGrecognisedthatstrokeisamultifaceted
conditionandthataccesstoservicesoutsidethosethatcanbeprovided
byacoremultidisciplinaryteamisimportant.ThereforetheGDG
specifiedtheseinthedescriptionoftheinpatientstrokerehabilitation
service.
5.2 Thecoremultidisciplinarystroketeam
5.2.1
EvidenceReview:Whatshouldbetheconstituencyofamultidisciplinaryrehabilitation
teamandhowshouldtheteamworktogethertoensurethebestoutcomesforpeople
whohavehadastroke?
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Components
Constituencyofamultidisciplinaryrehabilitationteam
Workingpractices,suchascommunicationandcoordinationofservices(teamand
familymeetings,coordinationofcarebetweenrehabilitationspecialtiesandother
agencies)
Outcomes
Patientandcarersatisfaction
NationalClinicalGuidelineCentre,2013.
78
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.2.2
Optimisedstrategiestominimiseimpairmentandmaximiseactivity/participation
Delphistatementswhereconsensuswasachieved
Table14: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Number
Statement
Acorestrokerehabilitationteam
shouldcompriseofmembershipfrom
thefollowingdisciplines:
Consultantneurology/strokemedicine
Nursing
Physiotherapy
Occupationaltherapy
SpeechandLanguageTherapy
ClinicalNeuropsychology
RehabilitationAssistant
Socialwork
81.0
89.1
99.0
99.0
99.0
74.0
72.2
71.2
28/101(28%)panelmembers
commented
Pharmacistsand
Nutritionist/Dieticiandidnotreach
consensus
Someotheroptionalteam
membersweresuggestedin
comments,forinstance:
Orthoptists
Counsellor
Familyorpatientsupportworker
Accesstorelevantotherssuchas
peerswithstroke,information
navigators,voluntarysector
organisations
Anopinionwasexpressedthat
differentspecialistsarerequiredat
differentstagesofrehabilitation
(Thecoreteamshouldbeavailable
althoughitisrecognisedthatat
differentstagesoftherehabilitation
pathwayanddependingonthe
needsofthepatientthelevelof
theseinputsmayvary.)
Theimportanceofvoluntarysector
involvementwasstatedwithregards
totheroleofacoordinator(This
rolecouldbeprovidedbythe
voluntarysector,thebestexample
beingtheStrokeAssociations
information,adviceandsupportco
ordinators.).
Throughoutthecarepathwayroles
andresponsibilitiesofthemulti
disciplinarystrokerehabilitationteam
servicesshouldbeclearlyoutlined,
documentedandcommunicatedto
thepatientandtheirfamily.
72.7
18/99(18%)panelmembers
commented
Informationtothefamilyofthe
personwhohashadastrokeshould
NationalClinicalGuidelineCentre,2013.
79
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.2.3
Number
Statement
Inordertoinformanddirectfurther
assessment,membersoftheMDT
shouldscreenthepersonwhohashad
astrokeforarangeofimpairments
anddisabilities.
Results
%
81.0
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
onlybegivenwithpatientsconsent
Communicationwasviewedas
integralinrehabilitationprocess
Extracts:
Verbalcommunicationshouldbe
supportedbyclear,unambiguous
writteninformationtoavoidany
subsequentdisputes/confusion.
Ithinkithelpscommunicationfor
patientsandstaff,howeverthe
frequencyandprocessofthishasto
berealisticinitsdelivery.
9/100(9%)commented:
Reliabilityandvalidityofscreening
instrumentswashighlighted
Reasonforscreening:
Screeningtoinformtreatment/
furtherassessmentratherthan
screeningforscreeningssake
Treatment:
Somepeoplecommentedthatthe
focusinstrokerehabilitationshould
beontreatmentratherthan
measurement.
Delphistatementwhereconsensuswasnotreached
Table15: Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
1.
Statement
Results
Thepersonwhohashadastrokeis
integratedinthestrokerehabilitation
team.
62.9
NationalClinicalGuidelineCentre,2013.
80
Amountandcontentofpanel
commentsorthemes
26/100(26%)panelmembers
commentedinround2;29/84(35%)
commentedinround3and24/70
(34%)commentedinround4:
Impairmentsofthepersonswho
havehadastrokethataffect
participationshouldbeconsidered
forthisstatement.(Some
individualscaneasilymakeavery
activeandsubstantialcontribution
totheworkoftheteamwhereas
othersbecauseoftheseverityofthe
strokeorofanycommunication
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Number
2.
Statement
Results
Amemberofthemultidisciplinary
strokerehabilitationteamshouldbe
taskedwithcoordinationandsteering
(forexamplecommunication,family
liaisonandgoalplanning)ofthe
rehabilitationofthepersonwhohas
hadastrokeateachstageofthecare
pathway.
62.5
NationalClinicalGuidelineCentre,2013.
81
Amountandcontentofpanel
commentsorthemes
difficultieswouldbemuchmore
limited.)
Patientpreference:
Itmaynotbethewishoftheperson
whohashadastroketoparticipate
intheteam(WhenIneedcareor
helpIwishtobetreatedwith
respect,dignityandasanequal,butI
viewtheMDTaspeoplewhosupport
me,advisemeandhaveclinical
expertise,theyaretheteamwho
helpme.).
Betweenrounds3and4the
statementwaschangedfrom:
isamemberof
to
isintegratedinthestroke
rehabilitationteam.
Mostpanelmembersobjectedto
theconceptofteammembership.
Theconceptofmembershipas
opposedtopartnershipwas
highlighted
Twopanelmembersexpressedthe
opinionthatthisstatementwas
redundant.
Adirectpromptwasgivenforthis
question(tolisttheroles).Inround2
61/100(61%)panelmembers
commented;48/85(56%)inround3
and34/72(47%)inround4:
Therewasalistofpossiblerolesfor
acoordinator:
Communication
Goalplanning
Familyliaison
Keyworking
Dischargeplanning
Singlepointofcontact
fewteamscoverthewholeofthe
strokecarepathwayandthiswould
notworkpractically.
whereamemberoftheteamis
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Number
5.2.4
Statement
Results
Amountandcontentofpanel
commentsorthemes
responsible,theprocessbecomes
sloweddown.
RecommendationsandlinkstoDelphiconsensussurvey
Statements
Recommendations
1. Acorestrokerehabilitationteamshouldcompriseofmembership
fromthefollowingdisciplines:
Consultantneurology/strokemedicine
Nursing
Physiotherapy
Occupationaltherapy
SpeechandLanguageTherapy
ClinicalNeuropsychology
Rehabilitationassistant
Socialworker
2. Throughoutthecarepathwayrolesandresponsibilitiesofthemulti
disciplinarystrokerehabilitationteamservicesshouldbeclearly
outlined,documentedandcommunicatedtothepatientandtheir
family.
3. Inordertoinformanddirectfurtherassessment,membersofthe
MDTshouldscreenthepersonwhohashadastrokeforarangeof
impairmentsanddisabilities.
3. Acoremultidisciplinarystrokerehabilitationteamshouldcomprise
thefollowingprofessionalswithexpertiseinstrokerehabilitation:
consultantphysicians
nurses
physiotherapists
occupationaltherapists
speechandlanguagetherapists
clinicalpsychologists
rehabilitationassistants
socialworkers.
4. Throughoutthecarepathway,therolesandresponsibilitiesofthe
coremultidisciplinarystrokerehabilitationteamshouldbeclearly
documentedandcommunicatedtothepersonandtheirfamilyor
carer.
5. Membersofthecoremultidisciplinarystroketeamshouldscreenthe
personwithstrokeforarangeofimpairmentsanddisabilities,in
ordertoinformanddirectfurtherassessmentandtreatment.
Otherconsiderations
SomeconcernwasexpressedthatasaresultoftheDelphisurvey
thepatientandfamilymemberswerenotpartoftheMDT(oneof
NationalClinicalGuidelineCentre,2013.
82
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
thestatementsthatdidnotreachconsensus).Alotofcomments
hadbeenmadeonthiswithinthesurvey,andtheGDGthoughtit
maybebecauseofdifferentinterpretationsofthemeaningofthe
termteam.TheMDTismadeupofagroupofprofessionalswho
areemployedtodeliveraservice,andalthoughthepatientand
familymemberswouldbeinvolvedtheywouldnotbeconsidered
asanintrinsicpartoftheteamdeliveringarehabilitationservice.It
wasagreedthatitwasimportantthatthepatientisclearwhatthe
teamsfunctionisandwhateachindividualroledoes.TheGDG
acknowledgedthatthereisalackofinformationforpatientsand
theirfamiliesonthestructureofthestrokepathway,andon
individualteammembersresponsibilities.Itisoftenjustassumed
patientsalreadyhaveanunderstandingofwhatrehabilitation
servicesare.
Itwasthereforefeltthatdocumentingandcommunicatingthiswas
veryimportant.Therewasadiscussionofwhetheritwaspossible
toprovideaclearerdescriptionofhowthiswouldtakeplacein
practice.However,theGDGcametotheconclusionthatthere
wouldbeawidevariationdependingonwhereinthecarepathway
peoplewouldbe,andaccordingtoindividualdifficultiesand
priorities.TheGDGthereforedidnotwanttobetooprescriptive
aboutthisprocess.
Thegroupacknowledgedthatwhilststatingaclinicalneuro
psychologistwouldbetheideal,itwasnotrealisticastherewere
notenoughoftheseprofessionalscurrentlyavailable.Thereforea
recommendationforaclinicalpsychologistwasmade.Howeverthe
groupwereinstrongagreementthatpsychologicalservicesshould
beacorepartoftheMDTandthiswasnotalwaysthecaseat
present.Althoughconsensuswasreachedforaconsultant
neurologist/strokemedicinethiswasmodifiedbythegroupin
recognitionthatstrokemedicineinthiscountryisoneyeartraining
andphysicianscomefromavarietyofdifferenthostroutes.
TheGDGrecognisedthattherewerearangeofotherservicesthat
peoplemayrequireafterastroke,notcoveredbythecoreMDT,
butvitalinprovidingacomprehensiveservice.TheGDGalsoraised
theimportanceofprovidingguidanceonaccesstoarangeof
servicesthatmayberequiredandtheimportanceofspeedy
referraltootherhealthprofessionalexpertisesuchasdieticians,
continenceadvisors,orthoptists,orthotistsorpharmacy.A
recommendationwasthereforeincludedprovidingguidanceon
accesstoservicesoutsidethecoreteambasedoncommentsfrom
theDelphipanel(seerecommendation2).
NationalClinicalGuidelineCentre,2013.
83
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.3 Healthandsocialcareinterface
5.3.1
Delphistatementswhereconsensuswasachieved
Table16: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Number
Statement
1.
Whereappropriate,socialworkers
shouldbeinvolvedwiththestroke
rehabilitationteamintheassessment
ofposthospitalcareneeds.
2.
Theroleofsocialcareandanyservice 72.7
provisionrequiredshouldbediscussed
withthepersonwhohashadastroke
anddocumentedwithinthesocialcare
plan.
10/99(10%)panelmembers
commented
Afewpanelmembershighlighted
therelationshipbetweenthis
statementandthejointcareplan
andthatthereshouldbeaccessto
onesetofnotes.
Acoupleofpeoplethoughtthatthis
shouldbediscussedfullywiththe
personwhohashadastrokeand
withthecarerornearestrelative.
Inanothercommentitwasstated
thatitisnotnecessarytodiscussthe
wholeplanwiththepersonwhohas
hadastrokeincasetheamountof
informationwasoverwhelming
3.
Whensocialneedsareidentifiedthere
needstobetimelyinvolvementof
socialservicestoensureseamless
transferfromprimarytocommunity
care.
11/100(11%)panelmembers
commented
Severalpanelmemberscommented
thatasocialworkershouldbepartof
theMDT.
Onepersoncommentedwhetherthe
statementshouldreadfrom
secondarytocommunitycarerather
thanfromprimarytocommunity
care.
72.0
76.8
NationalClinicalGuidelineCentre,2013.
84
11/100(11%)panelmembers
commented
Thepanelassumedthatasocial
workerwouldbepartoftheMDT
Somepeoplethoughtthattheterm
appropriateneededtobedefined.
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Anothercommentwasregardingthe
conceptsoftimelyandseamless
whichwerenotdefinedandthe
statementshouldbesetoutto
describeminimumstandards.
Number
Statement
4.
Coordinationbetweenhealthand
77.8
socialcareshouldincludeatimely,
accurateassessment(including
documentationandcommunication)
tofacilitatethetransitionalprocessfor
admission/returntocareornursing
homes.
10/99(10%)panelmembers
commented
Thisshouldalsoincludethe
managementstaffofthecarehome.
Socialworkershouldbepartofthe
MDT.
Therewouldbenoneedforthis
sinceintegratedhealthandsocial
careteamswoulddealwiththis.
Thetermtimelywasquestioned.
5.
Shouldfamilymemberswishto
participateinthecareoftheperson
whohashadastroketheyshouldbe
offeredtraininginassistingtheperson
whohashadastrokeintheiractivities
ofdailylivingpriortodischarge.
18/99(18%)panelmembers
commented
Thereweresomecommentsabout
theneedforconsentfromthe
personwhohashadastroke.
Thedifficultyofarrangingthisprior
todischargewasmentionedand
whetherthiscouldbedoneatthe
personshomewasraised.
Itwasalsostatedthatthereshould
notbeanassumptionthatpeople
arewillingtoprovidehighlevelsof
care.
Respitecareandcarersupport
optionsshouldalsobeidentifiedand
putinplace.
79.8
5.3.2
RecommendationsandlinkstoDelphiconsensussurvey
Statements
4. Whereappropriate,socialworkersshouldbeinvolvedwiththestroke
rehabilitationteamintheassessmentofposthospitalcareneeds.
5. Theroleofsocialcareandanyserviceprovisionrequiredshouldbe
discussedwiththepersonwhohashadastrokeanddocumented
withinthesocialcareplan.
NationalClinicalGuidelineCentre,2013.
85
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
6. Whensocialneedsareidentifiedthereneedstobetimely
involvementofsocialservicestoensureseamlesstransferfrom
primarytocommunitycare.
7. Coordinationbetweenhealthandsocialcareshouldincludeatimely,
accurateassessment(includingdocumentationandcommunication)
tofacilitatethetransitionalprocessforadmission/returntocareor
nursinghomes.
8. Shouldfamilymemberswishtoparticipateinthecareoftheperson
whohashadastroketheyshouldbeofferedtraininginassistingthe
personwhohashadastrokeintheiractivitiesofdailylivingpriorto
discharge.Ifthereisanewidentifiedneedforfurtherstroke
rehabilitationservices,thepersonwhohashadastrokeshouldbe
abletoselfreferwiththesupportofaGPorspecialistcommunity
services.
Recommendations
Forrecommendationsonlongtermhealthandsocialsupportsee15.2.4.
6. Healthandsocialcareprofessionalsshouldworkcollaborativelyto
ensureasocialcareassessmentiscarriedoutpromptly,where
needed,beforethepersonwithstrokeistransferredfromhospital
tothecommunity.Theassessmentshould:
identifyanyongoingneedsofthepersonandtheirfamilyor
carer,forexample,accesstobenefits,careneeds,housing,
communityparticipation,returntowork,transportandaccessto
voluntaryservices.
bedocumentedandallneedsrecordedinthepersonshealthand
socialcareplan,withacopyprovidedtothepersonwithstroke.
7. Offertrainingincare(forexample,inmovingandhandlingand
helpingwithdressing)tofamilymembersorcarerswhoarewilling
andabletobeinvolvedinsupportingthepersonaftertheirstroke.
Reviewfamilymembersandcarerstrainingandsupportneeds
regularly(asaminimumatthepersons6monthandannual
reviews),acknowledgingthattheseneedsmaychangeovertime.
Economicconsiderations Therearesomecostsassociatedwiththesocialcareassessmentandwith
thetrainingforfamilymembers(stafftimecost).TheGDGhasconsidered
theeconomicimplicationsandconcludedthattheseinterventionswill
improvethequalityoflifeofthepersonwithstroke;theimprovementin
qualityoflifewasconsideredlikelytooutweighthecosts.
Otherconsiderations
Asocialcareassessmenttoidentifyneedstosupportthepersonand
carersfollowingdischargeisessential,andthebenefitsofhavingasocial
workeraspartofthemultidisciplinarystrokerehabilitationteamhas
beenacknowledged.Itwasagreedhavingsocialcarefullyintegrated
withintheMDThelpstoensureinformationiscommunicatedand
planningsupportfordischargeisconductedadequately.Itwas
recognisedthatthereisoftenadeficiencyintheprovisionofaco
NationalClinicalGuidelineCentre,2013.
86
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
ordinatedapproachtodeliveryofservicesincurrentpractice.An
assessmentmayberequiredatdifferentpointsofthecarepathwayand
includeothersettingssuchascarehomes.Thediscussionalsohighlighted
aneedformorejoinedupserviceprovisionandneedforspeedy
distributionofinformation/documentationbetweenservices.
Communicationiscurrentlycommonlyslowwhichthenleadstodelays
andintherehabilitationprocess.TheGDGalsoagreedthattheperson
whohashadastrokeandtheirfamily/carerneedtobefullyintegrated
inthisprocessandassuchreceiveacopyofthehealthandsocialcare
plan.
Provisionoftrainingforthecarerwhoiswillingandabletoprovide
supporthasbeenhighlighted.Itwasagreedthattheseneedswouldvary
atdifferentstagesofthepersonsrecovery,andthereforeshouldbe
reviewedatregularintervals.TheGDGagreedthattrainingandsupport
forcarerswasextremelyimportant.
5.4 Transferofcarefromhospitaltocommunity
Rehabilitation can take place in either the hospital or at home. There are potential advantages to
rehabilitationathomeincludinginterventionstargetedmoreaccuratelyatthepatientsneedswithin
their own environment, better patient and carer outcomes in terms of wellbeing and mood, and
greater costeffectiveness. There are also potential disadvantages, for example, delivering high
intensitytherapymaybemoredifficulttoorganiseinacommunitysetting.
5.4.1
Earlysupporteddischarge
Early supported discharge is an approach that promotes discharge from hospital for community
based rehabilitation as soon as possible once appropriate support is in place for both patient and
carer.ItislikelythatsomestrokepatientswillbeunsuitableforthisESDapproachbecauseoftheir
level of physical disability or because of significant prior morbidity. The components of early
supported discharge vary from service to service, the integrated health and social care inputs
offered, and varying skill mix and number. Identifying the clinical and cost effectiveness of ESD is
thuscomplexandmultifaceted.
5.4.2
EvidenceReview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessofearly
supporteddischargeversususualcare?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
Earlysupporteddischargeforstroke
Comparison
Usualcare;strokehospitalunits
Outcomes
BarthelIndex
Lengthofhospitalstay
FunctionalIndependenceMeasure(FIM)
Caregiverstrainindex
Falls
Readmissionstohospital
HospitalAnxietyandDepressionScale(HADS)
NationalClinicalGuidelineCentre,2013.
87
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
ClinicalMethodologicalIntroduction
Mortality
QualityOfLife(anyoutcome)
NottinghamExtendedActivitiesofDailyLiving(NEADL)
5.4.2.1
Clinicalevidencereview
SearcheswereconductedforsystematicreviewsandRCTscomparingtheeffectivenessofearly
supporteddischargeversususualcareforpatientswithstroke.Onlystudieswithaminimumsample
sizeof20participants(10ineacharm)wereselected.Ten(10)RCTswereidentified.Table17
summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
Table17: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOME
Anderson,
200010
Acutestroke
patientsthat
were
medically
stableand
suitabletobe
discharged
earlyfrom
hospitaltoa
community
rehabilitation
schemeand
hadsufficient
physicaland
cognitive
function.
Patients
includedin
thisstudy
weremildly
disabled277
Earlyhospital
dischargeand
individuallytailored
home
based/community
rehabilitation
(medianduration,5
weeks)byafulltime
occupational
therapist,a
consultantin
rehabilitation,
physiotherapists,
occupational
therapists,social
workers,speech
therapists,and
rehabilitationnurses.
Effortsweremadeso
thatdischargefrom
hospitalcouldoccur
within48hoursof
randomisation.
(N=42)
BarthelIndexat
randomisation
[median(IQR)]:85
(8097)
Conventionalcareand
rehabilitationinhospital,
eitheronanacutecare
medicalgeriatricwardorin
amultidisciplinarystroke
rehabilitationunitrunby
specialistsinrehabilitation
orgeriatricmedicine.
(N=44)
BarthelIndexat
randomisation[median
(IQR)]:86(7795)
Askim,
200412
Acutestroke
patientswith
a
Scandinavian
StrokeScale
(SSS)score
greaterthan2
pointsand
lessthan58
points.Iscore
suchasthis
indicatesthat
Extendedservice
consistingofstroke
unittreatment
combinedwitha
homebased
programmeoffollow
upcarecoordinated
byamobilestroke
teamthatoffersearly
supporteddischarge
andworksinclose
cooperationwiththe
Ordinaryservicedefinedas
thestrokeunittreatmentof
choiceaccordingto
evidencebased
recommendations.
(N=31)
Barthelindex,
mean/median:54.0/55.0
BarthelIndex
CaregiverStrain
index
Mortality
Lengthofhospital
stay
NationalClinicalGuidelineCentre,2013.
88
SF36
Mortality
Falls
Barthelindex
Caregiverstrain
index
Readmissionto
hospital
Lengthofhospital
stay
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
STUDY
POPULATION
patientswere
moderately
disabled277
INTERVENTION
COMPARISON
primaryhealthcare
systemduringthe
firstfourweeksafter
discharge.Themobile
teamconsistedofa
nurse,a
physiotherapist,an
occupational
therapistandthe
consultingphysician.
(N=31)
Barthelindex,
mean/median:
57.7/55.0
Bautz
Holtert,
200220
Acutestroke
patients;not
severely
disabledprior
tostroke;had
noother
medical
condition
likelyto
preclude
rehabilitation
andwere
medically
stable.
Patients
includedwere
moderatelyto
mildly
disabled277
Earlysupported
dischargewitha
multidisciplinary
teamforeachstroke
patientwasoffered
andsupportand
supervisionwas
providedfromthe
projectteam
wheneverneeded.
Fourweeksafter
discharge,the
patientsintheESD
groupwereseenat
theoutpatientclinic.
(N=42)
BarthelIndexsum
scoreatday7:
[median(IQR)]:16.5
(1219)
Conventionalproceduresfor Lengthofhospital
dischargeandcontinued
stay
rehabilitation,whichwere
Nottingham
anticipatedtobelesswell
ExtendedActivities
organized.
ofDailyLiving
(N=40)
Mortality
BarthelIndexsumscoreat
day7:[median(IQR)]:14
(1118)
Donnelly,
200469
Acutestroke
patientswith
nopre
existing
physicalor
mental
disabilitythat
wasjudgedto
makefurther
rehabilitation
inappropriate.
Patients
includedwere
moderately
(1014)to
mildly
disabled(15
19)277
Earlierhospital
dischargecombined
withcommunity
based
multidisciplinary
stroketeam
rehabilitation
comprising0.33
coordinator,1
occupational
therapist,1.5
physiotherapists,1
speechandlanguage
therapist,and2
rehabilitation
assistants.On
averagethenumber
ofhomevisitsovera
3monthperiodwas
2.5perweekeach
lasting45minutes.
Usualhospitalrehabilitation
comprisinginpatient
rehabilitationinastroke
unitandfollowup
rehabilitationinaday
hospital
(N=54)
BarthelIndexatbaseline:
mean(SD):13.89(3.93);
Median(range):15(16)
NationalClinicalGuidelineCentre,2013.
89
OUTCOME
BarthelIndex
Nottingham
ActivitiesofDaily
Living
SF36
EuroQoL
CaregiverStrain
index
Lengthofstay
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
STUDY
POPULATION
INTERVENTION
COMPARISON
PatientsintheCST
groupweretobe
dischargedassoonas
theirhomewas
assessed.
(N=59)
BarthelIndexat
baseline:mean(SD):
14.14(3.38);
Median(range):14
(13)
OUTCOME
Fjaeartoft, Acutestroke
200482
patientswith
a
Scandinavian
StrokeScale
(SSS)score
greaterthan2
pointsand
lessthan57
points(i.e.
moderately
disabled).
ExtendedStrokeUnit
Service(ESUS)
definedasstrokeunit
treatmentsimilarto
OSUScombinedwith
servicefromamobile
teamthatoffersearly
supporteddischarge
andcoordinates
furtherrehabilitation
andfollowupinclose
cooperationwiththe
primaryhealthcare
system.Theteam
consistedofanurse,
aphysiotherapist,an
occupational
therapist,anda
physician.
(N=160)
OrdinaryStrokeUnitService
(OSUS)consistingof
treatmentinacombined
acuteandrehabilitation
strokeunitand/orthe
primaryhealthcaresystem.
Alsodefinedasstrokeunit
treatmentaccordingto
evidencebased
recommendations.
(N=160)
Caregiverstrain
index
GlobalNottingham
HealthProfile1&2
Indredavik, Acutestroke
patientswith
2000121
a
Scandinavian
StrokeScale
(SSS)score
greaterthan2
pointsand
lessthan57
points.
Patients
includedwere
moderately
disabled277
ExtendedStrokeUnit
Service(ESUS)
definedasstrokeunit
treatmentsimilarto
OSUScombinedwith
servicefromamobile
teamthatoffersearly
supporteddischarge
andcoordinates
furtherrehabilitation
andfollowupinclose
cooperationwiththe
primaryhealthcare
system.Theteam
consistedofanurse,
aphysiotherapist,an
occupational
therapist,anda
physician.
(N=160)
BarthelIndex,
mean/median:
OrdinaryStrokeUnitService
(OSUS)consistingof
treatmentinacombined
acuteandrehabilitation
strokeunitand/orthe
primaryhealthcaresystem.
Alsodefinedasstrokeunit
treatmentaccordingto
evidencebased
recommendations.
(N=160)
BarthelIndex,
mean/median:58.5/60
BarthelIndex
Mortality
12monthqualityoflifefollowuponIndredavikstudy
NationalClinicalGuidelineCentre,2013.
90
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
STUDY
POPULATION
INTERVENTION
60.4/65.0
COMPARISON
Mayo,
2000170
Acutestroke
patientswith
motordeficits
afterstroke
whohad
caregivers
willingand
ableto
providelivein
careforthe
subjectovera
4weekperiod
after
discharge
fromthe
hospital.
Patients
includedwere
mildly
disabled277
Rehabilitationat
homeafterprompt
dischargefrom
hospitalwiththe
immediateprovision
offollowupservices
byamultidisciplinary
teamoffering
nursing,physical
therapy(PT),
occupationaltherapy
(OT),speechtherapy
(ST),anddietary
consultation.
Durationof
interventionwas4
weeksforall
participants.
(N=58)
BarthelIndex:
8414.4
Usualcarepracticesfor
SF36
dischargeplanningand
BarthelIndex
referralforfollowup
services.Theseincluded
physiotherapy,occupational
therapyandspeechtherapy,
asrequestedbythe
patient'scareproviderand
offeredthroughextended
acutecarehospitalstay;
inpatientoroutpatient
rehabilitation;orhomecare
vialocalcommunityhealth
clinics
(N=56)
BarthelIndex:82.713.9
Rodgers,
1997219
Acutestroke
patientsthat
werenot
severely
handicapped
priortothe
incident
strokewithno
other
condition
likelyto
preclude
rehabilitation.
Patients
includedwere
moderately
disabled277
EarlySupported
Dischargewithhome
carefromtheStroke
DischargeTeam
(communitybased).
Theteamconsisted
ofanoccupational
therapist,
physiotherapist,
speechandlanguage
therapist,social
workerand
occupationaltherapy
technician.The
strokedischarge
rehabilitationservice
wasavailablefive
daysperweekbut
thehomecare
componentofthe
servicewasavailable
24hperdayand
sevendaysperweek
ifrequired.The
strokedischarge
servicewas
withdrawngradually
andacontactname
andnumberwas
providedtopatients
incaseofsubsequent
queriesorproblems
(N=46)
Inpatientandoutpatient
carewasprovidedforthe
controlgroupby
conventionalhospitaland
communityservices.
Dischargeplanningand
servicespostdischargefor
patientsrandomizedto
conventionalcarewere
arrangedandprovided
accordingtotheusual
practiceofeach
participatingwardorunit.
(N=46)
BarthelIndexat7dayspost
stroke:[median(range)]:13
(220)
NationalClinicalGuidelineCentre,2013.
91
OUTCOME
Lengthofhospital
stay
Mortality
Nottingham
ExtendedActivities
ofDailyLiving
Readmissionto
hospital
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
STUDY
POPULATION
INTERVENTION
BarthelIndexat7
dayspoststroke:
[median(range)]:15
(220)
COMPARISON
OUTCOME
Rudd,
1997224
Stroke
patientsable
toperform
functional
independent
transferor
ableto
perform
transferwith
assistance
Earlydischargewitha
plannedcourseof
domiciliary
physiotherapy,
occupationaltherapy,
andspeechtherapy,
withvisitsas
frequentlyas
considered
appropriate
(maximumoneday
visitfromeach
therapist)forupto3
monthsafter
randomization.
(N=167)
Barthelscoreat
randomisation
rangedfrom020
Usualcarewithno
augmentationofsocial
servicesresources.
(N=164)
Barthelscoreat
randomisationrangedfrom
020
BarthelIndex
HospitalAnxiety
andDepression
Scale(HADS),
Caregiverstrain
index
Mortality
vonKoch
2000275
Stroke
patientswith
moderateto
severe
impairment
Earlysupported
Routinerehabilitation.
dischargeand
(N=41)
continued
rehabilitationat
homebyaspecialised
team.The
rehabilitation
programmewas
tailormadeforeach
patient,continuedin
theirhomesfor3to4
months(meanof12
visits(range331)by
ahomerehabilitation
teamtherapist).
(N=42)
NationalClinicalGuidelineCentre,2013.
92
BarthelIndex
Falls
Lengthofhospital
stay
Readmissionto
hospital
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Comparisonofearlysupporteddischargeversususualcare
Table18: Earlysupporteddischargeversususualcareclinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Effect
Author(s)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
SMD/Risk
orP
Ratio(95%CI) value
Design
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Confidence
(ineffect)
BarthelIndex(6weeksfollowup)(Betterindicatedbyhighervalues)
Askim,200412
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(d)
75.2(30.6) 74(31.2)
1.20(14.71
to17.11)
MD1.20
higher
(14.71
lowerto
17.11
higher)
Moderate
Serious
imprecision
(d)
97.1(6.9)
95.1(10.6) 2.0(1.72to
5.72)
MD2.0
higher
(1.72
lowerto
Low
BarthelIndex(12weeksfollowup)(Betterindicatedbyhighervalues)
Mayo,2000170
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
93
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
Inconsistency
Indirectness
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
(m)
96.0(88.3
100)(h)
98.0(85.5
100)(h)
Serious
imprecision
(d)
75(32.9)
Serious
imprecision
Askim:
71.7
Imprecision
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
5.72
higher)
Confidence
(ineffect)
BarthelIndex(26weeksfollowup)(Betterindicatedbyhighervalues)
Anderson,
200010,
RCT
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
0(2.0to
2.0)(h)
(n)
Moderate
(m)
77.7(27.6) 2.70(19.59
to14.19)
MD2.70
lower
(19.59
lowerto
14.19
higher)
Moderate
Askim:
79.0
SMD
0.03
Low
BarthelIndex(26weeksfollowup)(Betterindicatedbyhighervalues)
Askim,200412
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(52weeksfollowup)(Betterindicatedbyhighervalues)
Askim,200412;
RCTs
Donnelly,200469;
Serious
limitations(
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
94
0.03(0.16to
0.22)
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Rudd,1997224
Design
Effect
Limitations
b)
Inconsistency
Indirectness
Imprecision
(d)
Absolute
effect/
Mean
Differenc
e(MD)
Early
or
Supported
Standard
Discharge Usualcare
ised
Mean
Mean
Mean
Mean
(SD)/
difference/
(SD)/
Differenc
Median
Median
e(SMD)
Median
difference/
(range)/fr (range)/
(95%CI)
equency
orP
Frequency SMD/Risk
(%)
(%)
Ratio(95%CI) value
(34.7);
(28.7);
higher
(0.16
Donnelly: Donnelly:
lowerto
17.98
17.15
0.22
(3.1);
(3.81);
higher)
Rudd:16.0 Rudd:16.0
(4.0)
(4.0)
Confidence
(ineffect)
BarthelIndex(26weeksfollowup)(Betterindicatedbyhighervalues)
Indredavik,
2000121
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
(f)
(g)
(g)
1.72(1.10
2.70)(h)
(g)
Moderate
(f)
(f)
(g)
(g)
2.75(0.77
9.77)(h)
(g)
Moderate
(f)
BarthelIndex(52weeksfollowup)(Betterindicatedbyhighervalues)
vonKoch,
2000275
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
95
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
Inconsistency
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
Indirectness
Imprecision
noserious
indirectness
Veryserious
imprecision
(e)
31/84
(36.9%)
32/85
(37.6%)
RR0.96(0.68
to1.35)
(m)
15.0(8.0
22.0)
30.0(17.3
48.5)
13.0(22.0to <0.001(h
6.0)
)
Confidence
(ineffect)
Falls(24and52weeksfollowup)(Betterindicatedbylowervalues)
Anderson,
200010;von,
2000275
RCT
Serious
limitations(c
)
noserious
inconsistency
15fewer Verylow
per1000
(from
120
fewerto
132
more)
Lengthofhospitalstay(6monthsfollowup)(Betterindicatedbylowervalues)
Anderson,
200010
RCT
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
96
Moderate
(m)
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Effect
Author(s)
Design
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
(m)
22(h)
31(h)
(n)
(n)
Moderate
(m)
(o)
6(h)
6(h)
(o)
(o)
Moderate
(o)
Noserious
imprecision
Askim:
23.5
(30.5);
Mayo:9.8
(5.3);
Askim:
30.5
(44.8);
Mayo:
12.4(7.4);
3.34(5.44,
1.24)
MD3.34
lower
(5.44to
1.24
lower)
Moderate
Confidence
(ineffect)
Lengthofhospitalstay(6monthsfollowup)(Betterindicatedbylowervalues)
BautzHoltert,
200220
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Lengthofhospitalstay(6monthsfollowup)(Betterindicatedbylowervalues)
vonKoch
2000275
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Lengthofhospitalstay(52weeksfollowup)(Betterindicatedbylowervalues)
Askim,200412;
Mayo,2000170;
Rudd,1997224
RCT
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
97
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
Inconsistency
Indirectness
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Rudd:12
(19)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Rudd:18
(24)
Veryserious
imprecision
(k)
(l)
(l)
8.00(23.25,
7.25)
MD8
lower
(23.25
lowerto
7.25
higher)
Verylow
Moderate
(m)
Imprecision
Confidence
(ineffect)
Lengthofhospitalstay(52weeksfollowup)(Betterindicatedbylowervalues)
Donnelly,
200469
RCT
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Lengthofhospitalstay(52weeksfollowup)(Betterindicatedbylowervalues)
Rodgers,1997219
RCT
Serious
limitations(
q)
Noserious
inconsistency
Noserious
indirectness
(m)
14(8
31)(h)
23(1158)
(h)
(n)
0.03(h)
Noserious
Noserious
Serious
49/444
65/482
RR0.75(0.53
34fewer Low
Mortality(1252weeksfollowup)
Anderson,
RCT
Serious
NationalClinicalGuidelineCentre,2013.
98
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
10
2000 ,Askim,
200412;
BautzHolter,
200220;
Indredavik,
2000121;Rodgers,
1997219;Rudd,
1997224
Effect
Limitations
limitations(
b)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
(i)
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
(10.1%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
(10.1%)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
to1.05)
per1000
(from63
fewerto
7more)
Confidence
(ineffect)
NottinghamADL(3monthsfollowup)(Betterindicatedbyhighervalues)
BautzHoltert,
200220
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
(m)
34.5(28
44)
30(1446)
(8to7)(h)
0.78(h)
Moderate(
m)
(m)
40(29
45)(h)
37(20
46)(h)
(8to7)(h)
0.93(h)
Moderate(
m)
NottinghamADL(6monthsfollowup)(Betterindicatedbyhighervalues)
BautzHoltert,
200220
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
99
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Effect
Author(s)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
Design
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Confidence
(ineffect)
NottinghamADL(52weeksfollowup)(Betterindicatedbyhighervalues)
Donnelly,200469 RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
12.0(6.34) 10.43
(5.92)
1.57(0.87to
4.01)
MD1.57
higher
(0.87
lowerto
4.01
higher)
Moderate
(m)
10.0(018) 7.0(021)
(n)
(n)
Moderate(
m)
30/128
(23.4%)
RR1.16(0.73
to1.82)
32more Verylow
per1000
(from55
fewerto
NottinghamADL(52weeksfollowup)(Betterindicatedbyhighervalues)
Rodgers,1997219
RCT
Serious
limitations(
q)
Noserious
inconsistency
Noserious
indirectness
Readmissiontohospital(24&52weeksfollowup)(Betterindicatedbylowervalues)
Anderson,
200010;Rodgers,
1997219;von
Koch
RCT
Serious
limitations
(c,q)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(e)
26/128
(17.8%)
NationalClinicalGuidelineCentre,2013.
100
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
2000275
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
167
more)
Confidence
(ineffect)
SF36AndersonPhysicalfunctioning(24weeksfollowup)(Betterindicatedbyhighervalues)
Anderson,
200010
RCT
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
MD1.2
lower
(13.3
lowerto
10.9
higher)
Moderate
SF36AndersonSocialfunctioning(24weeksfollowup)(Betterindicatedbyhighervalues)
Anderson,200010 RCT
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
j)
NationalClinicalGuidelineCentre,2013.
101
Low
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
Confidence
(ineffect)
SF36Physicalhealth(12weeksfollowup)(Betterindicatedbyhighervalues)
Mayo,2000170
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
MD5
higher
(0.82to
9.18
higher)
Low
Noserious
imprecision
MD0.2
lower
(4.73
lowerto
4.33
higher)
Moderate
Noserious
35.59
MD0.92
High
SF36Mentalhealth(12weeksfollowup)(Betterindicatedbyhighervalues)
Mayo,2000170
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
SF36Physicalhealth(52weeksfollowup)(Betterindicatedbyhighervalues)
Donnelly,200469 RCT
Noserious
Noserious
Noserious
34.67
NationalClinicalGuidelineCentre,2013.
102
0.92(11.71
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
limitations
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
(31.32)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
(32.01)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
to13.55)
higher
(11.71
lowerto
13.55
higher)
Confidence
(ineffect)
SF36Mentalhealth(52weeksfollowup)(Betterindicatedbyhighervalues)
Donnelly,2004
69
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
69.49
(18.26)
67.3
(20.07)
2.19(5.48to
9.86)
MD2.19
higher
(5.48
lowerto
9.86
higher)
Noserious
indirectness
Noserious
imprecision
66.36
(18.45)
68.21
(20.31)
1.85(9.60to MD1.85
5.90)
lower
(9.60
lowerto
Moderate
EuroQol(52weeksfollowup)(Betterindicatedbyhighervalues)
Donnelly,2004
69
RCT
Noserious
limitations
Noserious
inconsistency
NationalClinicalGuidelineCentre,2013.
103
High
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
5.90
higher)
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
(l)
(l)
2.70(0.02to
5.38)
MD2.70 Moderate
higher
(0.02
higherto
5.38
higher)
(l)
(l)
4.90(0.46to
10.26)
MD4.90
higher(
0.46
lowerto
10.26
higher)
Confidence
(ineffect)
GlobalNottinghamHealthProfile1(52weeksfollowup)(Betterindicatedbyhighervalues)
Fjaeartoft,
200482
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
GlobalNottinghamHealthProfile2(52weeksfollowup)(Betterindicatedbyhighervalues)
Fjaeartoft,
200482
RCT
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
NationalClinicalGuidelineCentre,2013.
104
Moderate
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Effect
Author(s)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
7/126
(5.6%)
RR2.65(1.16
to6.05)
92more Low
per1000
(from9
moreto
281
more)
Design
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Confidence
(ineffect)
HospitalAnxietyandDepressionScaleAnxiety(52weeksfollowup)(Betterindicatedbylowervalues)
Rudd,1997224
RCT
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(i)
20/136
(14.7%)
HospitalAnxietyandDepressionScaleDepression(52weeksfollowup)(Betterindicatedbylowervalues)
Rudd,1997224
RCT
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(e)
24/136
(17.6%)
21/126
(16.7%)
RR1.06(0.62
to1.8)
10more Verylow
per1000
(from63
fewerto
133
more)
Serious
1.5(2.3)
2.2(2.4)
0.70(1.91,
MD0.7
CaregiverStrainIndex(6weeksfollowup)(Betterindicatedbylowervalues)
Askim,200412
RCT
Noserious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
105
Moderate
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Design
Effect
Limitations
limitations
Inconsistency
inconsistency
Indirectness
indirectness
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
0.51)
lower
(1.91
lowerto
0.51
higher)
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
Serious
imprecision
(j)
Anderson:
0.2(0.4);
Askim:1.0
(1.6)
Anderson:
0.2(0.4);
Askim:1.8
(2.5)
0.03(0.26,
0.20)
MD0.03
lower
(0.26
lowerto
0.20
higher)
Low
Noserious
imprecision
Askim:1.2
(1.9);
Donnelly:
Askim:1.7
(2.7);
Donnelly:
0.13(0.98,
0.72)
MD0.13
lower
(0.98
lowerto
Moderate
Imprecision
imprecision
(j)
Confidence
(ineffect)
CaregiverStrainIndex(24&26weeksfollowup)(Betterindicatedbylowervalues)
Anderson,
200010;Askim,
200412
RCT
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
CaregiverStrainIndex(52weekfollowup)(Betterindicatedbylowervalues)
Askim,200412;
RCT
Donnelly,200469;
Fjaertoft,200482;
Serious
limitations(b
)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
106
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Summaryoffindings
Qualityassessment
Author(s)
Rudd,1997224
(a)
Effect
Design
Limitations
Inconsistency
Indirectness
Imprecision
Early
Supported
Discharge
Mean
(SD)/
Median
(range)/fr
equency
(%)
5.9(2.9);
Fjaertoft:
15.7(2.7);
Rudd:5(4)
Usualcare
Mean
(SD)/
Median
(range)/
Frequency
(%)
6(4.2);
Fjaertoft:
16.4(3.1);
Rudd:4(3)
Blindingnotdoneforoutcomeassessment.
BlindingofoutcomeassessmentnotdoneforRudd,1997.
(c)
Blindingnotdoneforoutcomeassessment(Anderson,2000).
(d)
ConfidenceintervalcrossedoneendofagreedMID.
(e)
.
ConfidenceintervalcrossedbothendsofthedefaultMID
(f)
Imprecisioncouldnotbeassessedbecauseonlyoddsratiowasreported.
(g)
Relativeandabsoluteeffectcouldnotbeassessedbecauseoddsratiowasreported.
(h)
Dataasreportedbytheauthor(s).
(i)
ConfidenceintervalcrossedoneendofthedefaultMID.
(j)
ConfidenceintervalcrossedoneendofdefaultMID.
(k)
ConfidenceintervalcrossedbothendsofdefaultMID.
(l)
Meandifferencereported.GenericInverseVarianceused.
(m)
Imprecisioncouldnotbeassessedbecauseonlymedianandinterquartilevaluesreported.
(n)
Meandifferencecouldnotbeassessedbecausemedianandinterquartilevaluesreported.
(o)
Imprecision/Relativeandabsoluteeffectcouldnotbeassessedbecauseonlythemeannumberofdayswasreported.
(b)
NationalClinicalGuidelineCentre,2013.
107
Absolute
effect/
Mean
Differenc
e(MD)
or
Standard
ised
Mean
Mean
difference/
Differenc
Median
e(SMD)
difference/
(95%CI)
orP
SMD/Risk
Ratio(95%CI) value
0.72
higher)
Confidence
(ineffect)
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.4.2.2
Economicevidence
Eightanalyseswereincludedthatcomparedearlysupporteddischargewithusualcare:onemodelledcostutilityanalysis183andsevencostconsequence
analysesthatreportedananalysisofcostsalongsideclinicaloutcomesfromarandomisedclinicaltrialincludedintheclinicalreview8,22,69,83,172,255,274.These
aresummarisedintheeconomicevidenceprofilebelow(Table19).FurtherdetailsoneachstudyareavailablefromtheevidencetablesinAppendixI.
ThreeidentifiedanalysescomparingESDwithusualcarewereexcludedformethodologicalreasons(Anderson20029,Larsen2006146,Saka2009228).
Table19: Economicevidenceprofile:earlysupporteddischarge(ESD)versususualcare
Study
Applicabilit
y
Anderson
20008
(Australia)
Partially
applicable
(b)(c)(d)
Potentially
serious
limitations
(g)(h)(l)
Costconsequenceanalysis
WithinRCTanalysisRCT
includedinclinicalreview
(Anderson200010)
Followup:6months
Beech
199922
(UK)
Partially
applicable
(b)(d)
Potentially
serious
limitations
(g)(i)(h)(j)
Donnelly
200469
(UK)
Partially
applicable
(e)(d)
Potentially
serious
Limitations
(g)(i)(h)(l)
Limitations Othercomments
Incremental
cost
Incrementaleffect(a)
ICER
Uncertainty
1217(m)
(ESDcost
saving)
FromclinicalreviewAnderson
200010
SF36(MD):physicalfunctioning
1.2(13.3,10.9);social
functioning8,1(19.89,3.69)
Barthel(MD):0(2.0,2.0)
Falls(RR):0.75(0.26,2.17)
Caregiverstrain(MD):0.00
(0.23,0.23)
N/A
IncrementalcostCI:2306
to127(o)
DSA:hospitalbasedcare
becamelesscostlythanESD
whenhospitalcostswere
reducedby50%
Costconsequenceanalysis
WithinRCTanalysisRCT
includedinclinicalreview
(Rudd1997224)
Followup:12months.
632(ESD
costsaving)
FromclinicalreviewRudd
1997224
Mortality(RR):0.75(0.47,1.19)
Barthel(SMD):0.0(0.24,0.24)
HADS(RR):2.65(1.16,6.05)
Caregiverstrain(SMD):1.00(
0.19,2.19)
N/A
IncrementalcostCI:NR;
p=NR
DSA:conclusionsnot
impactedunderplausible
variationsinlengthofstay
andoverheadrates
Costconsequencesanalysis
WithinRCTanalysisRCT
includedinclinicalreview
(Donnelly200469).
Followup:12months
1578(ESD
costsaving)
SeeclinicalreviewDonnelly
200469
Barthel020:0.24(0.16,0.64)
NottinghamADL(MD):1.57
(0.87,4.01)
N/A
IncrementalcostCI:
12,115,4851(o)
NoSA
NationalClinicalGuidelineCentre,2013.
108
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Study
Applicabilit
y
Limitations Othercomments
Incremental
cost
Incrementaleffect(a)
ICER
Uncertainty
SF36(MD):physicalfunctioning
0.92(11.71,13.55);mental
health2.19(5.48,9.86)
EuroQolVAS(e):1.85(9.60,
5.90)
Caregiverstrain(SMD):0.03
(0.52,0.57)
Fjaertoft
200583
(Norway)
Partially
applicable
(b)(c)(d)
Potentially
serious
limitations
(g)(h)(l)
Costconsequenceanalysis
WithinRCTanalysisRCT
includedinclinicalreview
(Indredavik2000121and
Fjaetoft200482)
Followup:12months
1491(m)
(ESDcost
saving)
FromclinicalreviewIndredavik
2000121andFjaetoft200482
Barthel(MD):1.72(1.102.70)
Mortality(RR):0.87(0.43,1.76)
Caregiverstrainindex(SMD):
0.24(0.00,0.49)
N/A
IncrementalcostCI:NR;
p=0.127
Stratificationbyfunctional
impairmentlevel:ESDnot
costsavingintheleast
severegroup(1477,CINR,
p=0.200)
DSA:varyingcostsdidnot
impactconclusions
McNamee
1998172
(UK)
Partially
applicable
(b)(d)
Potentially
serious
limitations
(g)(i)(h)(j)
Costconsequenceanalysis
WithinRCTanalysisRCT
includedinclinicalreview
(Rodgers1997219)
Followup:6months
325(ESD
costsaving)
FromclinicalreviewRodgers
1997219
Mortality(RR):0.25(0.03,2.15)
N/A
IncrementalcostCI:NR;
p=NR
Stratificationbyfunctional
impairment:ESDnotcost
savingintheleastsevere
group(2400,CINR,
p=0.001)
DSA:ESDnotcostsaving
whenthelowerrangeofthe
costofbeddayswasused
(578)
National
Audit
Office
Partially
applicable
(f)
Potentially
serious
limitations
Costutilityanalysis
Discreteeventsimulation
model
804
0.13QALYs
6184
(n)
UncertaintyaroundICERnot
reported
DSA:conclusionsnot
NationalClinicalGuidelineCentre,2013.
109
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Study
2010183
(NAO)
(UK)
Applicabilit
y
Limitations Othercomments
(k)
Timehorizon:10years
Healthstates:severe,
moderateandmild
disabilitydefinedbyBarthel
score
Treatmenteffects
(probabilityofbeingmild,
moderateorsevere)were
determinedat1year(data
fromRuddetal1997224)
Incremental
cost
Incrementaleffect(a)
ICER
Uncertainty
sensitivetodiscountrateor
extentofcoverageofESD
VonKoch
2001274
(Sweden)
Partially
applicable
(b)(c)(d)
Potentially Costconsequencesanalysis
serious
WithinRCTanalysisRCT
limitations
includedinclinicalreview
(g)(i)(h)(j)(l)
(vonKoch2000,2001274,275
Followup:12months
1333(m)
(ESDcost
saving)
FromclinicalreviewvonKoch
2000,2001274,275.
BarthelADL(MD):2.75(0.77,
9.77)
Falls(RR):1.02(0.72,1.43)
N/A
IncrementalcostCI:NR;
p=NR
NoSA
Teng
2003255
(Canada)
Partially
applicable
(b)(c)(d)
Potentially
serious
limitations
(g)(h)(i)
Costconsequenceanalysis
WithinRCTanalysisbased
onRCTincludedinclinical
review(Mayo2000170)
Followup:3months
1695(m)
(ESDcost
saving)
FromclinicalreviewMayo
2000170
SF36(MD):physicalcomponent
5.00(0.82,9.18);mentalhealth
0.20(4.73,4.33)
Barthel0100(MD):2.0(1.72,
5.72)
N/A
IncrementalcostCI:NR;
p=NR
DSA:conclusionsnot
sensitivetovaryingoverhead
rate
CI:confidenceinterval;DSA=deterministicsensitivityanalysis;ICER=incrementalcosteffectivenessratio;MD=meandifference;NR=notreported;RR=relativerisk;SA=sensitivity
analysis;SMD=standardisedmeandifference.
(a) ForwithinRCTcostconsequenceanalysesthehealthoutcomesreportedinclinicalreviewareincludedintableasreportedaspartofclinicalreview.
(b) QALYsnotused.
(c) SomeuncertaintyaboutapplicabilityofnonUKresourceuseandunitcosts.
(d) Someuncertaintyaboutapplicabilityofresourceuseandunitcostsfromover10yearsago.
(e) EuroQolreportedbutunclearifEQ5Dorvisualanaloguescalepartoftoolused.AssumedVASasreportsonscale0100.
(f) DiscountingnotinlinewithNICEmethodologicalguidance.
(g) RCTbasedanalysissofromonestudybydefinitionthereforenotreflectingallevidenceinarea.
NationalClinicalGuidelineCentre,2013.
110
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Someuncertaintyaboutwhethertimehorizonissufficient.
Somelocalcostsused;someuncertaintyastowhetherthesewillreflectnationalcosts.
Doesn'treportifresidentialcarehasbeenconsideredinanalysis.
Unclearhowthehealthoutcomes,healthandsocialcarecostsofeachhealthstateswerecalculated.Notclearwhetherthestudyconsideredthecostsoflongtermcaresuchas
residentialcare(nursinghomesandresidentialhomes).Unitcostsourcesunclear.
(l) Limited/nosensitivityanalysis.
194
(m) ConvertedtoUKpoundsusingrelevantpurchasingpowerparities .
(n) ICERcalculatedbytheNCGChealtheconomistusingtheincrementalcostsof804and0.13QALYs.TheactualNAOstudyreportedanICERof2,881butisunclearhowthisfigurewas
obtained.Theauthorofthereportwascontactedoverthisspecificissuebutnofeedbackwasreceivedatthetimeofwriting.
(o) Totalmeancosts,differenceinmeantotalcostsandconfidenceintervalfordifferencecalculatedbyNCGChealtheconomistbysummingcostcategories.Standarderrorofdifferencewas
calculatedassumingindependenceofcostcategoriesascovariancewasnotavailable;thisisjudgedlikelytounderestimateuncertainty.
(h)
(i)
(j)
(k)
NationalClinicalGuidelineCentre,2013.
111
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.4.2.3
Evidencestatements
Clinicalevidencestatements
Onestudy12comprising62participantsfoundnosignificantdifferenceintheBarthelindexat6and
26followupbetweentheEarlySupportedDischargegroupandtheusualcaregroup(MODERATE
CONFIDENCEINEFFECT).
Onestudy170comprising114participantsfoundnosignificantdifferenceintheBarthelindexat12
weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup(LOW
CONFIDENCEINEFFECT).
Threestudies12,69,224comprising506participantsfoundnosignificantdifferenceintheBarthelindex
at52followupbetweentheEarlySupportedDischargegroupandtheusualcaregroup(LOW
CONFIDENCEINEFFECT).
Twostudies10,275comprising169participantsfoundnosignificantdifferenceinfallsexperiencedin
theEarlySupportedDischargegroupcomparedtotheusualcaregroupat24and52weeksfollowup
(VERYLOWCONFIDENCEINEFFECT).
Threestudies12,170,224comprising507participantsfoundasignificantdifferenceinlengthofhospital
stayat52weeksfollowup(measuredbyinpatientstay)infavouroftheEarlySupportedDischarge
groupcomparedtotheusualcaregroup(MODERATECONFIDENCEINEFFECT).
Onestudy69comprising113participantsfoundnosignificantdifferenceinlengthofhospitalstayat
52weeksfollowupbetweentheEarlySupporteddischargegroupandtheusualcaregroup(VERY
LOWCONFIDENCEINEFFECT)
Sixstudies10,12,20,121,219,224comprising968participantsfoundnosignificantdifferenceinmortality
betweentheEarlySupportedDischargegroupandtheusualcareat12to52weeksfollowup(LOW
CONFIDENCEINEFFECT)
Onestudy69comprising113participantsfoundnosignificantdifferenceintheNottinghamADLat52
weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup(MODERATE
CONFIDENCEINEFFECT).
Threestudies10,219,275comprising356participantsfoundnosignificantdifferenceinreadmissionsto
hospitalat24and52weeksfollowupbetweentheEarlySupportedDischargegroupandtheusual
caregroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy10comprising86participantsfoundnosignificantdifferenceinthephysicalfunctionofthe
SF36at24weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(MODERATECONFIDENCEINEFFECT).
Onestudy10comprising86participantsfoundnosignificantdifferenceinthesocialfunctionoftheSF
36at24weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(LOWCONFIDENCEINEFFECT).
Onestudy170comprising114participantsfoundasignificantdifferenceinthephysicalhealthofthe
SF36at12weeksfollowupinfavouroftheEarlySupportedDischargegroupcomparedtotheusual
caregroup(LOWCONFIDENCEINEFFECT).
Onestudy170comprising114participantsfoundnosignificantdifferenceinthementalhealthofthe
SF36at12weeksbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(MODERATECONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
112
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Onestudy69comprising113participantsfoundnosignificantdifferenceinphysicalhealthoftheSF
36at52weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(HIGHCONFIDENCEINEFFECT).
Onestudy69comprising113participantsfoundnosignificantdifferenceinmentalhealthoftheSF36
at52weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(MODERATECONFIDENCEINEFFECT).
Onestudy69comprising113participantsfoundnosignificantdifferenceintheEuroQolat52follow
upbetweentheEarlySupportedDischargegroupandtheusualcaregroup(HIGHCONFIDENCEIN
EFFECT).
Onestudy82comprising320participantsfoundnosignificantdifferenceintheGlobalNottingham
HealthProfile1at52weeksbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(MODERATECONFIDENCEINEFFECT).
Onestudy82comprising320participantsfoundnosignificantdifferenceintheGlobalNottingham
HealthProfile2at52weeksbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(MODERATECONFIDENCEINEFFECT).
Onestudy224comprising262participantsfoundthatsignificantlylessproportionofpeopleinthe
usualcareexperiencedanxietyat52weeksfollowupcomparedtotheearlysupporteddischarge
group(LOWCONFIDENCEINEFFECT)
Onestudy224comprising262participantsfoundnosignificantdifferenceindepressionat52weeks
followupbetweentheearlysupporteddischargegroupandtheusualcaregroup(VERYLOW
CONFIDENCEINEFFECT)
Onestudy12comprising62participantsfoundnosignificantdifferenceincaregiverstrainat6weeks
followupbetweentheEarlySupportedDischargegroupandtheusualcaregroup(MODERATE
CONFIDENCEINEFFECT).
Twostudies10,12comprising148participantsfoundnosignificantdifferenceincaregiverstrainat24
and26weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(LOWCONFIDENCEINEFFECT).
Fourstudies12,69,82,224comprising826participantsfoundnosignificantdifferenceincaregiverstrain
at52weeksfollowupbetweentheEarlySupportedDischargegroupandtheusualcaregroup
(MODERATECONFIDENCEINEFFECT).
Economicevidencestatements
AUKcostutilitymodelfoundESDtobecosteffectivecomparedtousualcare(directlyapplicable,
potentiallyseriouslimitation)183.
SevenwithinRCTcostconsequenceanalyses(partiallyapplicable,potentiallyseriouslimitations)
foundcostswithESDtobesimilarorlowerthanusualcaretakingintoaccounthospitaland
communitycostswithfollowupover312months8,22,69,172,255,274.Thesestudiesalsogenerallyfound
healthoutcomestobeequivalentorimprovedwithESD.
5.4.3
Recommendationsandlinktoevidence
Recommendations
8. Offerearlysupporteddischargetopeoplewithstrokewhoareable
totransferfrombedtochairindependentlyorwithassistance,as
longasasafeandsecureenvironmentcanbeprovided.
NationalClinicalGuidelineCentre,2013.
113
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
9. Earlysupporteddischargeshouldbepartofaskilledstroke
rehabilitationserviceandshouldconsistofthesameintensityof
therapyandrangeofmultidisciplinaryskillsavailableinhospital.It
shouldnotresultinadelayindeliveryofcare.
10.Hospitalsshouldhavesystemsinplacetoensurethat:
peopleafterstrokeandtheirfamiliesandcarers(as
appropriate)areinvolvedinplanningfortransferofcare,and
carersreceivetrainingincare(forexample,inmovingand
handlingandhelpingwithdressing)
peopleafterstrokeandtheirfamiliesandcarersfeeladequately
informed,preparedandsupported
GPsandotherappropriatepeopleareinformedbeforetransfer
ofcare
anagreedhealthandsocialcareplanisinplace,andtheperson
knowswhomtocontactifdifficultiesarise
appropriateequipment(includingspecialistseatinganda
wheelchairifneeded)isinplaceatthepersonsresidence,
regardlessofsetting.
Relativevaluesofdifferent Thereviewoftheevidenceincludedthefollowingoutcomes:disability,quality
outcomes
oflife,andcarerstrainaswellasfalls,mortalityandlengthofstay.
TherewasconcernthatmeasuresofdisabilityusedintheBarthelindexwere
limitedbytheceilingeffectandthatmeasuresofqualityoflifedidnotcapture
thedomainsimportanttopatientssuchascognitiveandcommunication
difficulties.
DefinitionsoftheBarthelindexclassificationgiveninthesummarytableswere
takenfromthepaperbyDWadeasagreedwiththeGDG277.TheGDGnoted
thatpatientsrecruitedtostudieswereonaverageinthemildtomoderate
rangeoftheBarthelindex(1014moderate,1519mild)10,12,20,69,121,170,219.
TheGDGnotedthattheresultsshowninthemortalityoutcomesweredifficult
tointerpretduetoimprovementsinstrokecareandmortalityoutcomesover
thelastfewyearswhichwouldnotbereflectedinthestudiesincludedinthe
analysisofevidence(studiesrangedfrom1997to2004).
Tradeoffbetweenclinical
benefitsandharms
TheGDGstressedtheimportanceofdevelopingaconsensusonwhatearly
supporteddischargeshouldcompriseof,asthiswasvariableatpresent.The
GDGnotedthatearlysupporteddischargeservicesshouldbeabletooffer
similarintensityandskillmixavailableinhospitalwithoutadelayofdelivery.
TheGDGalsohighlightedthatthisinterventioncouldplaceaburdenonthe
carerandnotedtheimportanceoftheintegrationofhealthandsocialcareto
enableanadequateassessmentincludingequipmentneedsandacareneeds
assessmentundertakenandcareplanagreedforthepatientandtheirfamily.
TheGDGnotedtheimportanceofpatientsandtheirfamilieshavingapointof
contactifneeded.Existingcommunityrehabilitationteamsshouldalsobe
engagedinthisprocessandthepatientsGPkeptfullyinformed.
Economicconsiderations
TheGDGconsideredtheevidencetosuggestthatESDiscosteffective
comparedtousualcare.AllofsevenwithinRCTanalysesfoundthatESDwas
costsavingcomparedtousualcaretakingintoaccounthospitaland
communitycosts(whichoftenincludedsocialcarecosts)uptoayear;they
alsofoundthatitwasatleastaseffective.Amodelledcostutilityanalysis
NationalClinicalGuidelineCentre,2013.
114
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
foundESDtobecosteffectivecomparedtousualcare,withanincremental
costeffectivenessratiowellbelowthethresholdadoptedbyNICE.TheGDG
agreedthat,sincetheclinicalevidencesuggeststhatESDisatleastaseffective
asusualcare,andsincethereisevidencethatitisalsolikelytohavelower
costs,ESDrepresentsacosteffectiveinterventionforstrokepatients.
TheGDGnotedthatthemaincostsavingsofESDarelinkedtoapotentially
shorterlengthofhospitalstayandthatthisalsohasthepotentialtofreeup
acutecarehospitalbedsforotherstrokepatients.TheGDGalsonotedthat
ESDprogrammesarealreadycommonlyimplementedthroughouttheUKNHS
forappropriatepatients.
Qualityofevidence
ThreeofthestudiesshowedthatEarlySupportedDischargereducedlength
ofstayinhospital12,170,224.Theassessmentofconfidenceintheresultsforthis
outcomewasmoderate.Theredidnotappeartobeanysignificantdifference
inoutcomesthatrelatetodisability,qualityoflife,orcarerstrain.Confidence
intheresultsforotheroutcomeswaslimitedduetothestudydesignorto
variationsinhowtheresultswerereported.
TheGDGconsideredthatanecdotallyitwouldbeexpectedthatearly
supporteddischargewouldputagreaterburdenonthecarerbutthiswasnot
showninthestudiesbyAskimandAnderson10,12.TheGDGagreedwiththe
findingsofnodifferencebetweengroupsatoneyear12,69,82,224.
Otherconsiderations
TheGDGnotedthattherewasnoadequatedescriptionofthecompositionof
usualcareorearlysupporteddischargeinthestudiesanalysed.Thereforeit
wasnotpossibletospecifythecomponentsofESDwithinthe
recommendation.Whilstthismethodofdeliverywouldbesuitableformany
patients,theGDGagreedthatitwasnotsuitableforallandforsomepatients
rehabilitationwithinahospitalsettingwouldbemoreappropriate.Thisis
reflectedbythepatientsrecruitedintothetrialswhowerelessseverely
affectedaftertheirstroke.
TheGDGnotedthatoftenpatientsexperiencedistressatthepointof
discharge,feelingservicesaredisjointedandprovisionisinadequate.Inorder
toaddresstheseconcerns,consensusrecommendationsweremadeindicating
theplanning,supplyofequipmentandsupportforthepatientandtheircarers
thatneedtobeprovidedbythemultiagenciesinvolvedinthedeliveryofcare.
ESDteamswithinthestudiesvariedbutincluded:specialistphysiotherapists,
occupationaltherapists,speechandlanguagetherapists,rehabilitationnurses,
consultantsinrehabilitation,dieticiansandsocialworkers.Thegroupnoted
thatthestudiesdidnotincludeclinicalneuropsychologyinputandthismay
reflectpracticeatthetimeofthestudies.Theconsensusofthegroupwasthat
neuropsychologyshouldbeconsideredpartoftherehabilitationteam.
TheGDGthoughtitimportantthatfuturestudiesrecognisecarerandpatient
perspectivesandqualityoflifewereimportantoutcomestobemeasuredfor
bothgroups.
5.4.4
Transferofcarefromhospitaltocommunity
5.4.5
EvidenceReview:Whatplanningandsupportshouldbeundertakenbythe
multidisciplinaryrehabilitationteambeforeapersonwhohadastrokeisdischarged
fromhospitalortransferstoanotherteam/settingtoensureasuccessfultransitionof
care?
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Components
1. Dischargeplanning
2. Emotional/educationalsupport
3. Coordinationandresourcesofotherservices/agencies(suchassocialcare)
NationalClinicalGuidelineCentre,2013.
115
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Outcomes
5.4.6
4.
5.
6.
7.
Patientandcarersatisfaction
Successfuldischarge
Qualityoflife
optimisedstrategiestominimiseimpairmentandmaximiseactivity/participation
Delphistatementswhereconsensuswasachieved
Table20: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
Results
%
Statement
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Eachpatientshouldhavea
documenteddischargereportwhich
hasbeendiscussedwiththeperson
whohashadastrokeandtheircarer/s
priortotransferofcare,including
dischargestoresidentialsettings.
75.5
14/98(14%)panelmembers
commented
Thiswasseenasimportant,butit
wasquestionedwhetherthiswould
bedifferenttotheGPreport,acopy
ofwhichwouldbegiventothe
personwhohashadastroke.
Thisshouldbewritteninan
accessibleway.
Adischargereport(informingongoing
rehabilitationplanning)shouldcontain
informationaboutthefollowing:
Diagnosisandhealthstatus
Mentalcapacity
Functionalabilities
Transfersandmobility
Careneedsforwashing,dressing,
toiletingandfeeding
Psychologicalandemotionalneeds
Medicationneeds
Socialcircumstances
Managementofriskincludingthe
needsofvulnerableadults
Ongoinggoals
Waysofaccessingrehabilitation
services
86.8
69.7
86.8
82.8
82.8
77.7
84.8
76.7
74.7
76.5
74.4
31/99(31%)panelmembers
commented
Afewfurthersuggestionsand
commentsweremade:
Theindividualsnamedpointof
contact.
Jointhealthandsocialcareplan.
StrokeAssociationInformation
Furthercomments:
Thetermsmentalcapacitywas
queriedi.e.capacityforwhat,and
whethercognitivestatusmaybea
betterterm
Itwasfeltnotnecessarytohaveall
theseforallpeople.
Ahomevisit(withthepersonwhohas 69.8
hadastrokepresent)mayberequired
whensimulationofthehome
environmentsetupintheinpatient
settinghasbeeninconclusiveorthere
isanindicationforfurtherassessment.
14/96(14%)panelmembers
commented
Alimitednumberofpanelmembers
providedcommentsforthis
statement:
Onepersonfeltthattherewere
limitsonstafftimeandresources
NationalClinicalGuidelineCentre,2013.
116
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
5.4.7
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Anotherpersonstatedthatthis
dependedonwhetheranearly
supporteddischargeteamwas
available.
Thiscoulddelaydischargefrom
hospitalwasmentioned.
Thetermmaywasqueried.
Number
Statement
4.
Localsystemswithopen
communicationchannelsandtimely
exchangeofinformationshouldbe
establishedtoensurethattheperson
whohashadastrokeisableto
transfertotheirplaceofresidenceina
welltimedmanner.
71.7
10/99(10%)panelmembers
commented
Ofthetenpeoplewhocommented
onthisstatementsevenindicated
thatthephrasingofthestatement
wasconfusingandcontainedjargon.
Oftheotherthree,onecommented
ontheroleofthekeyworker,
anotherpersoncommentedthatthis
shouldminimiseduplicationand
administrationandthethirdperson
statedthatthisshouldbedoneas
soonasitissafetodoso.
5.
Localhealthandsocialcareproviders
shouldhaveestablishedstandard
operatingprocedurestoensureasafe
dischargeprocess.
74.0
11/100(11%)panelmembers
commented
Individualissueswereraisedinthe
comments:
Anychangestoproceduresneedto
becommunicatedintimelyfashion
Takeintoaccountpersonswishes
andbeawareofcarerstressand
vulnerableadultprocedures
Ideallyjointstandardprocedures
Aneedforflexibilityandbroad
guidancethatcanbeeasily
individualised,ratherthan
prescriptiveprocedures.
Delphistatementwhereconsensuswasnotreached
Results
Number
1.
Statement
Anaccessvisit(withouttheperson 36.6
present)canascertainsuitabilityof
accessto,fromandwithinthe
propertyinrespecttotheperson's
functional,cognitivestatusand
managingrisk.
NationalClinicalGuidelineCentre,2013.
117
Amountandcontentofpanel
commentsorthemes
Inround220/98(20%)panel
memberscommented;
15/84(18%)inround3and13/71
(18%)inround4:
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Results
Number
Statement
Amountandcontentofpanel
commentsorthemes
Themajorityofcomments
expressedthatthestatementwas
unspecificanddidnotsay
whetheritshouldbedoneorin
whatcircumstances(Theissueis
whenalways,sometimes,why,
howtodecide.).
Severalpeopleexpressedthe
opinionthatthisstatementwas
tooobvious,sinceitincludedthe
wordmayorlatertheword
can.
2.
Ahomevisitcanascertaina
person'spotentialformanaging
riskandcognitive/functional
impairmentwithinafamiliar
environment.
56.8
Inround211/99(11%)panel
memberscommented;
13/84(15%)inround3and8/70
(11%)inround4:
Themajorityofcomments
expressedthatthestatementwas
unspecificanddidnotsay
whetheritshouldbedoneorin
whatcircumstances.
(guidelinesshouldbegiven
guidanceabouttowhomand
underwhatcircumstancesavisit
eitheraccessorwiththepatient
shouldbedone.
usuallynotrequiredifESDteam
involvedincare.)
Severalpeopleexpressedthe
opinionthatthisstatementwas
tooobvious,sinceitincludedthe
wordmayorlatertheword
can.
3.
Bothaccessandhomevisitsshould 19.4
becoordinatedbyanoccupational
therapistandifthisisnotpossible
theyshouldhaveclinicaloversight
fromanoccupationaltherapist.
NationalClinicalGuidelineCentre,2013.
118
Inround238/95(40%)panel
memberscommented;
34/83(41%)inround3and15/72
(21%)inround4:
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Results
Number
Statement
Amountandcontentofpanel
commentsorthemes
Themainpointofcontentionwas
whetherornotanOTshould
overseethis.
althoughtheOTwouldusually
beinvolved,thisdoesnotneedto
bethecaseanditmaybe
appropriateforanothermember
oftheteamtoco
ordinate/conductthisdepending
onwhatlimitationsthept
presentedwith.
4.
Aspartofrehabilitationcare
52.1
planning,bothaccessandhome
visitscanbeusedseparatelyor
sequentially,toascertainsuitability
forrehabilitation,managementof
riskandmanagementoflifeafter
strokewithinthepersonshome
environment.
Inround29/99(9%)panel
memberscommented;9/83(11%)
inround3and11/71(15%)in
round4:
Itwasfeltthatthisstatement
wasvagueanddidnotdefinethe
circumstancesofwhenandhow
thisshouldhappen.
Therewasalsoacommentthat
thisshouldnotdelaydischarge
andthatthisissomethingthe
communitystroketeamcould
undertake.
5.4.8
RecommendationsandlinkstoDelphiconsensussurvey
Statements
9. Eachpatientshouldhaveadocumenteddischargereportwhichhas
beendiscussedwiththepersonwhohashadastrokeandtheircarer/s
priortotransferofcare,includingdischargestoresidentialsettings.
10.Adischargereport(informingongoingrehabilitationplanning)should
containinformationaboutthefollowing:
Diagnosisandhealthstatus
Mentalcapacity
Functionalabilities
Transfersandmobility
Careneedsforwashing,dressing,toiletingandfeeding
Psychologicalandemotionalneeds
Medicationneeds
NationalClinicalGuidelineCentre,2013.
119
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
Socialcircumstances
Managementofriskincludingtheneedsofvulnerableadults
Ongoinggoals
Waysofaccessingrehabilitationservices
11.Ahomevisit(withthepersonwhohashadastrokepresent)maybe
requiredwhensimulationofthehomeenvironmentsetupinthe
inpatientsettinghasbeeninconclusiveorthereisanindicationfor
furtherassessment.
12.Localsystemswithopencommunicationchannelsandtimely
exchangeofinformationshouldbeestablishedtoensurethatthe
personwhohashadastrokeisabletotransfertotheirplaceof
residenceinawelltimedmanner.
13.Localhealthandsocialcareprovidersshouldhaveestablished
standardoperatingprocedurestoensureasafedischargeprocess.
Recommendations
11.Beforetransferfromhospitaltohomeortoacaresetting,discuss
andagreeahealthandsocialcareplanwiththepersonwithstroke
andtheirfamilyorcarer(asappropriate),andprovidethistoall
relevanthealthandsocialcareproviders.
12.Beforetransferofcarefromhospitaltohomeforpeoplewithstroke:
establishthattheyhaveasafeandenablinghomeenvironment,
forexample,checkthatappropriateequipmentandadaptations
havebeenprovidedandthatcarersaresupportedtofacilitate
independence,and
undertakeahomevisitwiththemunlesstheirabilitiesandneeds
canbeidentifiedinotherways,forexample,bydemonstrating
independenceinallselfcareactivities,includingmeal
preparation,whileintherehabilitationunit.
13.Ontransferofcarefromhospitaltothecommunity,provide
informationtoallrelevanthealthandsocialcareprofessionalsand
thepersonwithstroke.Thisshouldinclude:
asummaryofrehabilitationprogressandcurrentgoals
diagnosisandhealthstatus
functionalabilities(includingcommunicationneeds)
careneeds,includingwashing,dressing,helpwithgoingtothe
toiletandeating
psychological(cognitiveandemotional)needs
medicationneeds(includingthepersonsabilitytomanagetheir
prescribedmedicationsandanysupporttheyneedtodoso)
socialcircumstances,includingcarersneeds
mentalcapacityregardingthetransferdecision
managementofrisk,includingtheneedsofvulnerableadults
plansforfollowup,rehabilitationandaccesstohealthandsocial
careandvoluntarysectorservices.
NationalClinicalGuidelineCentre,2013.
120
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
14.Ensurethatpeoplewithstrokewhoaretransferredfromhospitalto
carehomesreceiveassessmentandtreatmentfromstroke
rehabilitationandsocialcareservicestothesamestandardsasthey
wouldreceiveintheirownhomes.
15.Localhealthandsocialcareprovidersshouldhavestandard
operatingprocedurestoensurethesafetransferandlongtermcare
ofpeopleafterstroke,includingthoseincarehomes.Thisshould
includetimelyexchangeofinformationbetweendifferentproviders
usinglocalprotocols.
16.Aftertransferofcarefromhospital,peoplewithdisabilitiesafter
stroke(includingpeopleincarehomes)shouldbefollowedupwithin
72hoursbythespecialiststrokerehabilitationteamforassessment
ofpatientidentifiedneedsandthedevelopmentofshared
managementplans.
17.Provideadviceonprescribedmedicationsforpeopleafterstrokein
linewithrecommendationsinMedicinesadherence(NICEclinical
guideline76).
Economicconsiderations Noeconomicevidencewasfoundondischargeofpeopleafterstroke.
Therearesomecostsassociatedwiththeassessmentandfollowupvisits
(stafftimeandtravel/transportcost);theGDGhasconsideredthe
economicimplicationsandconcludedthatinsomecircumstancesthe
benefitoftheinterventionislikelytooutweighthecosts.
Otherconsiderations
Boththehealthandsocialcareplanoutliningrequirementsgoing
forwardaswellasasummaryofinformationontheadmissionand
treatmentsgiveninhospitalneedstobeprovidedtoappropriatepeople
(includingtheGP)andthepersonwhohadthestroke.Aspartofthe
dischargedocumentation,asummaryofrehabilitationactivitieswouldbe
includedasusualpractise.Havingalocalprotocoldrawnupbetween
healthandsocialcareproviderstoensureinformationisbeingrelayed
betweenbothagenciespriortodischargeisveryimportantinensuringa
smoothdischargeforthepersonandtheirfamilies.Itwasnotedthat
thereisoftenalackofinformationprovidedtofamilieswhentheperson
isgoingtoresidentialcare.TheGDGnotedthatitwasveryimportant
thatpeoplewhotransferfromhospitaltoacarehomeshouldreceivethe
samelevelofcareandtreatmentasthosewhoareabletoreturnhome.
TheGDGagreedthatthiswasaneglectedareaandfeltaconsensus
recommendationwaswarrantedtoinitiateanimprovementincurrent
practice.
Consensuswasnotreachedregardinghomevisitsthroughthemodified
Delphi.TheGDGrecognisethathomevisitsarenotrequiredinallcases;
howeverthereareacceptedsituationswhereahomevisitisindicated.
Clinicalindicationsforhomevisitsbeingcarriedoutmayincludetheneed
toassesswhetherthepersonisabletomobilisearoundtheirown
environment,andmanagenecessarytransferswithorwithout
equipment.Patientswithcognitiveorperceptualimpairmentsmayneed
NationalClinicalGuidelineCentre,2013.
121
StrokeRehabilitation
Organisinghealthandsocialcareforpeopleneedingrehabilitationafterstroke
tobeassessedintheirownenvironmenttoaiddecisionmakingregarding
whetherthepatientisabletosafelyreturnhome.Homevisitsorward
leavemayberequiredtoaidthepatientsacceptanceandtransitionback
homewithalteredabilitiesfollowingtheirstroke.
Whensufficientinformation(measurements,photographs)canbegained
andmockupcanbeachievedtofullyassessapatientsabilityinthe
hospitalsetting,visitsmaybeunnecessary.Thereshouldbelocally
agreedsituationswherehomevisitswouldandwouldn'tbeconducted,
andinwhatsituationsprofessionsotherthananOccupationaltherapist
couldconductthem.Onesuchexamplemaybeapatientwithminor
equipmentneedswhoismobilisingwithonepersononthewardmaynot
requireahomevisitiftheyarebeingdischargedwithimmediateESD
involvement.
Itiscurrentpracticethattheseareusuallycarriedoutbyoccupational
therapists,butattimesmaybeperformedbyotherappropriate
membersoftheMDT(forexamplephysiotherapist),dependingonthe
reasonforthehomevisit,andwouldbeoverseenbyanoccupational
therapistwithknowledgeofenvironmentalriskassessment,equipment
provisionandadaptation.TheGDGnotedalargetrial(HOVIS)whichis
soontobepublishedonthisarea.
Theneedforafollowuptobeundertakenbythestrokerehabilitation
teamoncethepersonhadtransferredtothecommunitywasviewedto
beimportanttoensuremanagementplanshavebeenfollowedandto
identifyanyfurthersupport.TheGDGnotedthiswasalready
documentedintheStrokeQualityStandardandagreedthisshouldbe
reinforcedbyaconsensusrecommendationfollowingcomments
receivedbystakeholders.
NationalClinicalGuidelineCentre,2013.
122
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
6 Planninganddeliveringstrokerehabilitation
Toensurethesafetyofthepersonwithstrokewhilemaintainingapatientcentredapproach,key
processesneedtobeinplace.Theseprocessesincludeassessmentonadmissiontothe
rehabilitationservice,individualisedgoalsettingandpatientcentredcareplanning.Thischapter
reviewsthoseprocesses.
Asearchforsystematicreviewswascarriedoutforassessmentforrehabilitation,goalsettingand
rehabilitationplanning.Directevidencefromsystematicreviewswasnotidentifiedforassessment
forrehabilitation(6.1)andrecommendationswerethereforedrawnfromthemodifiedDelphi
consensusstatement.Asystematicreviewforgoalsetting(6.2)wasidentifiedandupdated
(Rosewilliam2011221).Notallaspectsofgoalsettingwerecoveredbytheincludedsystematicreview
andthereforeadditionalDelphistatementsweredraftedfrompublishednationalandinternational
guidelinesandrecommendationsweremadebasedonboththereviewandtheDelphiconsensus
statements.Directevidencefromsystematicreviewswasnotidentifiedforrehabilitationplanning
(section6.3)andrecommendationswerethereforedrawnfromthemodifiedDelphiconsensus
statement.
6.1 Screeningandassessment
6.1.1
6.1.2
EvidenceReview:Inplanningrehabilitationforapersonafterstrokewhatassessments
andmonitoringshouldbeundertakentooptimisethebestoutcomes?
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Components
assessment
careplans
monitoring
Outcomes
Patientandcarersatisfaction
optimisedstrategiestominimiseimpairmentandmaximiseactivity/participation
Delphistatementswhereconsensuswasachieved
Table21: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
Results
%
Statement
Afteradmissiontohospitaltheperson
whohashadastrokeshouldhavethe
followingassessedassoonaspossible:
Positioning
Movingandhandling
Swallowing
Transfers
Pressurearearisk
Continence
Communication
82.0
92.0
94.9
79.5
90.0
86.8
NationalClinicalGuidelineCentre,2013.
123
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
34/100(34%)panelmembers
commented:
Anumberofadditional
assessments/measurementswere
suggested(alotofthesearecovered
inothersections):
Activitiesofdailyliving
Mood
Pain
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Number
Results
%
80.0
77.7
Statement
Nutritionalstatus
86.1
81.1
81.1
84.1
75.2
76.2
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Motorcontrol
Cognition
Anumberofpeoplecommentedthat
theterminologysensory
registration[theoneoptionthatdid
notreachconsensus]wasunclear.
1.
Comprehensiveassessmenttakesinto
account:
Previousfunctionalstatus
Impairmentofpsychological
functioning
Impairmentofphysiologicalbody
functionsandstructures
Activitylimitationsduetostroke
Participationrestrictionsinlifeare
stroke
Environmentalfactors(social
physicalandcultural)
2.
Familymembersand/orcarersshould 71.7
beinformedoftheirrightsforacarers
needsassessment.
11/99(11%)panelmembers
commented:
Thiswasgenerallyviewedasan
importantissue.
Extracts:
Thosecarerswhoarepassiveneed
tobeinformedthatthisisavailable
andmanymaybetootimidtoknow
theycanrequestthisassessment.
3.
Theimpactofthestrokeonthe
personsfamily,friendsand/orcarers
shouldbeconsideredandif
appropriatetheycanbereferredfor
support.
78.0
11/100(13%)panelmembers
commented:
Commentsweredivided:
Somethoughtthatthiswas
obvious
Othersthoughtthatinreality
thereisalackofavailable
supportmechanisms.
4.
Peoplewhohavehadastrokeshould
haveafullneurologicalassessment
includingcognition,vision,hearing,
power,sensationandbalance.
69.0
19/84(23%)panelmembers
commented:
Thiswasastatementthatwasadded
inRound3basedoncommentsin
Round2.
Commentstothisstatementwere
moreindividualthaninthemes:
Thephrasefullassessment
NationalClinicalGuidelineCentre,2013.
124
25/100(25%)panelmembers
commented:
Additionalissuestotakeinto
account:
Patientandcarerviews
Motivation
Comorbidities
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Number
Statement
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
wasqueriedbyone(Ifyou
meanthatafullneurological
assessmentincludesascreening
processthatcanleadtoamore
detailedassessmentasneeded
thenIstronglyagree)
Somepeoplewantedadditional
assessments(swallow,
coordination,movement
control,shouldersubluxationfor
instance)
Itwasmentionedthatthis
shouldbedoneaccordingto
needandthatpeopleshouldnot
beoverassessed.
Theneedtohaveaneurologist
doingthiswasquestioned.
5.
Delphipanelmembersagreedwith
screeningforthefollowing:
Mood
Pain
69.8
68.6
Inround2thiswasanopentext
questionand83peopleanswered;in
round3thiswasrephrasedintoa
statementwithmultipleoptions
formatand18/83(22%)commented:
Therewasconfusionaboutsomeof
theoptionsandadditionalscreening
toolsweresuggested:
Dysphagia/Swallowtests
Falls
CarersStrainIndex
6.
Routinecollectionandanalysisofa
rangeofmeasuresshouldinclude:
NationalInstituteofHealthStroke
Scale
BarthelIndex
HospitalAnxietyDepressionScale
(HADS)
74.0(of50)
selectedas
firstoption
46.5(of43)
assecond
option
56.3(of32)
asthird
option
Inround240/87(46%)panel
memberscommented;26/77(34%)
inround3.Thiswasincludedina
differentformatinRound3(toselect
thethreemain).
Thosethatdidnotreachconsensus
were:
ModifiedRankin
BergBalanceScale
EQ5D
GeneralHealthQuestionnaire
(GHQ)
GeriatricDepressionScale
Somepeopledislikedthefactthat
only3optionscouldbeselectedand
statedthatitdependsonthe
individualpatientswhichmeasures
NationalClinicalGuidelineCentre,2013.
125
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Number
6.1.3
Statement
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
wouldbeselected.
Otherspanelmembershighlighted
thatmeasuresdependonthestage
ofrehabilitation(NIHSSisa
reasonablebaselinewhereasthe
Bergismostusefulbeyondtheacute
phase.Italsodependsonwhatsort
ofanalysisyouareexpectingtobe
done.Isthedataforunderstanding
theseverityofstrokeortheoutcome
ofrehab?)
Itwasquestionedwhetherthe
statementreferstooutcomeor
baselinemeasures(Itdepends
whatyouaretryingtoshow?Ifits
outcomesandservicedemands?
Mayberehabilitationcomplexity
scalestoshowthedemandsand
resourcesyouneed.FIMtoshow
functionaloutcomesperhapsinstead
ofBarthel.).
Additionalmeasureswerealso
suggested:
TOM
PHQ
NottinghamExtendedActivities
ofDailyLivingScale
Delphistatementwhereconsensuswasnotreached
Table22: Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
1.
Statement
Thespecificlistofprofessional
screeningtoolstobeincluded:
MontrealCognitiveAssessment
(MOCA)
FrenchayAphasiaScreeningTest
(FAST)
MalnutritionUniversalScreening
Tool(MUST)
TheWaterlowPressurescorerisk
assessmenttool(pressureulcers)
Results
%
Amountandcontentofpanel
commentsorthemes
25.4
22.5
42.6
44.9
Inround248/93(52%)panel
memberscommented;40/72(56%)
inround3theoptionschanged
betweenrounds2and3:
Anumberofadditionalscales/tools
werementioned[someofwhich
werealreadyincludedinother
statements]:
BergBalancescale
ModifiedRivermeadMobility
Index
Mood
NationalClinicalGuidelineCentre,2013.
126
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Number
Results
%
Statement
Amountandcontentofpanel
commentsorthemes
Therapyoutcomemeasure
Screenformalnutrition
Validity,reliabilityandtrainingneed
tobetakenintoconsideration.(You
shouldstateusingarecognised
tool;
Thetoolisnotimportantsolongas
itisavalidatedtool.Thereisnoneed
todirectwhichtoolspeopleshould
use.)
Concernwasraisedaboutpossible
recommendationsbeingtoo
prescriptive(Thesetoolsshould
onlybesuggestedtoolsnot
prescriptiveastheclinicianshouldbe
abletomakethedecisionastothe
mostappropriatetool.
Thetoolisnotimportantaslongas
itisavalidatedtool.Thereisnoneed
todirectwhichtoolspeopleshould
use.)
Whetherthesewerescreeningtools
oroutcomemeasureswasalso
questioned.
2.
6.1.4
Datacollectionshouldbeoverseenby
anationalbody.
62.0
Inround227/97(28%)panel
memberscommented;21/81(26%)
inround3and16/71(23%)inround
4:
Itwashighlightedthatthisisalready
inexistenceinsomeplace(suchas
theRCPaudit,theScottishStroke
CareAuditortheNationalSentinel
StrokeAudit)
RecommendationsandlinkstoDelphiconsensussurvey
Statements
14.Afteradmissiontohospitalthepersonwhohashadastrokeshould
havethefollowingassessedassoonaspossible:
Positioning
Movingandhandling
Swallowing
Transfers
Pressurearearisk
Continence
Communication
Nutritionalstatus
NationalClinicalGuidelineCentre,2013.
127
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
15.Comprehensiveassessmenttakesintoaccount:
Previousfunctionalstatus
Impairmentofpsychologicalfunctioning
Impairmentofphysiologicalbodyfunctionsandstructures
Activitylimitationsduetostroke
Participationrestrictionsinlifearestroke
Environmentalfactors(socialphysicalandcultural)
16.Familymembersand/orcarersshouldbeinformedoftheirrightsfora
carersneedsassessment.
17.Theimpactofthestrokeonthepersonsfamily,friendsand/orcarers
shouldbeconsideredandifappropriatetheycanbereferredfor
support.
18.Peoplewhohavehadastrokeshouldhaveafullneurological
assessmentincludingcognition,vision,hearing,power,sensationand
balance.
19.Delphipanelmembersagreedwithscreeningforthefollowing:
Mood
Pain
20.Routinecollectionandanalysisofarangeofmeasuresshouldinclude:
NationalInstituteofHealthStrokeScale
BarthelIndex
HospitalAnxietyDepressionScale(HADS)
Recommendations
18.Onadmissiontohospital,toensuretheimmediatesafetyand
comfortofthepersonwithstroke,screenthemforthefollowing
and,ifproblemsareidentified,startmanagementassoonas
possible:
orientation
positioning,movingandhandling
swallowing
transfers(forexample,frombedtochair)
pressurearearisk
continence
communication,includingtheabilitytounderstandandfollow
instructionsandtoconveyneedsandwishes
nutritionalstatusandhydration(followtherecommendationsin
Stroke[NICEclinicalguideline68]andNutritionsupportinadults
[NICEclinicalguideline32]).
19.Performafullmedicalassessmentofthepersonwithstroke,
includingcognition(attention,memory,spatialawareness,apraxia,
perception),vision,hearing,tone,strength,sensationandbalance.
20.Acomprehensiveassessmentofapersonwithstrokeshouldtake
intoaccount:
theirpreviousfunctionalabilities
NationalClinicalGuidelineCentre,2013.
128
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
impairmentofpsychologicalfunctioning(cognitive,emotional
andcommunication)
impairmentofbodyfunctions,includingpain
activitylimitationsandparticipationrestrictions
environmentalfactors(social,physicalandcultural).
21.Informationcollectedroutinelyfrompeoplewithstrokeusingvalid,
reliableandresponsivetoolsshouldincludethefollowingon
admissionanddischarge:
NationalInstitutesofHealthStrokeScale
BarthelIndex.
22.Informationcollectedfrompeoplewithstrokeusingvalid,reliable
andresponsivetoolsshouldbefedbacktothemultidisciplinary
teamregularly.
23.Takeintoconsiderationtheimpactofthestrokeonthepersons
family,friendsand/orcarersand,ifappropriate,identifysourcesof
support.
24.Informthefamilymembersandcarersofpeoplewithstrokeabout
theirrighttohaveacarersneedsassessment.
Economic
considerations
Therearesomecostsassociatedwiththescreeningandfurther
assessment;theGDGhasconsideredtheeconomicimplications
andconcludedthattheseinterventionswillimprovethesafetyand
qualityoflifeofthepersonwithstroke;theimprovementinquality
oflifewasconsideredlikelytooutweighthecosts.
Otherconsiderations
TheGDGagreedthatinthiscontextscreeningisabriefevaluation
whichallowsthepatienttobetriagedandimmediatemanagement
tobeputinplacetoensurethepersonssafety.Wherethereis
evidenceoffunctionalimpairments,moredetailedassessmentwill
thenneedtotakeplace.Otherassessmentsshouldbeundertaken
wheretherearespecificneedsofthepatients.Itwasfeltthat
assessingformoodwasimportantandthiswasnotmadeexplicitin
thesurveyandshouldbeaddedintotherecommendation.The
GDGrecognisedthatsignsofimpairmentsinpsychological
functioning(includingmood)mightnotbedirectlyapparenttothe
personwhohashadthestrokeandthecliniciansonadmissionto
hospitalatthetimeofscreening.Thereforeitwasfeltthatthese
processesshouldbecomprehensivelyassessedatalaterstage.It
wasalsoagreedthatinadditiontolimitationsonactivity,an
assessmentofparticipationrestrictionsshouldalsobeundertaken.
Theanxietythatneurologicalassessmentimpliedthataneurologist
wouldhavetoundertaketheassessmentwasrecognisedby
NationalClinicalGuidelineCentre,2013.
129
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
substitutingthewordmedical.TheGDGfeltamedical
assessmentwasanintegralpartofacomprehensiverehabilitation
assessment.
ActivitylimitationsasdefinedbytheICFincludesocialattitudes,
architecturalcharacteristics,legalandsocialstructures,aswellas
climate,andterrain.TheGDGrecognisedthatarangeofadditional
measurestotheBarthel,andNationalInstituteofHealthStrokeScale
maybeused.Suchmeasuresshouldbeusedtocomparecohorts
ofdata,nottomonitorindividualprogressforrehabilitation.
Sincenoneofthespecificscreeningtoolsreachedconsensusthe
GDGwereunabletomakearecommendation.However,basedon
commentsofthenonconsensusstatementstheGDGrecognised
thatifmeasuresweretobecollectedtheyshouldbestandardised
measurementtoolswithpsychometricallyrobustproperties,and
staffshouldbetrainedintheiruseandfindingsshouldbefedback
totheteam.
TheGDGrecognisedthatthereisadistinctionbetweenmeasures
andscreeningtoolsthatshouldnotbeusedasoutcomes.
Opiniononsupportforfamilyandcarerswasdividedinthesurvey,
withsomethinkingthiswouldalwaysbedoneandothersthatin
realitythereisalackoforganisedmechanismstoprovidesupport.
TheGDGnotedthatitwouldbeusualtoreferthepersontotheir
GPifitwasfelttheyneededtobereferredforadditionalsupport.
TheMDTstroketeamwouldprovideinformationonwheresupport
couldbefound.
6.2 Settinggoalsforrehabilitation
6.2.1
EvidenceReview:Doestheapplicationofpatientgoalsettingaspartofplanningstroke
rehabilitationactivitiesleadtoanimprovementinpsychologicalwellbeing,functioning
andactivity?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention
Anypatientgoalsettingapproach
Comparison
Alternativerehabilitationgoalsettingapproaches
Outcomes
Psychologicalmeasuresandhealthrelatedqualityoflife
Physicalfunction
ActivitiesofDailyLiving(ADL)
Thesemayinclude:Barthel,Nottinghamextendedactivitiesofdaily
living,FIM,ratingscales,surveydata(quantitative),themes
identifiedbyqualitativestudies
NationalClinicalGuidelineCentre,2013.
130
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
6.2.1.1
ClinicalEvidenceReview
Asearchwasconductedforsystematicreviewscomparingtheclinicaleffectivenessofanypatient
goalsettingapproachestoalternativerehabilitationgoalsettingapproachestoimprove
psychologicalwellbeing,functionandactivityinadultsandyoungpeople16orolderwhohavehada
stroke.
Onesystematicreview(Rosewilliam2011221)matchingourprotocolwasidentified.Thisreview
includedtwentysevenstudies(eighteenqualitative,eightquantitativeandonemixedmethod
study).Weincludedtwentyonestudiesfromthisreviewmatchingourprotocol.Thesystematic
reviewexploredthenature,extentandeffectsofapplyingpatientcentredgoalsettinginstroke
rehabilitationpractice.
Afurthersystematicsearch(usingthesamesearchtermsasprovidedintheidentifiedsystematic
review)wasconductedforstudiespublishedsinceJune2010whichwasthesearchcutoffdateof
theincludedsystematicreview.Twostudies(Hale2010100;Worrall2011287)(Table23)matchingour
protocolwereidentifiedfromthisupdatesearchandwerealsoincludedforthisreview.
Table23: Overviewofthetwoadditionalstudiesfromthetopupsearchsincethesystematic
reviewsearchcutoffdate.SeeAppendixHforextraction
Studies
100
Hale2010
Worrall2011287
Population/setting
4communitybased
physiotherapistandseven
strokepatients(threemen,
fourwomen)
Aims
Reviewmethods
Toexplorethe
feasibilityand
acceptabilityofusing
*GoalAttainment
Scaling(GAS)inhome
basedstroke
rehabilitation(HBSR)
Qualitativedescriptive
studyinvolvingsemi
structuredindepth
interviews
50participantswith
aphasiapoststroke.All
participantshadtobeable
toparticipateinanin
depthinterviewinEnglish
usingspeech,gesture,
writing,pictures,and/or
drawings.
Todescribethegoals
ofpeoplewithaphasia
andtocodethegoals
accordingtothe
International
Classificationof
Functioning,Disability
andHealth(ICF)(WHO,
2001)
Qualitativedescriptive
studyinvolvingsemi
structured,indepth
interviews
*Astandardisedwayofscoringtheextenttowhichpatientsindividualgoalsisachievedinthecourseofintervention.
Intheincludedsystematicreview(Rosewilliam2011221)thefollowingmethodologywasadopted:
Bothqualitative(Table24)andquantitative(Table25)studydesignswereincludedinthereview
Qualityofincludedstudieswereassessedbyusingqualitycriteriaadaptedfrompublished
literatures256;106;184.Differentsetsofqualitycriteriawereusedforthequalitativeand
quantitativestudies
Studyqualityassessmentwasdoneinitiallybyoneresearcherandcrosscheckedbyoneofthe
twootherauthors
Themesfromallqualitativestudiesmatchingthereviewquestionswerepooled
Findingsweresynthesizedbyaggregatingthethemesfromthequalitativestudiesandrelating
themtofindingsfromquantitativestudies
Datafromthequantitativestudiescouldnotbemetaanalysedduetolackofrandomisedtrials
Effectsizes(forincludedquantitativestudies)werecalculatedwherepossible
NationalClinicalGuidelineCentre,2013.
131
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Forthisreview,wehaveaddedqualityratings(ourconfidenceinthestudies)tothequalitativeand
quantitativestudiesincludedinthesystematicreview.Thequalityratingswerebasedonquality
characteristics(reportedintheincludedsystematicreview)thatwereassessedinthereview.
Studiesfromthesystematicreviewwereexcludediftheyaddressedmixedneurologicalpopulations,
iftheproportionofpatientswithstrokeis<50%orifthenumberofstrokeparticipantsisunclear
Fortheadditionalqualitativestudiesidentifiedinourupdatesearch:
o ThestudyqualitiesofHale2010100andWorrall2011287wereassessedandratedusingthe
qualitycriteriaadaptedfromtheincludedsystematicreview(Table26)
o WemergedfindingsfromthethemesthatHale2010100identified:enthusiasticallycautious,a
toolintheboxofinterventions,timeconsuming,noteasytosetgoals.Findingswithinthese
themesmatchingthequalitativethemesinthesystematicreviewarepresented(inbold)in
oursummaryoffindingstable(Table27)
o ItwasnotpossibletomergefindingsfromWorrall2011287asthisstudywasstrictlyonaphasic
strokepatientsdescribingtheirgoalsandhowthesegoalscanbecoded(byclinicians)
accordingtotheInternationalClassificationofFunctioning,DisabilityandHealth(ICF)(WHO,
2001).Wethereforereportedthisstudyseparately
Table24: Qualitativestudiesintheincludedsystematicreview(Rosewilliam2011221)
Study
Alaszewski20045
Andreassenand
Wyller200511
Bendz200323
Stroke
samples/settings
Strokepatients,
professionalsfrom
strokerehabilitation
services
Strokepatientsnot
specified
Strokeunit
BoutinLesterand
Gibson200228
Strokepatients
Cott200449
Strokepatients,
occupational
therapist,
neurological
rehabilitationunit
Occupational
therapists
Occupational
therapists
Daniels200255
Foye200285
Datacollection
Semistructured
interviews
*Quality
characteristics
assessed
1,2,3,4,5,6,7,8,9,12
,13,14
Confidence(in
study)
Moderate
Semistructured
interviews
Openinterviews
withpatientsand
notesfrom
professionals
Unstructured
interviewsby
phoneandin
person
Focusgroupsused
tocollectdata
1,2,3,4,6,8,9,11,12,
13,14
2,3,4,8,9,11,12,13,
14
Moderate
1,2,3,5,7,8,9
Verylow
1,2,3,4,5,6,8,9,11,1
2,13,14
Moderate
Focusgroupsand
casenotes
Surveystodescribe
ethicallydifficult
situationsinown
words
1,2,3,4,5,6,7,8,9,10
,11,12,13,14
1,2,3,4,6,7,8,9,11,1
2,13,14
High
NationalClinicalGuidelineCentre,2013.
132
Moderate
Moderate
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Stroke
samples/settings
Stroke
physiotherapist
Strokepatients,
carersandspecialist
nurses
Professionalsfrom
strokerehabilitation
services
Strokepatients
Study
HaleandPiggot
2005101
Lawler1999149
Leach2010150
McGrathand
Adams1999171
Parry2004200
Strokeinpatient
rehabilitation
SuddickandDe
souza2006252
Timmermans
2009259
Wressle1999290
Strokeunits
Strokepatients
Strokepatients,
carer,professional
andclinicians
Confidence(in
study)
Moderate
Datacollection
Semistructured
interviews
Semistructured
interviews
*Quality
characteristics
assessed
1,2,3,4,5,6,8,9,10,1
1,12,13,14
1,2,3,4,5,8,9,10,11,
14
Semistructured
interview
1,2,3,4,6,7,8,9,10,1
1,12,13,14
Moderate
Structured
interviews
Videorecordswere
analysedusing
conversational
analysis
Semistructured
interview
Semistructured
interview
Interviewsand
dailyrecords
1,2,4,6
Verylow
1,2,3,4,5,6,8,9,11,1
2,13,14
Moderate
1,2,3,4,5,6,8,9
Low
1,2,3,4,5,6,9,11,12,
13,14
1,2,3,4,6,9
Moderate
Low
Low
*Qualitycharacteristicsassessed:1.Clearaims2.Adequatebackground3.Appropriatemethodology4.Appropriatedesign
5.Appropriaterecruitmentstrategy(sampleandsampling)Appropriatedatacollection6.Reliabilityofdatacollectiontool
7.Validityofdatacollectiontool8.Datacollectionmethodsdescribedadequately9.Dataanalysismethodsdescribed
adequately10.Reflexivity11.Ethicalissues12.Rigorousdataanalysis13.Clearfindings14.Valueofresearch
Table25: Quantitativestudiesintheincludedsystematicreview(Rosewilliam2011221)
Participants
andsample
size
Study
46
Design
Intervention
used(if
present)
*Quality
characteristics
assessed
Confidence(in
study)
Combs2010
caseseries
design
Useof
Canadian
Occupational
Performance
Measure
(COPM)to
exploregoals
1,5,6,9,10,11
Verylow
Gilbertson
200091
138stroke
patients
Singleblind
randomized
controltrial
Clientcentred
occupational
therapy
tailoredto
patientgoals
1,2,3,4,9,10,11,
13
Low
Monaghan
2005175
75stroke
patients
Serial
comparison
design
AStandard
meetingform
BNewform
toenhance
documentation
1,2,3,4,5,6,7,8,
11,12
Moderate
NationalClinicalGuidelineCentre,2013.
133
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Study
Participants
andsample
size
Design
*Quality
Intervention
characteristics
used(if
assessed
present)
ofpatient
needsgoalsand
involvement
CAboveform
andweekly
wardrounds
withpatients,
carersand
doctors
Confidence(in
study)
Phippsand
Richardson
2007205
CVApatients=
117
Retrospective
analysisof
records
Useof
Canadian
Occupational
Performance
Measure
(COPM)to
exploregoals
1,3,4,8,9,10,11,
13
Low
Roberts2005
214
9stroke
patients
preandpost
intervention
design
Useof
Canadian
Occupational
Performance
Measure
(COPM)to
exploregoals
1,2,4,5,6,9,11,1
2,13,14
Low
Timmermans
2009**259
40stroke
patents
Crosssectional
surveyusing
semistructured
interviews
1,2,9,11,13
Verylow
Wressle2002
289
206stroke
patients
Experimental
design
1,3,6,9,11
Verylow
Useof
Canadian
Occupational
Performance
Measure
(COPM)to
exploregoals
*Qualitycharacteristicsassessed1.Clearlyfocussedquestion2.Appropriatedesign3.Appropriatesamplesize4.Lackof
selectionbias5.Lackofperformancebias6.Appropriateintervention7.Lackofobserverbias8.LackofHawthorneeffect
9.Reliabilityofmeasures10.Validityofmeasures11.Appropriatestatistics12.Lackofconfoundingfactors13.Accurate
results
**Timmermans2009:acrosssectionalsurveyusingsemistructuredformatrequiringquantitativeandqualitativedata
(mixedmethodology)
Table26: Additionalqualitativestudiesfromtheupdatesearchsincesearchcutoffdateof
includedsystematicreviewherewithqualitycharacteristicsandratings
Study
Hale2010100
Stroke
samples/settings
4communitybased
physiotherapistand
sevenstrokepatients
Datacollection
semistructuredin
depthinterviews;
detailedclinicalcase
notesandresearcher
NationalClinicalGuidelineCentre,2013.
134
Quality
characteristics
assessed
1,6,7,8,12,13
Confidence(in
study)
Low
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Study
Worrall2011287
Stroke
samples/settings
50participantswith
aphasiapoststroke
Datacollection
fieldnotes
Qualitative
descriptivestudy
involvingsemi
structured,indepth
interviews
Quality
characteristics
assessed
Confidence(in
study)
1,2,3,4,5,6,7,8,9,1
1,12,13,14
High
*Qualitycharacteristicsassessed:1.Clearaims2.Adequatebackground3.Appropriatemethodology4.Appropriatedesign
5.Appropriaterecruitmentstrategy(sampleandsampling)Appropriatedatacollection6.Reliabilityofdatacollectiontool
7.Validityofdatacollectiontool8.Datacollectionmethodsdescribedadequately9.Dataanalysismethodsdescribed
adequately10.Reflexivity11.Ethicalissues12.Rigorousdataanalysis13.Clearfindings14.Valueofresearch
NationalClinicalGuidelineCentre,2013.
135
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Table27: Summaryoffindingsfromthequalitativethemesandquantitativeevidencefromsystematicreview(Rosewilliam2011)221and
additionalqualitativestudy(Hale2010)100fromupdatesearch
QUALITATIVEANDQUANTITATIVEFINDINGS
MATCHINGQUALITATIVEANDQUANTITATIVE
EVIDENCE
Patientsperceivedthatmakingprogresstowardspersonally
Quantitativeevidence:MaitraandErway2006164;
Wressle2002289;Timmermans2009259
Qualitativeevidence:Cott200449;Bendz200323;
Andreassen200511;McGrath1999171;Young2008295;
QUALITATIVETHEMES
Perceptionsofpatientsregardingpersoncenteredness
ingoalsettingandfactorsinfluencingit
meaningfulgoalshadbeengoodfortheirselfimageand
helpedasacopingmechanism171(VERYLOWCONFIDENCEIN
STUDY)
Otherreasonscitedaretogetbacktowork,independence,
nottobeaburdentoothersandtoavoidembarrassmentin
public259(VERYLOWCONFIDENCEINSTUDY)
Patientsperceivedthattheywerenotincontroloftheirgoals
andtheirinvolvementwithgoalsettingwaspassive295
(MODERATECONFIDENCEINSTUDY)
Passivitywasattributedto:
Limitedaccesstoinformation49(MODERATE
CONFIDENCEINSTUDY)
Inabilitytoaccepttheirconditionespeciallyinthe
earlystagesofstroke49(MODERATECONFIDENCEIN
STUDY)
Participationingoalsettingcouldbeimprovedbyprocesses
suchasformaldocumentationofthepatientsviews,
empoweringkeyworkerstobeproactive,respondingflexibly
totheirchangingneedsandtheuseofgradingsystemsto
measuretheirgoalachievement29549(MODERATE
CONFIDENCEINSTUDIES)
Evidencesuggesttheuseofexplicitmethodstoimprove
patientsperceptionofactiveparticipationingoalsetting
practice289(VERYLOWCONFIDENCEINSTUDY)
Professionalsperceptionsconcerningperson
centerednessingoalsetting
Patientssocialandoccupationalneedswerenotexplicitly
incorporatedintothetreatmentgoals,therebyreflectinga
perceptualpracticegap150(MODERATECONFIDENCEIN
STUDY)
Qualitativeevidence:Leach2010150;Daniels200255;
Hale2010100
Patientcenterednessingoalsettingwouldimprovepatients
NationalClinicalGuidelineCentre,2013.
136
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
QUALITATIVEANDQUANTITATIVEFINDINGS
QUALITATIVETHEMES
MATCHINGQUALITATIVEANDQUANTITATIVE
EVIDENCE
motivation,effectiveuseoftimeandcontributetoholistic
planning150(MODERATECONFIDENCEINSTUDY)
Professionalsascribedreasonsthatcouldlimitadoptionofa
patientcentredapproachsuchasconcernsaboutfuturerisks,
socioculturalbarriers,environmentalandresource
implications15055(MODERATETOHIGHCONFIDENCEIN
STUDIES)
*Setgoalsmightbeusedasameansofencouraging,
motivatingandpromptingpatient100(LOWCONFIDENCEIN
STUDIES)
*Ameasurementtool(GAS)wasfoundusefulinguiding
treatmentandassistingtherapiststosetpatientcentred
goals100(LOWCONFIDENCEINSTUDY)
*ProfessionalswereconcernedaboutthereliabilityofGoal
AttainmentScaling(GAS)inthatdifferenttherapistscould
setdifferentindicatorsforthesamepatient100(LOW
CONFIDENCEINSTUDY)
Statusofpatientcenterednessincurrentstroke
rehabilitationgoalsettingpractices
Evidencesuggeststhatcurrentgoalsettingpracticeisnot
largelypatientcentred
STUDIES)
150
Qualitativeevidence:Leach2010150;Hale2010100
(MODERATECONFIDENCEIN
*IndecisionbyprofessionalsabouttheuseofGASintheir
practice100(LOWCONFIDENCEINSTUDY)
Consequencesofdiscrepanciesinperceptionsand
practiceofgoalsettingprocess
Thereviewrevealeddiscrepanciesbetweenpatientand
professionalintheirperceptionsregardinglevelofpatient
involvementinthegoalsettingprocessandalsowithregard
torecoveryandfocusofrehabilitation164(LOWCONFIDENCE
INSTUDY)
Quantitativeevidence:MaitraandErway2006164;
Qualitativeevidence:Leach2010150;BoutinLester2002
28
;Alaszewski20045;Hale2005101
Thesediscrepanciesinperceptionofillnessandrecovery
betweenthepatientandprofessionalleadtoconflictsnotjust
inthegoalsettingprocessbutalsoimpactedonotherrealms
ofrehabilitationsuchasitsdeliveryandthetherapeutic
relationship150281015(VERYLOWtoMODERATECONFIDENCE
INSTUDIES)
NationalClinicalGuidelineCentre,2013.
137
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
QUALITATIVEANDQUANTITATIVEFINDINGS
MATCHINGQUALITATIVEANDQUANTITATIVE
EVIDENCE
Conflictarisingduetoamismatchinvaluesandprioritieswas
Noquantitativeevaluation
Qualitativeevidence:Foye200285
QUALITATIVETHEMES
Ethicalconflict
highlightedasanimportantdilemmaencounteredinpractice
85
(MODERATECONFIDENCEINSTUDY)
Challengestopatientparticipationingoalsetting
Inhibitoryfactorssuchaslimitedtime,presidingprofessional
routinesandthesingleopportunitytomeetclinicianspost
dischargeforsecondaryriskmanagement150252200(LOWto
MODERATECONFIDENCEINSTUDIES)
Quantitativeevidence:Monaghan2005175
Qualitativeevidence:Leach2010150;Suddick2006252;
Parry2004200;Cott200449;Lawler1999149;Hale2010
100
Patientsparticipationingoalsettingwashinderedby
psychosocialfactorssuchinabilitytoaccepttheoccurrenceof
stroke,depression,patientsguardingagainstexposingtheir
incompetence15049200(VERYLOWtoMODERATE
CONFIDENCEINSTUDIES)
Standardgoalsettingmeetingwhichisheldawayfromthe
patientandwithstandarddocumentationisnotconduciveto
patientcentredgoalsetting175(MODERATECONFIDENCEIN
STUDY)
Thefactormentionedbybothprofessionalsandpatientswas
thestrokepathologywithitshighlyunpredictablerecovery
prognosisanditseffects,suchasaphasia150149(LOWand
MODERATECONFIDENCEINSTUDIES)
*Settinggoalsandindicatorscouldbetimeconsuming
especiallywithpatientswithsevereimpairment(for
example,cognitiveimpairment)100(LOWCONFIDENCEIN
STUDY)
Strategiestodeveloppersoncenterednessingoal
settingpractices
Amultidisciplinaryteamapproachinvolvingthepatientalong
withspecialistssuchasspeechpathologistsimproves
discussionanddocumentationofpatientgoals150175
(MODERATECONFIDENCEINSTUDY)
Setpatientcentredgoalsandthentraining,either
Quantitativeevidence:Monaghan2005175;Wressle
2002289;Roberts2005214;Phipps2007205;Combs2010
46
;Gilbertson200091
Qualitativeevidence:Leach2010150;Hale2005101;
Daniels200255;Cott200449;Lawler1999149
conventionalorinnovative,tailoredtothosegoalsledto
shorttermimprovementinactivitiesofdailyliving,better
globaloutcome,bettermotoroutcomesandbetterself
214 205 46 91
perceivedperformanceandsatisfaction (VERY
NationalClinicalGuidelineCentre,2013.
138
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
QUALITATIVEANDQUANTITATIVEFINDINGS
QUALITATIVETHEMES
MATCHINGQUALITATIVEANDQUANTITATIVE
EVIDENCE
LOWtoLOWCONFIDENCEINSTUDIES)
Patientandfamilyeducationregardingthepathology,process
ofrehabilitationandgoalsetting150(MODERATECONFIDENCE
INSTUDY)
Encouragingpatientstoidentifygoalsthatareinlinewith
theirexpectation150(MODERATECONFIDENCEINSTUDY)
Activedecisionmakinginvolvingpatientsneededtobe
pitchedtotheirparticipatingability(gradeddecisionmaking)
49 55
(MODERATEtoHIGHCONFIDENCEINSTUDIES)
Theusestandardmeasurestoidentifyclientcentredgoals
improvedopportunityforpatientparticipationingoalsetting,
theirperceptionregardingparticipationandabilitytorecall
149 289 214 205 46
theirgoals (VERYLOWtoLOWCONFIDENCE
INSTUDIES)
*Findingsfromadditionalqualitativestudy(Hale2010)mergedherewithfindingsfromincludedsystematicreview
NationalClinicalGuidelineCentre,2013.
139
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Additionalqualitativestudyfromupdatesearchsincesearchcutoffdateofincludedsystematic
review
Summaryoffindings:
Worrall2011287(HIGHCONFIDENCEINEVIDENCEfromthisstudy):Describingthegoalsofpeople
withaphasiaandtocodethegoalsaccordingtotheICF
Returntoprestrokelife:
Participantsexpressedtheirdesiretobenormalagainandtoescapetheircurrentsituationand
returnhometothesecurityoftheiroldlife
Communication:
Participantswithaphasiaspokeoftheimportanceofrecoveringtheircommunicativefunction(for
example,communicationforbasicneedsaswellascommunicationtoexpresstheiropinions).
Theydescribedintensefeelingsoffrustration,hopelessness,isolation,anddepressionatnot
beingabletotalk
Manystressedthattheaphasiawasofhigherprioritytothemthantheirphysicalimpairments
Participantsspokeoftheneedforcommunicationrehabilitationtobeconnectedtoreallifeand
abouthowcommunicationgavethemconfidence
Information:
Participantswantedmoreinformationaboutaphasia,stroke,prognosis,andwhattoexpectat
differentstagesofrehabilitation
Havinginformationallowedpeopletostarttakingcontrolandtoparticipateindecisionsabout
theirowntherapyandtheirownrehabilitation
Speechtherapyandotherhealthservices:
Participantswantedspeechtherapythatmettheirneedsatdifferentstagesofrecovery,was
relevanttotheirlife,morefrequentandcontinuedforlonger.
Participantswantedpositiverelationshipsandinteractionswiththeirspeechtherapistsandother
healthserviceproviders
Controlandindependence:
Someexpressedfrustrationatnotbeingapartofthedecisionmakingintheircare,seeking
informationfromsourcesotherthanhealthprofessionals
Dignityandrespect:
Manypeoplereportedafeelingofbeingdisempoweredbytheiraphasia.Theywantedrespect,
statingthattheywerecompetentpeople,despitetheircommunicationdifficulties.
Social,leisure,andwork:
Tobeabletocarryoutsocialactivitiesandtofeelcomfortableinacrowd
Youngerpeoplewithaphasiawereparticularlyawareofthelossofworkandcareerandoften
helddeep,strongdesirestoreturntosomeemployment
6.2.2
Economicevidencesummary
Literaturereview
Norelevanteconomicevaluationswereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
NationalClinicalGuidelineCentre,2013.
140
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Basedonthedetailsoftheclinicalstudies,theresourcesassociatedwiththegoalsetting
interventionareequivalenttoanhourofmultidisciplinaryteamtimefortheinitialgoalsettingand
halfanhourforeachreview.ThesecostsaresummarisedinTable28.
Table28: Interventioncostsgoalsetting
Resources
Frequency
Unitcosts(a)
Costperpatient
Goalsettingwithmulti
disciplinaryteam
1hour
136perhourpsychologist
35perhournurse
45perhourphysiotherapist
45perhouroccupational
therapist
132perhourmedical
consultant
393
136perhourpsychologist
35perhournurse
45perhourphysiotherapist
45perhouroccupational
therapist
132perhourmedical
consultant
197
Reviewofgoalsettingwith 30minutes
multidisciplinaryteam
a)EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcarereportandthe
51
followingAgendaforChangesalarybandspsychologist(band8),physiotherapistandoccupationaltherapist(band6),nurse(band5)
(typicalsalarybandsidentifiedbyclinicalGDGmembers).
6.2.3
Evidencestatements
ClinicalEvidencestatements
Perceptionsofpatientsregardingpersoncenterednessingoalsettingandfactorsinfluencingit
Twostudies171259foundthatpatientsperceivedthatmakingprogresstowardspersonally
meaningfulgoalshadbeengoodfortheirselfimage,gettingbacktowork,independence,avoiding
embarrassmentinpublicandhelpedasacopingmechanism(MODERATECONFIDENCEINSTUDIES)
Onestudy49foundthatpatientsperceivedtheywerenotincontroloftheirgoalsandtheir
involvementwithgoalsettingwaspassive(MODERATECONFIDENCEINSTUDY)
Twostudies29549foundthatparticipationingoalsettingcouldbeimprovedbyprocessessuchas
formaldocumentationofthepatientsviews,empoweringkeyworkerstobeproactive,responding
flexiblytotheirchangingneedsandtheuseofgradingsystemstomeasuretheirgoalachievement
(MODERATECONFIDENCEINSTUDIES)
Professionalsperceptionsconcerningpersoncenterednessingoalsetting
Onestudy150foundthatpatientssocialandoccupationalneedswerenotincorporatedintothe
treatmentgoals,andthatpatientcenterednessingoalsettingwouldimprovepatientsmotivation,
effectiveuseoftimeandcontributetoholisticplanning(MODERATECONFIDENCEINSTUDY)
Twostudies15055highlightedconcernsaboutfuturerisks,socioculturalbarriers,environmental
andresourceimplicationsasreasonsthatcouldlimitadoptionofapatientcentredapproachingoal
setting(MODERATEtoHIGHCONFIDENCEINSTUDIES)
Onestudy100foundthatameasurementtool(GAS)wasfoundusefulinguidingtreatmentand
assistingtherapiststosetpatientcentredgoalsbutconcernswereraisedaboutthereliabilityofthis
tool(LOWCONFIDENCEINSTUDY)
NationalClinicalGuidelineCentre,2013.
141
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Statusofpatientcenterednessincurrentstrokerehabilitationgoalsettingpractices
Onestudy150foundthatcurrentgoalsettingpracticeisnotlargelypatientcentred(MODERATE
CONFIDENCEINSTUDY)
Onestudy100foundthatprofessionals(physiotherapist)wereundecidedabouttheuseofGoal
AttainmentScaling(GAS)intheirpractice(LOWCONFIDENCEINSTUDY)
Consequencesofdiscrepanciesinperceptionsandpracticeofgoalsettingprocess
Fourstudies15028,1015foundthatdiscrepanciesinperceptionofillnessandrecoverybetweenthe
patientandprofessionalleadtoconflictsinthegoalsettingprocesswhichalsoimpactedonother
realmsofrehabilitation(VERYLOWtoMODERATECONFIDENCEINSTUDIES)
Challengestopatientparticipationingoalsetting
Fivestudies149150252200175highlightedfactorsinhibitingpatientsfromparticipatingingoalsettings.
Thesefactorsinclude:limitedtime,presidingprofessionalroutines,goalsettingmeetingwhichis
heldawayfromthepatient,singleopportunitytomeetclinicianspostdischargeforsecondaryrisk
management,strokepathologywithitshighlyunpredictablerecoveryprognosisanditseffectssuch
asaphasiaand(LOWtoMODERATECONFIDENCEINSTUDIES)
Threestudies15049200highlightedpsychosocialfactorsinhibitingpatientsfromparticipatingingoal
settings.Thesefactorsinclude:inabilitytoaccepttheoccurrenceofstroke,depression,patients
guardingagainstexposingtheirincompetence(MODERATECONFIDENCEINSTUDIES)
Strategiestodeveloppersoncenterednessingoalsettingpractices
Twostudies150175highlightedthatamultidisciplinaryteamapproachinvolvingthepatientalong
withspecialistssuchasspeechpathologistsimprovesdiscussionanddocumentationofpatientgoals
(MODERATECONFIDENCEINSTUDY)
Fourstudies2142054691showedthatpatientcentredgoalsledtoshorttermimprovementinactivities
ofdailyliving,betterglobaloutcome,bettermotoroutcomesandbetterselfperceivedperformance
andsatisfaction(VERYLOWtoLOWCONFIDENCEINSTUDIES)
Onestudy150mentionedthatpatientandfamilyshouldbeeducatedwithregardsthepathology,
processofrehabilitation,settinggoalsandpatientsshouldbeencouragedtoidentifygoalsthatarein
linewiththeirexpectation(MODERATECONFIDENCEINSTUDY)
Goalsofpeoplewithaphasiapoststroke
Onestudy287foundthatpeoplewithaphasiapoststrokewantedgreaterautonomydignityand
respect.Theyalsowantedmoreinformationaboutaphasia,stroketoreturntotheirprestrokelife
tocommunicatetheirbasicneedsandtheiropinions(HIGHCONFIDENCEINSTUDY)
6.2.4
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
6.2.5
Recommendationsandlinkstoevidence
25.Ensurethatpeoplewithstrokehavegoalsfortheirrehabilitation
that:
aremeaningfulandrelevanttothem
focusonactivityandparticipation
Recommendations
arechallengingbutachievable
NationalClinicalGuidelineCentre,2013.
142
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
includebothshorttermandlongtermelements.
26.Ensurethatgoalsettingmeetingsduringstrokerehabilitation:
aretimetabledintotheworkingweek
involvethepersonwithstrokeand,whereappropriate,their
familyorcarerinthediscussion.
Relativevaluesofdifferent
outcomes
Theoutcomesofinterestwerepsychologicalmeasuresandhealthrelated
qualityoflife,physicalfunctionandActivitiesofDailyLiving(ADL)
AnyimpactgoalsettinghasonactivityandparticipationisclearlyImportant
butotheroutcomesincludingpatientssenseofself,autonomy,copingand
selfimagewerealsofelttobeimportant.
Tradeoffbetweenclinical
benefitsandharms
TheGDGagreedthatgoalsettingthatwaspatientcentredandinvolved
sharinginformation,andidentifyingpatientsvalues,beliefsandpreferences
waslikelytohavesignificantbenefitstothepatient,beingbothencouraging
andmotivating.Howevergoalsettingthatisdominatedbyprofessionalsmay
bebothtimeconsuming,anddisempowerpatients,focussingonrehabilitation
interventionsthathavelittleapparentrelevance,althoughtheycanassist
therapistsindevelopingatreatmentplan.
Economicconsiderations
Nocosteffectivenessstudieswerefound.Personnelcostfordeliveringagoal
settinginterventionwasestimatedat393fortheinitialinterventionand197
forthereviewofthegoalssetbasedonGDGestimatesoftheresourceuse
involved.TheGDGconsideredthattheadditionalcostswouldpotentiallybe
offsetbythelongtermbenefittopatientsintermsofimprovedqualityoflife.
Qualityofevidence
Thesystematicreview(Rosewilliam,2011)ofbothquantitativeandqualitative
studiesincludedinthereviewexploredthenature,extentandeffectsof
applyingpatientcentredgoalsettinginstrokerehabilitationpractice.Inthe
qualitativestudiesdatahadbeencollectedbyinterviews,focusgroupsand
surveys.Thequantitativestudieshadusedrandomised,crosssectionalsurvey,
retrospectiveanalysisofrecordsandcaseseriesdesigns.
TwootherqualitativestudiesevaluatedtheGoalAttainmentScaling(GAS)in
homebasedstrokerehabilitation(Hale,2010),andgoalsofpeoplewith
aphasiaandhowthesegoalscanbecoded(byclinicians)accordingtothe
InternationalClassificationofFunctioning,DisabilityandHealth(ICF)(Worrall,
2011).Thesewerebothdescriptivestudiesusingsemistructured,indepth
interviews.
Thethemesexploredbythestudiesincludedperceptionsofpatientsregarding
personcenterednessingoalsettingandfactorsinfluencingit,professionals
perceptionsconcerningpersoncenterednessingoalsetting, challengesto
patientparticipationingoalsettingandstrategiestodevelopperson
centerednessingoalsettingpractices.
Thequalityofincludedstudies(Rosewilliam,2011)wereassessedbyusing
qualitycriteriaadaptedfrompublishedliteraturewithdifferentsetsofquality
criteriausedforthequalitativeandquantitativestudies.Themesfromall
qualitativestudiesmatchingthereviewquestionswerepooled.Thefindings
fromallofthestudiesweresynthesisedbyaggregatingthethemesfromthe
qualitativestudiesandrelatingthemtofindingsfromquantitativestudies.The
studyqualitiesofHale2010andWorrall2011wereassessedandratedusing
thequalitycriteriaadaptedfromtheincludedsystematicreviewandwe
mergedfindingsfromthethemesthatHale2010identified.Confidenceinthe
effectsreportedwithinthestudiesrangedfromverylowtohigh.TheGDG
notedthatthemajorityofstudiesweresmallqualitativestudiesfocussingon
patients'perceptions,professionalsperceptions,theneedforpatient
NationalClinicalGuidelineCentre,2013.
143
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
centrednessandhowtodevelopthis.
Otherconsiderations
6.2.6
TheGDGnotedthatthefindingsfromthestudiesofgoalsettinginstrokewere
similartothosereportedingoalsettinginotherdisablingconditions.
Theimportanceofdevelopingstructurestosupportpatientinvolvementin
goalsettingincludingstafftrainingwashighlighted.Goalsettingneedstobe
adaptedaccordingtotheenvironmentandthestageofacceptancewiththe
individual.Thestudieshighlightedthatsettinggoalsattheveryacutestageis
notalwaysappropriate.Afterastroke,thepersonhasanenormous
adjustmenttomakeinacceptingandcomingtotermswithwhathas
happened.TheGDGagreedthatthereweredifferentlevelsofparticipationby
thepatientingoalsetting,andattheacutestagethismaybelimiteduntilthe
personfeelsreadyandmoreconfidentwhentheycanparticipatemore.
Delphistatementswhereconsensuswasachieved
Table29: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Number
Statement
Bothprofessionspecificaswellas
multidisciplinarystroketeams'goals
shouldbepersonfocused.
81.8
17/99(17%)panelmembers
commented
Thiswasseenimportantinthe
processofgoalplanningbysome
panelmembers(Absolutely.We
dontdothisenoughyetandwe
needtogetmuchbetteratthisto
useoutcomemeasuresproperlyand
reallyeffectively.)
Itwasseenasmostimportantthat
goalsshouldbesetbyorset
collaborativelywiththepersonwho
hashadastroke(Goalsneedtobe
genuinelypersongenerated.
Goalsettingshouldbe
collaborative,setwiththepatient,
andmultidisciplinaryratherthanuni
disciplinary
Thereshouldbeonesetofpatient
agreedpatientcentredgoals)
Fourpeopleexpressedtheopinion
thatthiswasnotasensible
statement.
Effortsshouldbemadetoestablish
thewishesandexpectationsofthe
personwhohashadastrokeandtheir
carer/family.
86.9
13/99(13%)panelmembers
commented
NationalClinicalGuidelineCentre,2013.
144
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
6.2.7
Results
%
Number
Statement
Thefollowingcriteriashouldbeused
whensettinggoalswiththeperson
whohashadastroke:
Meaningfulandrelevant
Shouldbefocusedonactivitiesand
participation
Challengingbutachievable
Bothshortandlongtermtargets
MayinvolveoneMDTteammember
ormaybemultidisciplinary
Involvecarer/familywherepossible,
withconsentofpersonwhohashada
stroke
Usedtoguidetherapyandtreatment
92.0
69.7
76.0
70.1
76.0
81.0
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Itwashighlightedthatthese
expectationsneedtoberealistic.
Somepeoplequestionedtheterm
effortsandwhatthiswouldmeanin
realterms.
Onepersonindicatedtheopinion
thatthiswasaredundantstatement.
20/100(20%)panelmembers
commented
Ratherthanthemesindividualissues
werehighlighted:
Thetypeofgoaldependsonthe
stageandsettingofrehabilitation
(Initialgoalsintheacutesetting
maybelessfocussedonactivities
andparticipationasthetreatment
beginstodevelopabasefromwhich
furthergoalsmaybeset,forexample
increasingthelengthoftreatment
thatcanbetolerated.Notall
objectivescanbeidentifiedwithin
recognisedassessmenttoolsinthe
earlystages.)
Somegoalsmightnotbeeasily
measurable(Goalsdonothaveto
bemeasurableasimprovementin
engagementandmotivationcanbea
goalthatwillbedifficultto
quantify.)
Goalsshouldbejargonfree.
Onepersonindicatedtheopinion
thatthiswasaredundantstatement.
Delphistatementswhereconsensuswasnotachieved
Table30: Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
Statement
Goalsshouldhavepredicteddatesfor
completion.
Results
%
Amountandcontentofpanel
commentsorthemes
36.5
Inround224/98(24%)panel
memberscommented;19/85(22%)
inround3:
Themes:
Flexibilitytimingofgoalsshould
notbetoorigidandprescriptive.
NationalClinicalGuidelineCentre,2013.
145
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Results
%
Amountandcontentofpanel
commentsorthemes
Typeofgoalssomegoalsdont
lendthemselvestopredictanend
point
Effectonpatientsfocusondates
andfailurecanleadtodistressand
haveanimpactonconfidenceand
esteem
ProgressionRatherthangivingone
date,regularreviewsleadtoa
feelingofprogress
Number
Statement
Areviewofgoalsofthepersonwho
hashadastrokeshouldbeconducted
betweenthepersonandthe
multidisciplinaryteammember
deliveringtheinterventionatthe
expecteddateofcompletion.
42.4
Inround214/99(14%)panel
memberscommented;13/85(15%)
inround:
Thepanelscommentshavethe
followingthemessomeofthese
aremirroringthoseforexpected
datesofgoals:
Expecteddateitwasqueried
whethertherewouldbeanexpected
date(Idontagreethatgoalsalways
needtohaveanexpecteddateof
completion.)
Regularreviewsgoalsshouldbe
regularlyreviewedasanongoing
process(Butshouldbeconstantly
reviewedthroughouttherapy.).
Flexibilitywhenandhowthe
reviewwouldtakeplaceshouldbe
flexible(Thesepeopleshouldbe
involvedbuttheredoesneedtobe
someflexibility).
TeamorindividualmemberCould
involveanindividualteammember,
butsometimesalsothewholeteam
(Thisshouldbepartoftheweekly
MDTmeetingwhichthepatient
shouldtakepartin.).
Onepersonobjectedtothis
statementsinceitrepresentsand
idealscenarioratherthanwhatcan
beachievedinclinicalpractice(if
youdidallthesethings,youdnever
havetimetodoanyactual
therapy.).
Thereasonsforunattainedgoalsand
goalsthathavebeenreassessedneed
tobedocumented.
56.5
Inround211/99(11%)panel
memberscommented;6/85(7%)in
round3:
Generallythiswasseenaspositive,
butitwasstatedthatthismaybetoo
reflectiveforsomeandthatitneeds
NationalClinicalGuidelineCentre,2013.
146
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
6.2.8
Number
Statement
Patientsshouldhaveawrittencopyof
theirgoals.
Results
%
Amountandcontentofpanel
commentsorthemes
tobenefittheindividualratherthan
beameasureofoutcome.
Itishelpfultoknowwhyagoalis
notbeingmettolearnabout
patternsofrecoveryandwhat
affectsprogress.
52.4
Inround3(thisstatementwasfirst
introducedinround3)17/84(20%)
panelmemberscommented
Therewasafeelingthattheformat
ofthisdocumentationwouldnot
alwaysbeaccessibletotheperson
whohashadastroke(cognitiveor
languageimpairedpersonsfor
instance).
Itmightbehelpfulifthisstatedthat
thesegoalsshouldbeinlanguage
appropriatetothepatient(notMDT
language)andthatwherepossible,
theyshouldreflectthepatientsown
wordsinsettingthegoals.
Forpatientswithmemoryproblems
thisisparticularlyimportantbutalso
writtengoalsaidcommunication
betweenthepatient,teamand
family.
RecommendationsandlinkstoDelphiconsensussurvey
Statements
Recommendations
21.Bothprofessionspecificaswellasmultidisciplinarystroketeams'
goalsshouldbepersonfocused.
22.Effortsshouldbemadetoestablishthewishesandexpectationsof
thepersonwhohashadastrokeandtheircarer/family.
23.Thefollowingcriteriashouldbeusedwhensettinggoalswiththe
personwhohashadastroke:
Meaningfulandrelevant
Shouldbefocusedonactivitiesandparticipation
Challengingbutachievable
Bothshortandlongtermtargets
MayinvolveoneMDTteammemberormaybemultidisciplinary
Involvecarer/familywherepossible,withconsentofperson
whohashadastroke
Usedtoguidetherapyandtreatment
27.Ensurethatduringgoalsettingmeetings,peoplewithstrokeare
providedwith:
NationalClinicalGuidelineCentre,2013.
147
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
anexplanationofthegoalsettingprocess
theinformationtheyneedinaformatthatisaccessibletothem
thesupporttheyneedtomakedecisionsandtakeanactivepart
insettinggoals.
28.Givepeoplecopiesoftheiragreedgoalsforstrokerehabilitation
aftereachgoalsettingmeeting.
29.Reviewpeoplesgoalsatregularintervalsduringtheirstroke
rehabilitation.
Otherconsiderations
TheDelphitechniquewasusedtoelucidatethestrokerehabilitation
community'sviewsofgoalsettingandconsensuswasachievedonthe
importanceofmeaningful,relevantachievablegoalsthatfocussedonactivity
andparticipationandincludedbothshorttermandlongtermtargets.
TheGDGconsideredtheareasthatachievedconsensusthatwouldsupplement
therecommendationsalreadymadebasedontheevidencereviewundertaken.
TheGDGnotedthosestatementsthatdidnotachieveconsensus,andagreed
thesedidnotseemtobeparticularlycontroversial.Itwasagreedthatemphasis
shouldbeplacedonhavinggoalsthataremeaningfulandrelevanttothe
patient.TheGDGagreedthatitwasveryimportantthatpatientsshouldreceive
acopyoftheirgoals,andarguedthatitwasnotpossibletoprovidepatient
centredgoalsiftheydidnothaveacopytheycouldreferto.Thegroupagreed
withmanyofthecommentsfromthesurveythatinformationongoalsshouldbe
inaformataccessibletothepatienttotakeintoaccountcognitiveor
languageimpairmentsAlthoughtherewasnoagreementaboutreviewing
goalsatspecifieddatestheGDGagreedthatareviewshouldbeconductedat
appropriatetimepointstomonitoranddiscussprogressandreassesstheneeds
andwishesofthepatient.
6.3 Planningrehabilitation
6.3.1
Delphistatementswhereconsensuswasachieved
Table31: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
Statement
Results
%
Documentationrelatedto
rehabilitationshouldbe
individualised,andcontainthe
followingminimuminformation:
Basicdemographicsincluding
92.9
93.9
96.9
NationalClinicalGuidelineCentre,2013.
148
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
17/99(17%)panelmembers
commented:
Anumberofadditional
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Number
Results
%
Statement
contactdetailsandnexttokin
Diagnosisandrelevantmedical
information
78.7
Listofcurrentmedications
includingallergies
93.9
Standardisedscreening
assessmentstoincludethose
identifiedinearlierquestions
87.8
79.5
Personfocusedrehabilitationgoals 85.8
Multidisciplinaryprogressnotes
76.5
Keycontactfromthestroke
rehabilitationteamtocoordinate
healthandsocialcareneeds
79.5
Dischargeplanninginformation
Jointhealth/socialcareplansif
developed
Followupappointments
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
documentsweresuggested:
Returntoworkinformationwas
mentionedmostfrequently
Informationonadditional
supportavailableafterdischarge
(forexample,carersupport
organisationsandstrokesupport
groups)
Strokeeducation/lifestyle
information
1.
Inthedevelopmentof
rehabilitationplans,effortsshould
bemadetoencouragetheperson
whohashadastrokeandcarersto
beinvolvedandactively
participate.
86.9
17/99(17%)panelmembers
commented:
Thiswasseenasimportantin
personcentredcare.
Itwasmentionedthatthewishes
ofthepersonwhohashada
strokeshouldbetakeninto
consideration.Somepeoplefind
thisastressfulexperience.
Threepeopleexpressedan
opinionthatthiswasaredundant
statement.
2.
Rehabilitationplansshouldbe
reviewedbythemultidisciplinary
teamatleastonceperweek.
71.4
NationalClinicalGuidelineCentre,2013.
149
Inround241/95(43%)panel
memberscommented;
34/77(44%)inround3
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Number
Statement
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Thephaseofrehabilitationwas
commentedon.Weeklyreviews
earlyonintheacutephase,or
whenthepersonwhohashada
strokeisaninpatient,reducingto
longerintervalsasthe
rehabilitationprogresses.
notsensible.Infirst6weeks
weeklyisneededthereaftertwo
weeklyisreasonableorlonger
inlightofthequickthroughput
ofhospitalstrokepatientsthe
reviewmayneedtobe
undertakentwiceaweek.
Therewasaconcernnottobe
tooprescriptiveabouttiming.
becauseeachpersonwhohas
hadastrokeisdifferent,the
reviewshouldtakeplace
accordingtoneedsofthe
individualandthiswillvary
Typeofplanandtypeofgoal
wasalsoseenasimportant:
Thisdependsonhowyoudefine
rehabilitationplans.Arethey
broad,forexampletogohome
independentlywalkingandself
careandreturningtoworkor
morespecifictothemomentfor
exampletobeabletostandfor5
minutesinastandingframe?
6.3.2
Delphistatementwhereconsensuswasnotreached
Table32:
Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
NationalClinicalGuidelineCentre,2013.
150
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Results
Number
1.
Statement
Whenthereisasignificantchange,
orwhenaplateau/potentialis
reached,orbeforedischarge,a
meetinginvolvingthestroke
rehabilitationteam,withan
invitationtothepersonandtheir
family/carer,shouldbeconducted
todiscussthesepoints.
63.4
Amountandcontentofpanel
commentsorthemes
Inround222/99(22%)panel
memberscommented;
16/85(19%)inround3and11/72
(15%)inround4:
Therewereseveralthemes:
MDTsomemembersofthe
panelthoughtthatthisdoesnot
havetoinvolvethewholeteam
(Themeetingsshouldhappen
butonlyincludetherelevant
staff,notthewholestroke
rehabilitationteam).
Beforedischargethiswasseen
asthemostimportantaspectof
thestatement.
Needforanadditionalmeeting
ifthereareregularreviewsthen
changes/plateaushouldnot
comeasasurprise
Meetingtypethisneedstobe
tailored(formalorinformal)to
theindividualandtheir
carer/family
Statementthestatementitself
wasseenashavingtoomany
differentcomponentstoanswer
withoneresponse.
Severalpeoplecommentedthat
thetermsplateauorpotential
wasunclear.(Whatisplateau?
Onedayofnochange,oneweek,
onemonth?)
6.3.3
RecommendationsandlinkstoDelphiconsensussurvey
Statements
24.Documentationrelatedtorehabilitationshouldbeindividualised,and
containthefollowingminimuminformation:
Basicdemographicsincludingcontactdetailsandnexttokin
Diagnosisandrelevantmedicalinformation
NationalClinicalGuidelineCentre,2013.
151
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Listofcurrentmedicationsincludingallergies
Standardisedscreeningassessmentstoincludethoseidentifiedin
earlierquestions
Personfocusedrehabilitationgoals
Multidisciplinaryprogressnotes
Keycontactfromthestrokerehabilitationteamtocoordinate
healthandsocialcareneeds
Dischargeplanninginformation
Jointhealth/socialcareplansifdeveloped
Followupappointments
25.Inthedevelopmentofrehabilitationplans,effortsshouldbemadeto
encouragethepersonwhohashadastrokeandcarerstobeinvolved
andactivelyparticipate.
26.Rehabilitationplansshouldbereviewedbythemultidisciplinaryteam
atleastonceperweek.
Recommendations
30.Provideinformationandsupporttoenablethepersonwithstroke
andtheirfamilyorcarer(asappropriate)toactivelyparticipatein
thedevelopmentoftheirstrokerehabilitationplan.
31.Strokerehabilitationplansshouldbereviewedregularlybythe
multidisciplinaryteam.Timethesereviewsaccordingtothestageof
rehabilitationandthepersonsneeds.
32.Documentationaboutthepersonsstrokerehabilitationshouldbe
individualised,andshouldincludethefollowinginformationasa
minimum:
basicdemographics,includingcontactdetailsandnextofkin
diagnosisandrelevantmedicalinformation
listofcurrentmedications,includingallergies
standardisedscreeningassessments(seerecommendation18)
thepersonsrehabilitationgoals
multidisciplinaryprogressnotes
akeycontactfromthestrokerehabilitationteam(includingtheir
contactdetails)tocoordinatethepersonshealthandsocialcare
needs
dischargeplanninginformation(includingaccommodationneeds,
aidsandadaptations)
jointhealthandsocialcareplans,ifdeveloped
followupappointments.
NationalClinicalGuidelineCentre,2013.
152
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Economicconsiderations Therearesomecostsassociatedwiththereviewingoftherehabilitation
planbythemultidisciplinaryteam.TheGDGhasconsideredthe
economicimplicationsandconcludedthatthebenefitsofthe
interventionintermsofimprovementinqualityoflifewereconsidered
likelytooutweighthecosts.
Otherconsiderations
TheGDGoverallagreedwiththestatementonwhatinformationshould
beincludedinplanningrehabilitation,whilstacknowledgingthiswasnot
exhaustiveandshouldbethoughtofasacorelist.Itwasfeltthatthere
wouldbeavarietyofopinionsonadditionalinformationthatshouldbe
included,butwereinagreementwiththeconsensusview.
Itwasthoughtthatcareplanningisanelementofgoalsetting.Although
somecommentshadbeenmadethatthestatementswererather
obvious,theGDGthoughtthatprovidingsupporttoenabletheperson
andcarerstobeinvolvedinthedevelopmentoftheirrehabilitationplans
throughhavingknowledgeandfeelingempoweredtoparticipatewasa
keyrecommendationtomake.
TheGDGthoughtthatspecifyingwhenrehabilitationplansshouldbe
reviewedwasnothelpful,andagreedwiththecommentsfromthe
Delphisurvey,thatthiswouldbevariable,withreviewsbeingcarried
outveryfrequentlyintheearlystagesandlesssolateron.Thegroup
agreedthatitshouldbebasedontheneedsofthepatientatdifferent
stagesoftherehabilitationpathway.
6.4 Intensityofstrokerehabilitation
Thedoseofrehabilitationthatindividualsreceivevariesfromcountrytocountryandserviceto
service.Inspecialistneurorehabilitationservicespatientsmayreceive5hoursoftherapyeachday,
inothers1or2hourseachday.Durationoftherapymayvaryfrom2weeksto3or6monthswith
somepatientsaccessingorreaccessinginputsomeyearsaftertheonsetofstroke.
TheNationalStrokeStrategy61statesPeoplewhohavehadstrokesaccesshighqualityrehabilitation
and,withtheircarer,receivesupportfromstrokeskilledservicesassoonaspossibleaftertheyhave
astroke,availableinhospital,immediatelyaftertransferfromhospitalandforaslongastheyneed
it.TheNICEstrokequalitystandard189specifiesthatPatientswithstrokeareofferedaminimumof
45minutesofeachactivetherapythatisrequired,foraminimumof5daysaweek,atalevelthat
enablesthepatienttomeettheirrehabilitationgoalsforaslongastheyarecontinuingtobenefit
fromthetherapyandareabletotolerateit.Manyfrailolderpatientswithcomorbiditiescannot
toleratesuchintensityintheearlystagesafterstroke,otherpatientscantoleratefarmore.Inother
sphereswheremotorlearningisimportantitisacceptedthatthedegreeofperformance
improvementisdependentontheamountofpractice.Instrokewherethereisarangeof
impairmentsandaspatientsmovearoundinchangingenvironmentsthereisuncertaintyaboutthe
benefitsofincreasingthetotaldoseoftherapywhetherintermsofintensity(hoursperday)or
durationoftherapy(weeks).
6.4.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
intensiverehabilitationversusstandardrehabilitation?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention:
Intensiverehabilitation(inpatientandoutpatient)mixedpackage
oftherapydeliveredbyaMDT.
NationalClinicalGuidelineCentre,2013.
153
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
6.4.1.1
ClinicalMethodologicalIntroduction
(hoursperday,numberofdaysoftreatment,weeksversus
months,largeversussmalldose)
Comparison:
Standardrehabilitationornone
Outcomes:
Lengthofstay
FunctionalIndependenceMeasure(FIM)
BarthelIndex
QualityofLife(anymeasure)
NottinghamActivitiesofDailyLiving
Rankin
RivermeadMobilityIndex
FrenchayActivitiesIndex
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingtheeffectivenessofintense
rehabilitationwithusualcareforrehabilitationafterstrokeforadultsandyoungpeople16orolder
thathavehadastroke.Onlystudieswithaminimumsamplesizeof20participants(10ineacharm)
andincludingatleast50%ofparticipantswithstrokewereselected.Four(4)RCTswereidentified.
Table33summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
Table33:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
zdemir,
2001196
Patientsaged>80
yearswhohad
strokeorrecurrent
strokeandhadbeen
referredafter
medical
stabilisation.
Followup:60days
Therapeuticand
neuromuscular
exerciseswith
occupational
therapywith
professional
supervisionfor2
hoursaday,5days
aweek(intense
multidisciplinary
inpatient
rehabilitation
service).(N=30)
Conventional
Functional
exerciseswith
Independence
familycaregiverand
Measure(FIM)
limitedprofessional
supervisiongivenat
homefor2hours
onceaweek.(N=30)
Ryan,
2006225
Patientsaged>=65
yearsrecently
dischargedfrom
hospitalafter
sufferingastrokeor
hipfracture(only
thesubgroupresults
ofpeoplewith
strokewereused
includedinthe
reviewhere)
Followup:3months
Domiciliary
intensive
rehabilitation:sixor
morefacetoface
contactsperweek
frommembersofa
multidisciplinary
rehabilitationteam.
Maximumlengthof
treatmentlastedfor
12weeks.(N=45)
Standard
rehabilitation:three
orlessfacetoface
contactsperweek
frommembersofa
multidisciplinary
rehabilitationteam.
(N=44)
Smith,1981241
Patientsadmittedto Intensive
hospital,witha
rehabilitation:
recentconfirmed
physiotherapyand
NationalClinicalGuidelineCentre,2013.
154
Standard
rehabilitation:
physiotherapyand
OUTCOMES
Barthel
Index
Frenchay
Activities
Index(FAI)
EuroQol5D
(EQ5D)
Euroqol
Visual
Analogue
Scale(EQ
VAS)
Activitiesof
DailyLiving
(ADL)
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
STUDY
POPULATION
stroke,whowere
abletomanagethe
mostintensiveof
the3regimens.
Followup:12
months
INTERVENTION
occupational
therapyingroups
andindividuallyfor
fourfulldaysa
weekuptosix
months(exceptfor
fourpatientswho
madeafullrecovery
earlier)(timespent
intherapywas
recorded).(N=46)
COMPARISON
OUTCOMES
occupational
therapyingroups
andindividuallyfor
threehalfdaysa
weekuptosix
months(exceptfor
fivepatientswho
madeafullrecovery
earlier)(timespent
intherapywas
recorded).(N=43)
'Noroutine'
rehabilitation:
regularhomevisits
byahealthvisitor,
(onaverageof
sevenvisits(range
313)toeach
patient).Thesevisits
usuallylastedoneto
twohoursduring
thesixmonthsafter
dischargefrom
hospital.(N=44)
Werner,
1996282
Patientswhowere
atleast1yearpost
stroke,with
evidenceof
functional
limitationsinthe
areaofdressing,
walking,eating,or
bathing.
Followup:9months
Intensive12week
outpatient
rehabilitation
programconsisting
ofanhoureachof
physicaland
occupational
therapy,fourtimes
perweek,for12
weeks;therapy
focusedon
neuromuscular
facilitationand
functionaltasks.
(N=33)
Norehabilitation.
(N=16)
NationalClinicalGuidelineCentre,2013.
155
Functional
Independence
Measure;
motormeasure
(FIMMM)
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Comparison:Intensiverehabilitationversusstandardrehabilitationornone
Table34:IntensiverehabilitationversusstandardrehabilitationClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
No.of
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Mean
Differenc
e(MD)
(95%CI)
Intensive
rehabilitation
Mean(SD)
Standard
rehabilitation
Mean(SD)
Mean
difference
(95%CI)
Noserious
imprecision
2.75(2.1)
2.65(2.1)
0.10
(0.77,
0.97)
MD0.1
higher
(0.77
lowerto
0.97
higher)
Moderate
Noserious
imprecision
0.09(0.2)
0.01(0.1)
0.08
(0.01,
0.15)
MD0.08
higher
(0.01to
0.15
higher)
Moderate
Noserious
imprecision
0.14(0.25)
0.0(0.25)
0.14
(0.04,
0.24)
MD0.14
higher
(0.04to
0.24
higher)
Moderate
8.87(7)
8.08(7.7)
0.79(
MD0.79
Low
Imprecision
Confidence
(ineffect)
Barthelindex(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Ryanet
al225
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
EuroqolVAS(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Ryanet
al225
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Euroqol5D(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Ryanet
al225
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Frenchayactivitiesindex(3monthsfollowup)(Betterindicatedbyhighervalues)
1
RCT
Serious
Noserious
Noserious
Serious
NationalClinicalGuidelineCentre,2013.
156
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Summaryoffindings
Qualityassessment
No.of
studies
Ryanet
al225
Design
single
blinded
Effect
Limitations
limitations
(a)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
(c)
Intensive
rehabilitation
Mean(SD)
Standard
rehabilitation
Mean(SD)
Mean
Mean
Differenc
difference e(MD)
(95%CI)
(95%CI)
2.27,
higher
3.85)
(2.27
lowerto
3.85
higher)
Confidence
(ineffect)
FunctionalIndependenceMeasure(totalscore)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Ozdemiret unblinde
d
al.196
Veryserious
limitations
(d)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
59.63(14.19)
12.3(13.38)
47.33
(40.35,
54.31)
MD47.33
higher
(40.35
lowerto
54.31
higher)
Low
3.54
2.87
(h)
(h)
P<0.01(i)
Low(g)
3.50
2.89
(h)
(h)
Low(g)
ActivitiesofDailyLivingindex(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Smithetal
241
RCT
unblinde
d
Veryserious
limitations
(e)
Noserious
inconsistency
Noserious
indirectness
(g)
ActivitiesofDailyLivingindex(12monthsfollowup)(Betterindicatedbyhighervalues)
1
Smithetal
241
RCT
unblinde
d
Veryserious
limitations
(e,f)
Noserious
inconsistency
Noserious
indirectness
(g)
(a)Unclearrandomization.Thestudydidnotachievetheprespecifiedratioof2:1(intensive/nonintensive)25%strokepatientlosstofollowup.
(b)MeandifferencedidnotreachtheagreedMIDof1.85points.
(c)ConfidenceintervalcrossedbothendsofdefaultMID.
(d)Unblindedwithinadequaterandomisationandunclearallocationconcealment.
(e)Unblindedwithnodetailsonrandomisationprocessandallocationconcealment.
(f)20%patientsdroppedoutat1year.
(g)Imprecisioncouldnotbeassessedbecauseonlymeansofdatawerereported.
NationalClinicalGuidelineCentre,2013.
157
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
(h)Relative/absoluteeffectcouldnotbeestimatedasnostandarddeviationwasprovidedinthestudy.
(i)Pvalueasreportedbytheauthors.
Table35: IntensiverehabilitationversusnorehabilitationClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
No.of
studies
Design
Effect
Limitations
Inconsistency
No
rehabilitation
Mean(SD)
Mean
difference
(95%CI)
Indirectness
Imprecision
Intensive
rehabilitation
Mean(SD)
Noserious
indirectness
(d)
3.54
1.50
(e)
(e)
P<0.01
(f)
Low(d)
Noserious
indirectness
(d)
3.50
0.60
(e)
(e)
P<0.05
(f)
Low(d)
(d)
6.6
1.5
(e)
(e)
Low(d)
(d)
0.7
1.0
(e)
(e)
P=0.03
(f)
Low(d)
Pvalue
Confidence(in
effect)
ActivitiesofDailyLivingindex(3monthsfollowup)
1
RCT
Smithetal unblinded
241
Very
serious
limitations
(a)
Noserious
inconsistency
ActivitiesofDailyLivingindex(1yearfollowup)
1
RCT
Smithetal unblinded
241
Very
serious
limitations
(a,b)
Noserious
inconsistency
FunctionalIndependenceMeasure(Motor)(3monthsfollowup)
1
Werneret
al282
RCT
unblinded
Very
serious
limitations
(c)
Noserious
inconsistency
Noserious
indirectness
FunctionalIndependenceMeasure(Motor)(3to9monthsfollowup)
1
Werneret
al282
RCT
unblinded
Very
serious
limitations
(c)
Noserious
inconsistency
Noserious
indirectness
(a)Unblindedstudy,nodetailsonrandomisationprocessandunclearallocationconcealment.
NationalClinicalGuidelineCentre,2013.
158
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
(b)20%patientsdroppedoutat12months
(c)Singleblindedstudywithunclearallocationconcealment,highdropoutrateinbotharms10ofthe33patientsintheinterventiongrouplosstofollowup(5droppedoutat3monthsandanother5dropped
outat9months);9ofthe16controlslosstofollowup;5additionalcontrolpatientswererecruitedafterthetreatmentended.
(d)Imprecisioncouldnotbeassessedbecauseonlymeansofdatawerereported.
(e)Relative/absoluteeffectcouldnotbeestimatedasnostandarddeviationwasprovidedinthestudy.
(f)Pvalueasreportedbytheauthors.
NationalClinicalGuidelineCentre,2013.
159
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
6.4.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingdifferentintensitiesofmultidisciplinaryrehabilitation
wereidentified.
Newcosteffectivenessanalysis
FullmethodsandresultsarepresentedinAppendixK;asummaryisprovidedbelow.
TheGDGidentifiedthecomparisonofmoreintensiveprogrammesofrehabilitationforpeoplewith
strokewithlessintensiveprogrammesasahighpriorityareaforeconomicanalysis.
Moreintensiverehabilitationmaybemorecostlytodeliverthanlessintensiverehabilitationbecause
itmayrequireadditionalstafftime.However,additionalcostsmaybeoffsetbyanimprovementin
outcomesforthepatient(suchasindependencyinactivitiesofdailyliving),leadingtoincreased
QALYsandpotentiallyareductioninfuturehealthcareandsocialcarecosts.
Thefollowinggeneralprincipleswereadheredtoindevelopingthecosteffectivenessanalysis:
TheGDGwasconsultedduringtheconstructionandinterpretationofthemodel.
Modelinputswerebasedonthesystematicreviewoftheclinicalliteraturesupplementedwith
otherpublisheddatasourceswherepossible.
Whenpublisheddatawasnotavailableexpertopinionwasusedtopopulatethemodel.
Modelinputsandassumptionswerereportedfullyandtransparently.
Theresultsweresubjecttosensitivityanalysisandlimitationswerediscussed.
ThemodelwaspeerreviewedbyanotherhealtheconomistattheNCGC.
Modeloverview
Acostutilityanalysiswasundertakentoevaluatethecosteffectivenessofmoreintensiveversusless
intensivestrokerehabilitation.Lifetimequalityadjustedlifeyears(QALYs)andcostswereestimated
fromacurrentUKNHSandpersonalsocialservicesperspective.Asisstandardpracticeineconomic
evaluation,bothcostsandQALYSwerediscountedtoreflecttimepreference;arateof3.5%per
annumwasusedinlinewithNICEmethodologicalguidance187.Thecosteffectivenessoutcomeofthe
modelwascostperQALYgained.
TheanalysiswasprimarilybasedondatafromtheUKclinicalstudyreportedbyRyanandcolleagues,
2006225describedintheclinicalreviewabove.
Aprobabilisticanalysiswasundertakentoevaluateuncertaintyinthemodelinputestimates.In
addition,varioussensitivityanalyseswereundertakentotesttherobustnessofmodelassumptions
anddatasources.Inthese,oneormoreinputswerechangedandtheanalysisreruntoevaluatethe
impactonresults.
TheGDGnotedthattheintensitylevelinthemoreintensiverehabilitationarminthestudyreported
byRyanandcolleagueswaslikelytobelowerthanthatnowspecifiedbythestrokequality
standard188.Wethereforeundertookexploratorythresholdanalysestoprovideinformationtohelp
informtheGDGdecisionmaking.
NationalClinicalGuidelineCentre,2013.
160
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Population
Thepopulationforthecosteffectivenessanalysiscomprisedadultsandyoungpeopleaged16or
olderwhohavehadastrokeandrequiredrehabilitation.
Comparators
Thecomparatorsinthemodelwere:
Lessintensivemultidisciplinaryrehabilitation
Moreintensivemultidisciplinaryrehabilitation
FollowingRyanetal.(2006)225,theinterventionwasassumedtobedeliveredathome.Lessintensive
rehabilitationwasthreeorlessfacetofacecontactsperweek,for12weeksmaximum.More
intensiverehabilitationinthestudywassixormorefacetofacecontactsperweek,for12weeks
maximum.
Modelstructure
Alifetableapproachwastakentotheanalysis.LifetablesforEnglandandWaleswereadjustedfor
theincreasedmortalityinpeoplewhohavehadastroke.Thisestimatedthenumberofpeoplealive
aftereach3monthperiod(eachcycle)andthiswasusedtoestimatelifeyearsforpeopleinthe
model.Itwasassumedthatmortalityisnotimpactedbythetypeofrehabilitationreceivedandso
lifeexpectancydidnotvarybycomparatorinthemodel.
Aqualityoflife(utility)valuewasattributedtopeoplewhowerealiveinthemodelthatdepended
onthetypeofrehabilitationreceived(moreintensiveorlessintensive).Thisresultedin
differencesinQALYsbetweenpatients.
Differencesintotalcostsbetweenthemoreandlessintensiverehabilitationgroupsweredueto
differencesinthecostofdeliveringrehabilitationthiscostwasincurredinthefirst3monthcycle.It
wasassumedinthebasecaseanalysisthatinthepostrehabilitationperiodcostsdidnotvary
betweenthemoreintensiveandthelessintensiverehabilitation.
Modelinputs
Modelinputswerebasedonclinicalevidenceidentifiedinthesystematicreviewundertakenforthe
guideline,supplementedbyadditionaldatasourcesasrequired.Modelinputswerevalidatedwith
clinicalmembersoftheGDG.Asummaryofthemodelinputsusedinthebasecase(primary)
analysisisprovidedinTable36below.Moredetailsaboutsources,calculationsandrationalecanbe
foundinthefulltechnicalreportinAppendixK.
Table36: Summaryofbasecasemodelinputs
Input
Data
Source
Probability
distribution
Comparators
Lessintensiverehabilitation
Moreintensiverehabilitation
Population
Peoplewhohavehadastroke
andneedrehabilitation
Perspective
UKNHS&PSS
NICEreferencecase187
Timehorizon
Lifetime
187
Discountrate
Costs:3.5%
Outcomes:3.5%
NICEreferencecase
n/a
Cohortsettings
NationalClinicalGuidelineCentre,2013.
161
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Probability
distribution
Input
Data
Source
Ageonentrytomodel
77years
Ryanetal.2006225
%female
225
61%
Ryanetal.2006
Mortality
Fixed
Fixed
Mortalityrate
Agedependent
EnglandandWales200709life
tables192
Fixed
Mortalityrate
adjustmentforstroke
(SMR)
Female:2.85(CI:2.66,3.05)
Male:2.58(CI:2.43,2.75)
BronnumHansenetal.200134
Lognormal
Qualityoflife(utility)
Beforerehabilitation
0.54
225
Fixed
225
Ryanetal.2006
Changeafterless
intensiverehabilitation
0(SE0.04)
Ryanetal.2006
Normal
Differenceinchange
withmoreversusless
intensiverehabilitation
0.14(SE0.05)
Ryanetal.2006225
Normal
Longtermutility
assumption
Scenario1:differenceis
maintainedoverlifetime
Scenario2:difference
disappearsovertime(3
months,1yearor5years)
Assumptions
n/a
Costs
Rehabilitationcosts
Lessintensive:634
Moreintensive:865
Derivedfromresourceuseand
unitcostsbelow
n/a
Totalnumberof
rehabilitationsessions
Less:17.9(SE1.19)
Difference,moreless:6.5(SE
1.76)
Ryanetal.2006225
Gamma
Normal
Lengthofrehabilitation 45minutes
session
Assumptionbasedontrial
Fixed
range(3060minutes)(Personal
communicationAWRyan,email
January2011)
Personneldelivering
rehabilitation
Professional:75%sessions
Assistant:25%sessions
Assumption
Fixed
Costperhourhome
visit:rehabilitation
professional(a)
54
PSSRU2010:Community;hour
costofhomevisiting50;band
6(b);includingqualifications
Fixed
Costperhourhome
visit:rehabilitation
assistant
27
PSSRU2010:Clinicalsupport
Fixed
workernursing(community);
perhourspentonhomevisits50;
band3(b);including
qualifications
Postrehabilitation
costs
Nodifference
Assumption
Fixed
CI=95%confidenceinterval;n/a=notapplicable;PSSRU=PersonalSocialServicesResearchUnit;SMR=standardised
mortalityratio;SE=standarderror
(a) Physiotherapist,occupationaltherapistandspeechandlanguagetherapist
(b) CostswerecalculatedusingPSSRUdataandapproachbutwiththesalarybandstated
NationalClinicalGuidelineCentre,2013.
162
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Results
Theanalysisfoundthatmoreintensiverehabilitationwascosteffectivecomparedtolessintensive
rehabilitation,basedonlevelsofinterventionandoutcomesfromtheRyanetal.2006study225.
Therewasanadditionalcostassociatedwithmoreintensiverehabilitationasmorerehabilitation
sessionswereprovided;howeverthiswasoffsetbytheadditionalimprovementinqualityoflifethat
resultsinhigherQALYs.Thisconclusionwasseenwithalllongtermutilityscenarios.Therewaslow
withinanalysisuncertaintyaboutthisconclusion.Itwasalsorobusttoarangeofsensitivityanalyses
aroundinputparameters
Table37: Basecaseresultsmoreintensiveversuslessintensiverehabilitation(probabilistic
analysis)
Analysis
Meancost
difference
(moreless)(a)
MeanQALY
difference
(moreless
Incremental
cost
effectiveness
ratio(ICER)
%simulations
more
intensivecost
effective
(20K/QALY)
Scenario1differenceinutilitymaintainedovertime
Maintainedoverlifetime
226
0.70
324
99%
Scenario2utilitydifferencedisappearsovertime
Disappearsover3months
228
0.03
6,722
95%
Disappearsover1year
228
0.08
2,751
99%
Disappearsover5years
226
0.29
776
100%
(a) Minordifferenceareduetoresultsbeingfromdifferentrunsoftheprobabilisticanalysis
Thresholdanalyses
FullresultstablesareshowninthefulltechnicalreportinAppendixK.
Costs:
Ananalysiswasundertakentodeterminethecostdifferencethresholdwhereintensive
rehabilitationwasnolongercosteffective(usinga20,000perQALYgainedcosteffectiveness
threshold).Underthemostconservativelongtermutilityassumption(wheretheutilitydifference
observedattheendofrehabilitationhaddisappearedover3months),moreintensiverehabilitation
wouldnolongerbecosteffectiveifthedifferenceinrehabilitationcostwasmorethan685
(equivalenttoadifferenceofabout17sessions,of45minutes,witharehabilitationprofessional).
Underthemostfavourableutilityassumption(wherethedifferenceobservedattheendof
rehabilitationwasmaintainedindefinitely),moreintensiverehabilitationremainedcosteffective
untilthedifferenceinrehabilitationcostsexceeded13,433(equivalenttoadifferenceofover300
sessionswitharehabilitationprofessional).
QALYs:
Wealsoundertookathresholdanalysiswherewevariedthedifferenceinthenumberof
rehabilitationsessionsbetweenthegroupsandthencalculatedwhatQALYdifferencewouldbe
requiredforittobeconsideredcosteffective.TheGDGestimatedthatincurrentUKpracticealevel
ofinputinlinewiththecurrentNICEqualitystandardwouldbe45minutesofeachrelevanttherapy
atleast5daysaweekaslongastheyarecontinuingtobenefitfromit.Thusover6weeksan
individualmightreceive6090sessionsofinput.TheGDGrecognisedthattherecentStrokeSentinel
audithighlightedthataboutathirdofpatientsreceivedlessthanthiswhileinhospital123.Nodatais
availableforcommunitybasedrehabilitationservices.TheGDGestimatedthatatypicallevelofinput
wouldbethreephysiotherapysessionsperweek,oneoccupationaltherapysessionperweek,and
onespeechandlanguagetherapysessionperweek(thatis30sessions).Thiswouldbeadifference
NationalClinicalGuidelineCentre,2013.
163
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
of60sessionstotalbetweenidealandtypicalinput.Thedifferenceinnumberofsessionswas
thereforevariedbetween6.5(fromtheRyanetal.2006study)and60(basedontheGDGestimate).
ThelifetimeQALYgainrequiredformoreintensiverehabilitationtobecosteffectiverangedfrom
0.010.11whenthedifferenceinnumberofrehabilitationsessionswasvariedbetween6.5and60.
Wethenalsocalculatedthenumberofmonthsforwhich,differentqualityoflife(utility)gainswould
needtobemaintained,inordertoachievetheseQALYgains.Withadifferenceof60rehabilitation
sessionswithmoreintensivecomparedtolessintensiverehabilitation,itwasfoundthatautilitygain
of0.14(asobservedintheRyanetal.2006study)wouldneedtobemaintainedfor9monthsin
orderformoreintensiverehabilitationtobecosteffective.Whenutilitygainwasvariedbetween
0.02and0.24,thisvariedfrom5monthsto64months.
Discussion
Ryanetal.(2006)studygeneralisability
Thekeylimitationsofthisanalysisarethelimitationsoftheclinicaleffectivenessdataforthe
comparisonofmoreandlessintensiverehabilitation.Onlyonestudyreportedutilitydatathatcould
beusedtocalculateQALYsandtheamountofrehabilitationreceivedinthisstudycomparedwith
thecurrentqualitystandard,andevencurrentUKpracticeisverydifferent.Instudyreportedby
Ryanandcolleaguesmoreintensiverehabilitationwasatotalof17sessionsonaverageperperson
andlessintensivewas11.TheGDGestimatedthatalevelofinterventionsimilartothat
recommendedbythecurrentNICEqualitystandardwouldbemorelike90rehabilitationsessionsper
patient(spreadacrossspecialities),andthattypicallevelsofinputintheUKwouldbearound30
sessions.
Itwasnotedthatrehabilitationisacomplexintervention,thatis,theoutcomedoesnotvarylinearly
withinputs.Onepossibilityisthatthereisacriticalthresholdforimprovement.Forexample,ifone
legisweakthepatientwillbeunabletowalk.Thestrengthmayincreaselinearlyfor6weeks,but
onlyinweek7willthepatientwalk.Ifafunctionaloutcomeisused,thepatientwillappearto
plateaufor6weeksandthenmayshowasignificantchangeinfunctionalstatus.Thisagainmakesit
difficulttoextrapolatefromthestudyreportedbyRyanandcolleagues.
Stratification
Itwasnotedthatyoungerpatientsalsooftenhavethecapacitytoparticipateinmoresessionsof
rehabilitationasthisislinkedtocardiovascularfitness,frailtyandcomorbidity,allofwhichtendto
beworseinolderpatients.Theyalsooftenhaveagreaterrangeofneeds(education,work,and
parenting).Yetoftenyoungerpatientsdonotgetmorerehabilitation.Itwasnotpossibleto
undertakesubgroupanalysisonthisbasisinthemodelasnotclinicalstudieshadexaminedthis.
Qualityoflifeassumptions
ThestudyreportedbyRyanandcolleaguesreportedEQ5Dqualityoflifedataat3monthsbutdid
nothaveanylongertermfollowupandsoassumptionsweremaderegardingwhathappenstothe
differenceinqualityoflifeovertimebetweenthegroups.Howeverbothconservativeandmore
favourableassumptionswereexploredinthemodeltotesttheimpactonresults.
Theanalysisdoesnotincludeanyimpactoncarerqualityoflifeastherewasnoevidenceavailable.It
isplausiblethatgreaterfunctionalabilityforthepersonwhohashadastrokemayalsomeanless
burdenontheircarerandthismayleadtoanimprovementinthecarersqualityoflifeaswell.Ifthis
werethecase,thiswouldincreasetheQALYgainwithmoreintensiverehabilitation,makingitmore
costeffective.
NationalClinicalGuidelineCentre,2013.
164
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
Postrehabilitationcosts
Inthebasecaseanalysisweassumednodifferenceinpostrehabilitationcosts;howevergreater
functionalabilitycouldplausiblyresultinlowerdependencyandpotentiallylowersocialcarecosts.
Thiswouldfurtherfavourmoreintensiverehabilitation.
Rehabilitationsetting
ThestudyreportedbyRyanandcolleagueswasbasedoncommunityrehabilitationandsocostsin
themodelarealsobasedoncommunityrehabilitation.TheGDGconsideredthattheamountof
rehabilitationshouldbethesamewhetherdeliveredinthecommunityorinhospital.Inadditionif
rehabilitationwastakingplaceinhospitaltheintensityofrehabilitationwouldmostlikelynotchange
thelengthofstaybutwouldjustimpacttheamountofinputfromdifferentprofessionalswhilstin
hospital.Thereforeineithersettingthecostimpactwouldlargelybeaboutpeoplestimeratherthan
changesinhospitalcapacity,overheadsorhotelcostsandsothiswasnotconsideredlikelytogreatly
impacttheresults.Itwasnotedthatpotentiallymoreintensiverehabilitationduringtheinitial
hospitalisationmayevenreducehospitalstayaspatientsbecomemorefunctionallyablemore
quickly.
6.4.1.3
Evidencestatements
Clinicalevidencestatements
Onestudy225with89participantsfoundnosignificantdifferencebetweentheintensive
rehabilitationgroupandthestandardrehabilitationgroupat3monthsontheBarthelIndex
(MODERATECONFIDENCEINEFFECT).
Onestudy225with89participantsfoundastatisticallysignificantimprovementintheintensive
rehabilitationgroupcomparedwiththestandardrehabilitationgroupat3months,ontheEuroqol
VisualAnalogueScale(MODERATECONFIDENCEINEFFECT)
Onestudy225with89participantsfoundastatisticallysignificantimprovementintheintensive
rehabilitationgroupcomparedwiththestandardrehabilitationgroupat3months,ontheEuroqol
5D(MODERATECONFIDENCEINEFFECT)
Onestudy225with89participantsfoundnosignificantdifferenceontheFrenchayActivitiesIndex
betweentheintensiverehabilitationgroupandthestandardrehabilitationgroupat3monthsfollow
up(LOWCONFIDENCEINEFFECT)
Onestudy196with60participantsfoundthattherewasastatisticallysignificantimprovementinthe
FunctionalIndependenceMeasureintheintensiverehabilitationgroupovera60dayfollowup,
comparedwiththelessintensivehomebasedgroup(LOWCONFIDENCEINEFFECT).
Evidencestatementscouldnotbeproducedforthefollowingoutcome(s)asresultswerenot
presentedinawaythatenabledthesizeoftheinterventionseffecttobeestimated:
ActivitiesofDailyLivingIndex241
FunctionalIndependenceMeasure(Motor)282
Economicevidencestatements
Moreintensiverehabilitationwasfoundtobecosteffectivecomparedtolessintensive
rehabilitation,basedonamodelledanalysisusinglevelsofinterventionandoutcomesfromtheRyan
etal.2006study(24versus18rehabilitationsessions;EQ5Ddifference0.14at3months)andarange
oflongtermutilityassumptions.However,theseconclusionsarelimitedbyconcernsregarding
NationalClinicalGuidelineCentre,2013.
165
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
applicabilityofthestudyreportedbyRyanandcolleaguestocurrentUKpractice.Exploratory
thresholdanalysesfound:
Underthemostconservativelongtermutilityassumption(wheretheutilitydifferenceobserved
attheendofrehabilitationhaddisappearedover3months),moreintensiverehabilitationwould
nolongerbecosteffectiveifthedifferenceinrehabilitationcostwasmorethan685(equivalent
toadifferenceofabout17sessions,of45minutes,witharehabilitationprofessional).
Underthemostfavourablelongtermutilityassumption(wherethedifferenceobservedatthe
endofrehabilitationwasmaintainedindefinitely),moreintensiverehabilitationremainedcost
effectiveuntilthedifferenceinrehabilitationcostsexceeded13,433(equivalenttoadifference
ofover300sessionswitharehabilitationprofessional).
Assumingadifferenceof60sessionsbetweenmoreandlessintensiverehabilitation:autility
differenceof0.14wouldneedtobemaintainedfor9monthsformoreintensivetobecost
effective;adifferenceof0.24for5months;andadifferenceof0.02for64months(about4
years).
6.4.2
Recommendationsandlinktoevidence
33.Offerinitiallyatleast45minutesofeachrelevantstroke
rehabilitationtherapyforaminimumof5daysperweekto
peoplewhohavetheabilitytoparticipate,andwhere
functionalgoalscanbeachieved.Ifmorerehabilitationis
neededatalaterstage,tailortheintensitytothepersons
needsatthattimeg.
34.Considermorethan45minutesofeachrelevantstroke
rehabilitationtherapy5daysperweekforpeoplewhohave
theabilitytoparticipateandcontinuetomakefunctional
gains,andwherefunctionalgoalscanbeachieved.
35.Ifpeoplewithstrokeareunabletoparticipatein45minutes
ofeachrehabilitationtherapy,ensurethattherapyisstill
offered5daysperweekforashortertimeatanintensitythat
allowsthemtoactivelyparticipate.
Recommendations
Relativevaluesofdifferent
outcomes
Theoutcomesofinterestincludedinthereviewwere:
lengthofstay,FunctionalIndependenceMeasure(FIM),BarthelIndex,
QualityofLife(anymeasure),NottinghamActivitiesofDailyLiving,
Rankin,Rivermeadscore,FrenchayActivitiesIndex
Thelimitednumberofstudiesavailableshowedanimprovementin
everymodelofrehabilitation.Twostudies(Smith1981,Werner
1996241,282)whichwerebothpostacutesuggestedanimprovementwith
outpatientintensiverehabilitation.
Onestudy(Ryan2006225)showedabenefitonEQ5Dsocialparticipation
healthrelatedqualityoflifemeasurebutnotonBarthel.Itwasnoted
thattheBarthelbaselinewas16andthemeanBarthelgainwas2.7.The
GDGconsideredthereasonadifferencewasnotseenbetweenthetwo
groupsmayhavebeenduetoceilingeffectsastheBarthelscaleonly
goesto20.Anaveragescoreof18.7wouldindicatethatthepatientsin
Intensity
NationalClinicalGuidelineCentre,2013.
166
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
thestudywerelessseverelydisabled.Thegroupacknowledgedthismay
alsoaccountforthegainsfoundintheEQ5D.
ThepatientsinthestudybyOzdemir196weremoreacuteanditwas
recognisedbytheGDGthattheFIMoutcomegainswereclinicallyhighly
significantreinforcingthevalueofrehabilitationbutthattherewere
limitationsinthestudydesign.
TheGDGnotedthatpatienttolerancetothetherapiesshouldbetaken
intoconsiderationaspatientstolerancewouldvary.TheGDGagreed
thatthereisnolinearrelationshipbetweenoutcomeandintervention.
Tradeoffbetweenclinical
benefitsandharms
ThestudyreportedbyRyanandcolleagues,whichprovidesinformation
ontheEQ5Doutcomeat12weeks,showsthereissignificantdifference
inEQ5Dandthisisclinicallysignificant.TheEQ5Disastandardized
measureofhealthoutcome,domainscovermobility,selfcare,pain,
anxietyanddepressionandusualactivity.Theinterventionwouldaimto
restoreusualactivityandtheGDGagreedthattheywouldexpectthisto
bemaintainedafterthe12weekperiod.
Thegroupinthepaper(Ryan2006)wasarelativelyablegroupsoitis
reasonabletoassumethesegainswouldbemaintained.TheWerner
study282showedthatovera3monthperiod3yearspoststrokethe
intensivegroupimprovedontheFIMoutcomescaleandthiswas
maintainedoverthefollowing9months.ItwasnotedthatFIMcovers
twooftheitemswithintheEQ5D.TheGDGagreedthatacohortthat
wasmoredisabledwouldbeexpectedtomakegreatergainsfromhaving
hadmoreintenserehabilitation.
TheGDGagreedthattherewerenoparticularharmsassociatedwithany
oftheinterventionsdeliveredwithinthestudiesandtheyconsideredthe
benefitsofprovidingrehabilitationattheappropriateindividuallevel
wereclearandthosereceivingmoreintensivetherapywouldbe
expectedtoachievethegreatergains.
Economicconsiderations
Nopublishedeconomicevaluationscomparingmoreandlessintensive
rehabilitationwereundertaken.TheGDGidentifiedthisareaasahigh
priorityforanalysisandacosteffectivenessmodelwasdevelopedbased
onthestudyreportedbyRyanandcolleagues(thiswastheonlystudy
thatreportedqualityoflifedata[EQ5D]suitableforcalculatingQALYs).
Thisanalysisfoundmoreintensiverehabilitationtobecosteffective
comparedtolessintensiverehabilitation.TheGDGnotedthatthese
conclusionswerelimitedbyconcernsregardingapplicabilityofthestudy
reportedbyRyanandcolleaguestocurrentUKpractice,inparticularthe
fairlylowlevelsofrehabilitationinbothgroupscomparedtocurrent
standards;otherlimitationstothisstudyarenotedelsewhereinthis
table.Itwasalsonotedthattheanalysisincorporatedtheadditionalcost
ofmoreintensiverehabilitationbutdidnotincorporateanydownstream
costdifferencesduetoalackofevidenceonwhichtobasethese.
Potentiallytheremaybecostsavingsdownstreamofmoreintensive
rehabilitation;forexample,ifpatientsaremorefunctionallyable,social
carecostsmaybereduced.Ifthisweretobethecasethiswouldfurther
favourmoreintensiverehabilitation.
Duetotheconcernsdescribedaboveaboutapplicability,exploratory
thresholdanalyseswereundertakentohelpinformGDGdecision
making.Thecostdifferencethresholdrangedbetween685(equivalent
toadifferenceofabout17sessionsof45minuteswitharehabilitation
professional)and13,433(equivalenttoadifferenceofover300
sessionswitharehabilitationprofessional),dependingonthe
assumptionmadeabouthowshorttermqualityoflifedifferencesare
maintainedinthelongerterm.Themostconservativeutilityassumption
wasthatthequalityoflifedifferenceobservedat3monthsdisappeared
NationalClinicalGuidelineCentre,2013.
167
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
by6months.Themostfavourableutilityassumptionwasthatthe
differencewasmaintainedindefinitely.Itwasagreedthatwhilethere
maybesomeconvergencebetweengroups,itwaslikelythatoverall
somedifferencewouldbemaintained.
TheGDGestimatedthattherewouldbedifferenceofaround60sessions
betweencurrentpracticeandrehabilitationprovisionatthelevelofthe
NICEqualitystandard.Withthisdifferenceinnumberofrehabilitation
sessionswithmoreintensivecomparedtolessintensiverehabilitation,it
wasfoundthatautilitygainof0.14(asobservedinthestudyreported
byRyanandcolleagues)wouldneedtobemaintainedfor9monthsin
orderformoreintensiverehabilitationtobecosteffective.Whenutility
gainwasvariedbetween0.02and0.24,thisvariedfrom5monthsto64
monthsrespectively.
TheGDGnotedthatthisanalysiswaslargelyexploratorygiventhe
limitationsofthedata.Itwasalsonotedthat,astherelationship
betweenintensitylevelandoutcomeswerenotlinear,extrapolationwas
difficult.However,theyconcludedthatbasedonthethresholdanalyses
itseemedlikelythatifmoreintensiverehabilitationprovidedqualityof
lifebenefitsitwaslikelyitwouldbecosteffective.Thereforeitwas
agreedthatincreasingintensitytothelevelinthecurrentquality
standardwaslikelytobecosteffective.Inaddition,theGDGconsidered
thatabovethiswherepeoplecontinuetomakefunctionalgainsitis
likelythatqualityoflifegainswouldmeanthatprovisionwouldbecost
effective.
Qualityofevidence
WhilstallthestudieshadsomelimitationsmethodologicallytheGDG
consideredthattherewasmodestevidencethatshowedmoreintensive
rehabilitationatthelaterstagespoststrokewasbeneficialas
demonstratedinthestudiesbyRyanandWerner225,282.Moderate
confidenceineffect(Ryan2006)wasfoundforthequalityoflife
outcomeEuroqol5Dwhichdemonstratedasignificantimprovement.
ConfidenceintheresultsshownfortheBarthelandFrenchayoutcomes
wasmoderateandlowanddemonstratednosignificantdifference.A
significantimprovementwasshownfortheFunctionalIndependence
measureovera60dayfollowup.(Ozdemir2001).
TheGDGwereconcernedthatthepatientsinbothgroupsintheRyan
study225werehigherfunctioninginbothgroupsandthereforemaynot
demonstratealotofdifference.ThepatientsintheOzdemirpaper196
wasconsideredtobemorerepresentativeoffunctioninglevelsofstroke
patientsseeninclinicalpractice.
Otherconsiderations
Onlyonestudy(Ozdemir196)waswithinthehospitalsetting,otherswere
outpatient/communitysettings.Noneofthestudieswerestartedwithin
2weeksofonsetofstrokebutsomeaddressedrehabilitationneedsin
thesubacuteandchronicphases.
TheGDGagreeditwasdifficulttostatewhatcouldbeconsidered
intensivefromthestudiesreviewed.Twoofthestudieshad2hours45
daysperweek(Werner1996,Ozdemir2001),whilethestudybyRyan
(2006)describedthenumberofcontactsmade.
TheGDGnotedthattheamountoftherapyhighlightedinthestudies
wouldnotreflecthighlyintensivepracticeversuswhatwouldnowbe
acceptedasconventional.TheGDGnotedthatintensityofrehabilitation
couldbeconsideredintermsoffrequency,time,andduration,andthat
studiesofintensitymaybeconfoundedbyothervariablessuchas
expertise,modeofdelivery,andanyspecificdeficitbeingtargeted.The
GDGagreedthattheevidencedemonstratedthatmorerehabilitation
wasbetter,butwhatremainsuncleariswhatmorerehabilitation
constitutes.TheGDGagreedthelevelofintensitydeliveredwithinthe
NationalClinicalGuidelineCentre,2013.
168
StrokeRehabilitation
Planninganddeliveringstrokerehabilitation
studiesdidnotappeartobeconsistentwithcurrentmedicalpracticeor
aspirations.Itwasnotedthatfurtherresearchisrequired.
Becausethestudiesreviewedprovidednodetailsontheinterventions
delivered(otherthanstatingamixofphysiotherapyandoccupational
therapy),itisnotpossibletomakerecommendationsonwhatshouldbe
deliveredwithinapackageofintensiverehabilitation.Thegroupagreed
thatbestpracticewouldofferinterventionsthataregoaldirectedand
taskorientatedaccordingtoindividualneed.
ThegroupacknowledgedandagreedwiththeStrokeQualityStandard189
whichdefinesrehabilitationtherapyasphysiotherapy,occupational
therapyandspeechandlanguagetherapywithothertreatmentsas
requireddeliveredineitherahospitalorcommunitysetting.Each
therapyisprovidedthroughfacetofacecontacteitherindividuallyoras
partofagrouptreatmentanddoesnotincludeadministrativetasks
relatedtopatients.Thisshouldbeofferedtoallwhohavethephysical
andmentalabilitytoparticipateandwhodemonstratethroughtheir
individualgoalsthattheycontinuetobenefitfromthetherapy.
TheGDGagreeditwasimportantthatpeopleshouldbeabletoreaccess
rehabilitationatanystageofthestrokepathwaywhenneeded.
NationalClinicalGuidelineCentre,2013.
169
StrokeRehabilitation
Supportandinformation
7 Supportandinformation
7.1 Providingsupportandinformation
Provisionofappropriate,accurateandtimelyinformationisakeycomponentofpoststrokecare.It
isacorerecommendationofmanypolicydocuments,suchastheNationalStrokeStrategy61.Despite
this,manyresearchreportsindicatethatpatientsandtheirfamiliesfeeltheirinformationneedshave
beenpoorlymet.Howeverinformationprovisionisanebulousconceptanditisdifficulttodetermine
anappropriateobjectiveoutcome.Itisacknowledgedthatinformationiscommonlypassively
availablethroughleaflets.TheGDGsoughttoidentifyeffectiveactivemethodsofinformation
provisionwhichwouldprovidepositivebenefitsintermsofmoodandactivitiesofdailyliving.
7.1.1
7.1.1.1
Evidencereview:Whatistheclinicalandcosteffectivenessofsupportedinformation
provisionversusunsupportedinformationprovisiononmoodanddepressioninpeople
withstroke?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehada
stroke
Intervention:
Supportedinformationgiving(activeinformation
provision,encouragefeedback,peersupport,
interactivecomputerprogramme)
Comparison:
UnsupportedInformation(suchas,leafletsand
noticeboardinformation)
Outcomes:
Impactonmood/depression:
HospitalAnxietyandDepressionScale
GeneralHealthQuestionnaire
VisualAnalogueMoodScale
StrokeAphasicDepressionQuestionnaire(SADQ)
GeriatricDepressionScale
BeckDepressionInventory
Selfefficacy
GeneralSelfefficacyScale
StrokeSelfefficacyQuestionnaire
LocusofControlScale
Extendedactivitiesofdailyliving
Nottinghamextendedactivitiesofdailyliving
FrenchayActivitiesIndex
Yalemoodscale
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparinginterventionsofsupported
informationwithunsupportedinformationforadultsoryoungpeopleof16yearsoldafterstroke.
Onlystudieswithaminimumsamplesizeof20participants(10ineacharm)wereselected.Five(5)
RCTswereidentified.
Table38summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
NationalClinicalGuidelineCentre,2013.
170
StrokeRehabilitation
Supportandinformation
Table38: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
Ellis,2005
Patientswith
strokeinthe
previous3
monthswithno
severecognitive
impairments.
Additionalinputfrom
theStrokeNurse
Specialist(SNS),who
reviewedpatientsat
monthlyintervalsfor
approximately3
months.Individual
adviceonlifestyle
changes,the
importanceof
medication
complianceandits
relevanceto
secondaryprevention
wasgiven.(N=94)
Usualcare,which
GeriatricDepression
includedgenericrisk
Scale
factoradvicefrom
medicalstaffaswell
astheSNS,given
withintheoutpatient
context.Following
enrolmentthe
controlgrouphadno
furtherinputfrom
theSNS.(N=98)
Hoffmann,
2007112
Patientswith
stroke(mean8.4
dayspost
onset)whohada
reportedEnglish
proficiencylevel;
corrected
hearingand
vision;no
reportedor
observable
dementiaand
weremedically
stable.
Computergenerated
tailoredwritten
informationdesigned
sothatthehealth
professionalproviding
theintervention(in
thistrial,theresearch
nurse)communicates
andcollaborateswith
thepatientto
establishhisorher
informationneeds.
(N=69)
Genericwritten
Selfefficacy
information;aseries HospitalAnxietyand
ofthreestrokefact
DepressionScale
sheetsproducedby
theStroke
Associationof
Queenslandwhich
coveredtopicssuch
ashowstrokeoccurs,
riskfactors,and
physical,cognitive
andemotional
changesfollowinga
stroke.(N=69)
Lowe,
2007159
Patientswitha
primary
diagnosisof
acutestroke,
withoutsevere
cognitiveor
communication
problems
CareFileproject(an
individualised
informationbooklet)
inadditiontousual
care.(N=50)
Usualcare,including
StrokeAssociation
informationleaflets
andfollowupin
StrokeReviewClinic.
(N=50)
Rodgers,
1999218
Medicallystable
patients(5and9
dayspostonset).
Nofurther
detailsprovided.
Multidisciplinary
StrokeEducation
Program(SEP)
consistingofarolling
programofone1hour
smallgroup
educationalsessions
forinpatientsandtheir
informalcarer
followedbysix1hour
educationalsessions
afterdischargefrom
hospital.(N=121)
Informationleaflet
HospitalAnxietyand
(onanumberof
DepressionScale
topics)androutine
Nottingham
communicationwith
ExtendedActivities
nurses,doctorsand
ofDailyLiving
therapystaff
membersthroughout
inpatientstay.
(N=83)
Smith,
2004243
Patientswitha
diagnosisof
Specificallydesigned
strokeinformation
Usualpractice:
membersofthe
75
NationalClinicalGuidelineCentre,2013.
171
OUTCOMES
Mood(Yalesingle
question)
FrenchayActivities
Index.
StrokeRehabilitation
Supportandinformation
STUDY
POPULATION
acutestroke;no
receptive
aphasia;no
cognitive
impairmentand
proficientin
English.
INTERVENTION
(StrokeRecovery
Programme)manual
andpatientswere
invitedtoattend
educationmeetings
everytwoweekswith
membersoftheir
multidisciplinaryteam.
(N=84)
NationalClinicalGuidelineCentre,2013.
172
COMPARISON
strokeunit
multidisciplinary
teamwerefreeto
discussaspectsof
treatmentand
respondtoany
specificqueries.
(N=86)
OUTCOMES
HospitalAnxietyand
DepressionScale
StrokeRehabilitation
Supportandinformation
Comparison:Supportedinformationversusunsupportedinformation
Table39:Supportedinformationversusunsupportedinformationclinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Supported
information
Mean(SD)
Median
(IQR)/
Frequency
Imprecision (%)/
Unsupported
information
Mean(SD)
Median(IQR)/
Frequency
(%)/
Mean
Difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
GeriatricDepressionScore(5monthsfollowup)(Betterindicatedbylowervalues)
Ellis,200575
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
4.3(3.17)
5.1(3.24)
0.80(1.71,
0.11)
MD0.8
lower(1.71
lowerto
0.11higher)
Moderate
0.7
0.50(1.39,
0.39)
MD0.5
lower(1.39
lowerto
0.39higher)
Moderate
0.2
0.20(0.59,
0.19)
MD0.2
lower(0.59
lowerto
0.19higher)
Moderate
0.1
0.40(0.21,
1.01)
MD0.4
Moderate
higher(0.21
lowerto
Selfefficacy(togetinformationaboutthedisease)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
0.2
Selfefficacy(toobtainhelpfromfamily,community,andfriends)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
0.0
Selfefficacy(tocommunicatewiththedoctor)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
0.3
NationalClinicalGuidelineCentre,2013.
173
StrokeRehabilitation
Supportandinformation
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Supported
information
Mean(SD)
Median
(IQR)/
Frequency
Imprecision (%)/
Unsupported
information
Mean(SD)
Median(IQR)/
Frequency
(%)/
Mean
Difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
1.01higher)
Confidence
(ineffect)
Selfefficacy(tocontrol/managedepression)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
0.0
0.3
0.30(0.83,
0.23)
MD0.3
lower(0.83
lowerto
0.23higher)
Moderate
0.4
0.3
0.10(0.18,
0.38)
MD0.1
High
higher(0.18
lowerto
0.38higher)
0.0
0.2
0.2(0.64to
1.04)
MD0.2
Low
higher(0.64
lowerto
1.04higher)
1.5
1.40(0.14,
2.66)
MD1.40
Moderate
higher(0.14
to2.66
higher)
Selfefficacy(tomanagethediseaseingeneral)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Selfefficacy(tomanagesymptoms)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(b)
Anxiety(HospitalAnxietyandDepressionScale)(3monthsfollowup)(Betterindicatedbylowervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
0.1
Anxiety(scoreinHospitalandAnxietyDepressionScale>=11)(3monthsfollowup)(Betterindicatedbylowervalues)
NationalClinicalGuidelineCentre,2013.
174
StrokeRehabilitation
Supportandinformation
Summaryoffindings
Qualityassessment
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Smith,
2004243
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Supported
information
Mean(SD)
Median
(IQR)/
Frequency
Imprecision (%)/
Serious
imprecision
(c)
5/49(10.2%)
Unsupported
information
Mean(SD)
Median(IQR)/
Frequency
(%)/
11/45
(24.4%)
Mean
Difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
RR0.42
(0.16to
1.11)
142fewer
per1000
(from205
fewerto27
more)
Moderate
RR0.76
(0.55to
1.06)
96fewerper Moderate
1000(from
181fewerto
24more)
Anxiety(scoreinHospitalAnxietyandDepressionScale>=11)(6monthsfollowup)(Betterindicatedbylowervalues)
Rodgers,
1999218,
Smith,
2004243
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
Serious
inconsistency imprecision
(c)
44/140
(31.4%)
43/107
(40.2%)
Depression(HospitalAnxietyDepressionScale)(3monthsfollowup)(Betterindicatedbyhighervalues)
Hoffmann,
2007112
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Very
serious
imprecision
(b)
0.4
0.3
0.1(1.46
to1.66)
MD0.1
Low
higher(1.46
lowerto
1.66higher)
31/44
(70.5%)
26/40(65%)
RR1.08
(0.81to
1.46)
52moreper Moderate
1000(from
123fewerto
299more)
Mood(YaleScale)(6monthsfollowup)(Betterindicatedbyhighervalues)
Lowe,
2007159
RCTSingle Noserious
blinded
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecisio
n(c)
Depression(scoreinHospitalAnxietyDepressionScale>=11)(3monthsfollowup)(Betterindicatedbylowervalues)
NationalClinicalGuidelineCentre,2013.
175
StrokeRehabilitation
Supportandinformation
Summaryoffindings
Qualityassessment
Noof
studies
Smith,
2004243
Design
Effect
Limitations
Inconsistency
Indirectness
RCTSingle Noserious
blinded
limitations
Noserious
inconsistency
Noserious
indirectness
Supported
information
Mean(SD)
Median
(IQR)/
Frequency
Imprecision (%)/
Very
serious
imprecisio
n(d)
Unsupported
information
Mean(SD)
Median(IQR)/
Frequency
(%)/
5/49(10.2%) 9/45(20%)
Mean
Difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
RR0.51
(0.18to
1.41)
98fewerper Low
1000(from
164fewerto
82more)
RR0.76
(0.51to
1.14)
76fewerper Moderate
1000(from
156fewerto
44more)
Depression(scoreinHospitalAnxietyDepressionScale>=11)(6monthsfollowup)(Betterindicatedbylowervalues)
Rodgers,
1999218,
Smith,
2004243
RCTSingle Noserious
blinded
limitations
Noserious
inconsistency
Noserious
inconsistency
Serious
imprecisio
n(c)
35/140
(25%)
34/107
(31.8%)
NottinghamExtendedActivitiesofDailyLiving(6monthsfollowup)(Betterindicatedbyhighervalues)
Rodgers,
1999218
RCTSingle Noserious
blinded
limitations
Noserious
inconsistency
Noserious
indirectness
(e)
7(022)
8(021)
(f)
0.69(h)
Moderate
(e)
0(023)
(f)
(f)
High(e)
(f)
High(e)
FrenchayActivitiesIndex(3monthsfollowup)(Betterindicatedbyhighervalues)
Smith,
2004243
RCTSingle Noserious
blinded
limitations
Noserious
inconsistency
Noserious
indirectness
(e)
1(030)
FrenchayActivitiesIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
Smith,
2004243
RCTSingle Noserious
blinded
limitations
Noserious
inconsistency
Noserious
indirectness
(e)
5(032)
3(033)
(a)
ConfidenceintervalcrossesoneendofdefaultMID.
ConfidenceintervalcrossesbothendsofdefaultMID.
(c)
ConfidenceintervalcrossesoneendofdefaultMID.
(d)
ConfidenceintervalcrossesbothendsofdefaultMID.
(e)
Imprecisioncouldnotbeassessedbecauseonlymedianandinterquartilerangesofdatareported.
(f)
Relativeandabsoluteeffectcouldnotbeassessedbecausemedianandinterquartilerangesofdatareported.
(b)
NationalClinicalGuidelineCentre,2013.
176
(f)
StrokeRehabilitation
Supportandinformation
7.1.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingsupportedinformationprovisionwithusualcarewere
identified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Thestudiesincludedintheclinicalreviewuseddifferentinterventions.Typicalunitcostsrelevantto
theinterventionsinthestudiesincludedintheclinicalreviewwerereviewedbytheGDGin
conjunctionwiththestudyinterventiondescriptionstoaidconsiderationofcosteffectiveness.The
studyinterventionsaredescribedinfullinTable38.Estimatedunitcostshforrelevantpersonnelare
listedbelow.
AmultidisciplinarystrokeeducationprogramwasdescribedbyRodgers,1999218consistingofone
1hourgroupsessionandsix1hoursessionspostdischarge.Eachsessionwasledbyamemberof
theteam.Theusualcarecomparatorincludedroutinecommunicationwithhealthcare
professionalsandatelephonehotlinenumber.
o Districtnurse(band6)51perhourspentwithapatient
o Clinicalpsychologist(band8a)136perhourofclientcontact
o Speechandlanguagetherapist(band6)47perhourofclientcontact
o Occupationaltherapist(band6)45perhourofclientcontact
o Physiotherapist(band6)48(community)and45(hospital)perhourofclientcontact
o Socialworker54perhourofclientrelatedwork
Ellis,200575lookedataninterventionprovidedbyaStrokeNurseSpecialist.Thepatientswere
reviewedmonthlyfor3months.Thisinterventionwasadditionaltousualcare.
o Nursespecialist(band7nurseadvanced)81perhourofclientcontact.
Lowe,2007159assessedtheprovisionofinformationbookletstopatients.Thebookletincluded
generalinformationaboutstrokeaswellassectionswerepatientspecificinformationcouldbe
entered.Adiscussion(1520minutes)aboutthecontentofthebookletwasheldwithpatientsby
amemberofthemultidisciplinaryteampriortodischargeseerelevantunitcostsabove.This
interventionwasadditionaltousualcare.
ComputergeneratedtailoredinformationwasprovidedtopatientsinthestudybyHoffman,
2007.Patientswereabletoselectthetypeandamountofinformationfromarangeoftopics112.
Aresearchnursealsoelaboratedonthetopicsandplacedthebooklet(generatedfromMicrosoft
Word)inpersonalisedfolders.Thisinterventionwasadditionaltousualcare.
o Nurse(band6nursespecialist)43perhourofpatientcontact.
o Alicenceforthe'Whatyouneedtoknowaboutstrokeeducationpackagecomputer
program40developedbytheUniversityofQueensland,Australiacosts86(excludingVAT)i
h EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandrelevantAgendaforChangesalarybands50(typicalsalarybandidentifiedbyclinicalGDGmembers).
i AU$199(2011)convertedtoUKpounds(2010)usingpurchasingpowerparities194.
NationalClinicalGuidelineCentre,2013.
177
StrokeRehabilitation
Supportandinformation
Patientsweregivenastrokerecoverymanualandinvitedtoattendeducationmeetingseverytwo
weeksinthestudybySmith,2004243.Themanualcontainedinformationaboutstroke,agreed
goalsasdiscussedatthemeetingsaswellasasectionforcarers.Themeetings(approximately20
minutes)werewithamultidisciplinaryteam(doctor,nurse,physiotherapistandoccupational
therapist).Inusualcarecomparatorarminformationleafletswerefreelyavailableandstaff
respondedtospecificquestions.
o Medicalconsultant132percontracthour
o Unitcostsforotherteammembersareaslistedabove.
Evidencestatements
Clinicalevidencestatements
Onestudy75comprising192participantsfoundnosignificantdifferenceindepressionat5months
afterstrokebetweenthegroupthatreceivedsupportedinformationandthegroupthatreceived
unsupportedinformation(MODERATECONFIDENCEINEFFECT).
Onestudy112comprising138participantsfoundnosignificantdifferencebetweenthegroupthat
receivedsupportedinformationandthegroupthatreceivedunsupportedinformationat3months
afterstrokeinselfefficacywiththefollowingsections:
Gettinginformationaboutthedisease(MODERATECONFIDENCEINEFFECT)
Obtaininghelpfromfamily,community,andfriends(LOWCONFIDENCEINEFFECT)
Communicatingwiththedoctor(MODERATECONFIDENCEINEFFECT)
Controlling/managingdepression(MODERATECONFIDENCEINEFFECT)
Managingthediseaseingeneral(HIGHCONFIDENCEINEFFECT)
Managingsymptoms(LOWCONFIDENCEINEFFECT)
Onestudy112comprising138participantsshowedsignificantimprovementinanxietyat3months
afterstrokewiththegroupthatreceivedunsupportedinformationcomparedtothegroupthat
receivedsupportedinformation(MODERATECONFIDENCEINEFFECT)
Onestudy243comprising170participantsfoundnosignificantdifferenceintheproportionof
participantsexperiencedanxietyat3monthsafterstrokebetweenthegroupthatreceived
supportedinformationandtheunsupportedinformationgroup(MODERATECONFIDENCEIN
EFFECT).
Twostudies218,243comprising374participantsfoundnosignificantdifferenceinanxietyat6months
afterstrokebetweenthegroupthatreceivedsupportedinformationandtheunsupported
informationgroup(MODERATECONFIDENCEINEFFECT).
Onestudy112comprising138participantsfoundnosignificantdifferenceindepressionat3months
afterstrokewiththegroupthatreceivedsupportedinformationandthegroupthatreceived
unsupportedinformation(LOWCONFIDENCEINEFFECT)
Onestudy243comprising170participantsfoundnosignificantdifferenceinintheproportionof
participantsexperienceddepressionat3monthsafterstrokebetweenthegroupthatreceived
supportedinformationandtheunsupportedinformationgroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy159comprising100participantsfoundnosignificantdifferenceinmoodat6monthsafter
strokebetweenthegroupthatreceivedsupportedinformationandtheunsupportedinformation
group(MODERATECONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
178
StrokeRehabilitation
Supportandinformation
Twostudies218,243comprising374participantsfoundnosignificantdifferenceindepressionat6
monthsafterstrokebetweenthegroupthatreceivedsupportedinformationandtheunsupported
informationgroup(MODERATECONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
7.1.2
Recommendationsandlinktoevidence
36.Workingwiththepersonwithstrokeandtheirfamilyorcarer,
identifytheirinformationneedsandhowtodeliverthem,taking
intoaccountspecificimpairmentssuchasaphasiaandcognitive
impairments.Pacetheinformationtothepersonsemotional
adjustment.
37.Provideinformationaboutlocalresources(forexample,leisure,
housing,socialservicesandthevoluntarysector)thatcanhelpto
supporttheneedsandprioritiesofthepersonwithstrokeand
theirfamilyorcarer.
38.Reviewinformationneedsatthepersons6monthandannual
strokereviewsandatthestartandcompletionofanyintervention
period.
39.NICEhasproducedguidanceonthecomponentsofgoodpatient
experienceinadultNHSservices.Followtherecommendationsin
PatientexperienceinadultNHSservices(NICEclinicalguideline
138)j.
Recommendations
Relativevaluesofdifferent Itisdifficulttoidentifyandcapturethedifferentoutputsofinformation
outcomes
provision.Arangeofpotentialoutputsinclude:abetterunderstandingof
stroke,changesinbehaviour(forexamplecompliancewithmedication,
increasedsatisfactionwithservices,decreasedanxietyanddepression,
increasedactivityandparticipationinsocialrolesafterstroke).
TheGDGconsideredthattherelationshipbetweeninformationprovisionand
theoutputsareunlikelytobelinearandwillbemoderatedbyalargerangeof
factorsincluding:personalfactors(patientseducationallevels,premorbid
mentalhealthstatus),diseasefactors(suchascognitivefactorsandaphasia),
andsocialfactors(suchasfamilybeliefs).Thetimingandpacingofinformation
topatientsneedsisalsocritical.Patientgroupsrepeatedlyaskformore
informationandthereforefactorstobeconsideredarewhatinformationis
required,theappropriatemethodofdeliveryforthepatientandthe
timelinessofprovision.
Tradeoffbetweenclinical
benefitsandharms
Onthebasisofthesestudiesitappearsthatadditionalsupportedinformation
provisiondoesnotaffectimprovementinmood.
Thebaselinescoresweresuchthatthemajorityofthepatientswerenot
depressedandthechangescoreswerenotclinicallysignificant.
For recommendations on continuity of care and relationships see section 1.4 and for recommendations on
enabling patients to actively participate in their care see section 1.5.
NationalClinicalGuidelineCentre,2013.
179
StrokeRehabilitation
Supportandinformation
TheGDGnotedthatstandardinformationbriefingmaynotberelevant;and
thatperhapsguidingpatientstowardarticulatingwhatinformationtheyneed
wouldbeofmorebenefittopatients.
Economicconsiderations
Supportedinformationprovisionmayhavearesourceimpactoverusualcare
butthiswouldvarydependingonthespecificintervention,andonthe
patientsneeds.Theclinicalstudiesrevieweddidnotprovideevidencethat
patienthealthoutcomeswereimproved;however,asnotedabove,theGDG
consideredthatthebenefitsofinformationprovisionwerehardtomeasure
andtheremaybeadditionalaimsandbenefitsofinformationgivingvaluedby
patientsbutnotcapturedbytheseoutcomes.
Qualityofevidence
TheGDGthoughtthatthepatientsperceptionsandattributionsareinformed
byawiderangeofsources,muchofwhichisavailableinsideandoutsideofthe
healthcareenvironment.Theincludedstudiesexaminedtheaddedvalueofa
morestructuredapproachtoinformationprovisionprovidedbyhealthcare
professionals.Thestudiesarenecessarilyreductionistinacomplex
environment.
Thecomponentsoftheinterventionswereinadequatelydescribedandthe
evidencewasgenerallyofhightolowqualityfortheoutcomesassesseddueto
imprecisionoftheeffectestimate.
Therewasconsensusthatprovisionofinformationwasuseful.Therewasvery
littleconsensusonhowandwhenthisshouldbedone,somethingthatis
reflectedinthestudydesigns.TheGDGnotedthatthestudybyEllis75was
focusedonassessingtheroleofthenursespecialistratherthanthe
intervention.TheHoffmanstudy112includedonlyEnglishspeakersand
thereforeitdidnotreflectclinicalpractice.
Otherconsiderations
TheGDGagreedthatinformationprovidedislikelytovaryfrompatientto
patientandneedstoreflectpatientsneedsandpriorities,familyexpectations,
andthelocalresourcesprovidedbyleisure,housing,socialservicesandthe
voluntarysectortosupportthese.Informationneedsarelikelytovaryat
differentstagesafterstroke.
TheGDGnotedthatspecificgroupssuchasthosewithdysphasiaorcognitive
impairmentsmayhaveparticularinformationneeds.
NationalClinicalGuidelineCentre,2013.
180
StrokeRehabilitation
Cognitivefunctioning
8 Cognitivefunctioning
Followingstroke,manypeopleexperiencedifficultiesinarousal,attention,concentration,memory,
perception,problemsolving,decisionmaking,insightandotherareasofcognitionthatimpedetheir
abilitytofunctionineverydayactivities.Cognitiveabilitiesanddisabilitiesmustbeconsideredin
addressingallareasoffunctioningincludingcommunication,mobility,selfcare,socialinteraction,
recreationalpursuits,andotherproductiveactivitiessuchasschoolorwork.
Cognitiverehabilitationcanbeconceptualisedintwoways.Itcanbedesignedtofacilitate
restorationoforcompensationforunderlyingimpairment(s)withtheaimofimprovingfunctional
performance.Oftenbothrestorativeandcompensatoryapproachesareintegratedinorderto
maximisefunction.Theseinterventionsshouldbebasedonthenatureandscopeof
neuropsychologicalimpairmentsidentifiedonneuropsychologicalassessmentsusingvalidated
standardisedtests,andanassessmentoftheimpactoftheseimpairmentsonfunction.
Inpracticalterms,attention,memory,spatialawareness,apraxiaandperceptionarecriticalto
successfulrehabilitationinmanyotherdomains.However,thischapteroftheguidelinefocuseson
visualneglect,memoryandattention.
Forthereviewofpsychologicaltherapiesinrelationtoemotionalfunctioningforpeopleafterstroke
pleaseseechapter9
8.1 Visualneglect
Themoststrikingfeatureofneglectisaninabilityofthepatienttoorienttowardsandattendto
stimulieventheirownbodypartsinthecontralesionalspace(theleftsideforpatientswithright
hemispherelesions)12,despiteanabilitytomakesuchexploratorymovementswhenprompted.The
severityoftheinattentionmayvaryaccordingtocontext.Incircumstanceswherepatientsarealso
unawareoftheirdeficit(anosognosia),thedisorderbecomesaparticularlydifficultsyndrometo
rehabilitate3.Persistentneglectisoftenassociatedwithpoorfunctionaloutcome4,impactingon
everydaytaskssuchasdressing,feedingandreading.
Neglectisdifficulttotreatinclinicalpractice.Thisdifficultycanbeattributedtothefactthatitisa
syndromeanddoesnotseemtobeduetoadisruptionofjustonecognitiveprocessbutratherdue
todifferentcombinationsofneuropsychologicaldeficits4.Itis,therefore,unlikelythatasingle
therapeuticinterventionwillsuitallindividuals.
Neglectcanpresentindifferentmodalitiesforexample,sensory,motororvisual.Unilateralvisual
neglectisarelativelycommonproblemparticularlyfollowinghemisphericstroke.Approachesto
treatmentincludebothrestorativeandcompensatoryapproaches,includingtheuseofgoggleswith
prismsthatinducearightwardopticalshiftof~515hasbeentried.Theinducedopticalshift
initiallyleadstoerrorsofpointingtotherightofthevisualtarget,leadinginturntocompensatory
leftwardmanualcorrections.Inpatients,thiscompensatorybehaviouristypicallyfollowedbyan
aftereffectwhentheprismsareremovedwithmanualerrorsnowbeingbiasedtowardstheleft
instead.
8.1.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
cognitiverehabilitationversususualcaretoimprovespatialawarenessand/orvisual
neglect?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention:
Prisms,eyepatchesandgoggles,
NationalClinicalGuidelineCentre,2013.
181
StrokeRehabilitation
Cognitivefunctioning
ClinicalMethodologicalIntroduction
Tracktoleft,
Approachessuchascubecopying.
8.1.1.1
Comparison:
UsualCare
Outcomes:
BehaviouralInattentionTest(BIT),
Drawingtests(clockdrawingetc.),
LineBisectiontests,
Allcancellationtests(linecancellation,bellcancellationetc.),
Sentencereading,
Targetscreenexaminations(lumptogetherallcancellationtests
anddrawingtests),
RivermeadPerceptualAssessmentBattery(RPAB)
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingtheclinicaleffectivenessof
cognitiverehabilitationtherapieswithusualcaretoimprovespatialawarenessand/orvisualneglect
foradultsandyoungpeople16orolderwhohavehadastroke.Onlystudieswithaminimumsample
sizeof10participants(5ineacharm)wereselected.Nine(9)RCTswereidentifiedwhichaddressed
visualneglect.Table40summarisesthepopulation,interventionandoutcomesforeachofthe
studies.
Table40: Summaryofstudiesofincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixF.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Fanthome,
199580
Patientswithaprevious
strokeaffectingthe
rightsideofthebody;
under80yearsofage
withnohistoryof
dementiaorpsychiatric
problems.Allpatients
wereinhospitaland
receivingphysiotherapy
andoccupational
therapybutnoprevious
treatmentfortheir
visualneglect.
Wearingtheeye
movementdetection
glasseswhich
providedareminder
bleepifpatients
failedtomovetheir
eyestotheleftfor
15secondsfor2
hoursand40
minutes/weekfor4
weeks.(N=9)
Notreatment
wasprovidedfor
theirvisual
inattentionor
otherperceptual
deficitsfor4
weeks.(N=9)
BIT
conventional
subset
BIT
behavioural
subset.
Kalra,1997130
Acutestrokepatients
(themedianduration
betweentheacute
episodeand
randomizationwas6
days(range214days).
Patientswithvisual
neglectwereidentified
bycomprehensive
multidisciplinary
assessments(including
linebisectiontest).
Modifiedapproach
toconventional
therapyinvolving
spatiomotorcueing
basedonthe
attentionalmotor
integrationmodel
andearlyemphasis
onrestorationof
function.(N=25)
Conventional
therapyinput
concentratingon
restorationof
normaltone,
movement
patternsand
motoractivity
before
addressing
skilledfunctional
activity.
(N=25)
Rivermead
Perceptual
Assessment
Battery(RPAB)
RPAB
cancellation
subtest
RPABbody
imagesubtest.
NationalClinicalGuidelineCentre,2013.
182
StrokeRehabilitation
Cognitivefunctioning
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Nys,2008
Patientswithstrokeand
visualneglectas
assessedbytheBIT.
Patientswhoperformed
belowthecutoffonat
leasttwoofthefour
subtasksoftheBITwere
includedinthestudy.
Fourdayinarow
experimental
treatmentwith10
degreerightward
deviatingprisms.
(N=10)
Fourdayina
row
experimental
treatment
withoutprism.
(N=6)
Linebisection
test
Star
cancellation
test
Representatio
naldrawing
test,
BIT(total
score).
Robertson,
1990215
Patientswithsignificant
unilateralleftfield
visualneglectaccording
toBITanddefinedas
failurein3outof9
tests.
Computerised
scanningand
attentiontraining,14
sessionsof75
minuteseachusually
2times/week.
(N=20)
Exposureto
plausible
computer
activitiesthat
wereconsidered
nottoimprove
cognitive
function
wogames,
quizzesand
simplelogical
gamessuchas
redsandgreens
foranaverageof
11.4hours(SD
5.2)
(N=16)
BIT(total
score)
Letter
cancellation
test.
Robertson,
2002216
Patientswithdiagnosis
ofrighthemispheric
strokeandunilateral
visualneglect(as
definedbyascoreof51
orlessonthestar
cancellationtestofthe
BITorascoreof7or
lessonthelinebisection
test).Participantshad
nootherexisting
comorbiditiesthat
preventorinfluencethe
assessment.
Perceptualtraining
pluslimbactivating
deviceprovided
in12sessionsof45
minutesduration
overa12week
period.(N=19)
Perceptual
BIT
trainingplus
Behavioural
dummy
subset
(inactive)limb
Letter
activatingdevice
cancellation
providedin12
test.
sessionsof45
minutesduration
overa12week
period.(N=21)
Rossi,1990222
Patientswithstrokeand
homonymous
hemianopiaor
unilateralvisualneglect.
15diopterplastic
pressonfresnel
prismsplusreceiving
routine
rehabilitation
programme
(physical,
occupationalspeech
therapy).(N=18)
Noprismbut
receivingroutine
rehabilitation
programme
(physical,
occupational
speechtherapy)
(N=21)
191
NationalClinicalGuidelineCentre,2013.
183
Linebisection
test
Line
cancellation
testtask
Tangent
Screen
Examination.
StrokeRehabilitation
Cognitivefunctioning
STUDY
POPULATION
INTERVENTION
COMPARISON
Tsang,2009
Participantswere
inpatientswithsub
acutestroke(meantime
sincestroke3wks.)with
leftvisualneglectbased
onthetotalscoreon
theBIT.
4weeksof
conventional
occupationaltherapy
withrighthalffield
eyepatchingglasses,
whichwereworn
throughoutthe
occupationaltherapy
treatmentsessions.
Fiveoccupational
therapysessionsof
60minuteseach
session/
week.(N=17)
4weeksof
BIT
conventional
Conventional
occupational
subset.
therapywithout
eyepatching.
Fiveoccupational
therapysessions
of60minutes
each
session/week.
(N=17)
Turton2010264
Righthemisphericfirst
timestrokepatients(at
least20dayspost
stroke)withunilateral
spatialneglect
Participantswere
instructedto
performrepeated
pointingmovements
totargets,usingthe
rightunaffected
handwhilewearing
theprismglasses
(using10dioptre,6
degreeprisms)each
weekdayfor2
weeks.Before
wearingtheglasses,
participantswere
givensomepointing
practice,withvision
oftheterminalpoint
ofmovement,to
ensurethey
understoodthetask
(N=17)
Shamtreatment BIT
usingplain
Conventional
glasseseveryday
subset
duringtheweek
for2weeks.
Participantswere
giventhesame
pointingpractice,
withvisionofthe
terminalpointof
movementasthe
intervention
group.(N=19)
Mancuso2012166
Outpatientswithleft
visualneglectresulting
fromrighthemisphere
vascularlesion.All
patientswereselected
inaccordancewithtests
forneglectwhohad
verylowscoresonat
leasttwo(outofhow
manyisabitunclear)
visualneglecttests.
Participantscarried
outapointing
exercisewhilst
wearingprismatic
lensesproducing
opticalshiftof5
degreestotheright.
Therewereoverall
fiverehabilitation
sessionslasting
about30minutes
eachforoneweek.
Participants
receivedthe
samepointing
exercisewhilst
wearingneutral
lenses.
263
NationalClinicalGuidelineCentre,2013.
184
OUTCOMES
Line
cancellation
tests
Bells
cancellation
tests
Lines
orientation
test
Fourssubtests
ofBIT(line
bisection,
copying
drawings,
findingobjects
anddealing
playingcards
tests)
StrokeRehabilitation
Cognitivefunctioning
Comparison:Cognitiverehabilitationforspatialawarenessand/orvisualneglectversususualcare
Table41:Cognitiverehabilitationforspatialawarenessand/orvisualneglectversususualcareClinicalstudycharacteristicsandclinicalsummaryof
findings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Cognitive
rehabilitation
Frequency
(%)/mean
(SD)/median
(range)
Usualcare
Frequency
(%)/mean
(SD)/
median
(range)
Serious
imprecision
(b)
52(24)
59.9(20.2)
7.9(
22.34,
6.54)
MD7.9
lower
(22.34
lowerto
6.54
higher)
Low
Serious
imprecision
(b)
Nys
60.2(21.9)
Robertson
60.1(18.6)
Nys
61.2(21.2)
Robertson
61.8(21.5)
1.51(
12.86,
9.85)
MD1.51
lower
(12.86
lowerto
9.85
higher)
Low
Veryserious
imprecision
(e,f)
Fanthome
93.4(41.3)
Turton14.8
(18.8)
Fanthome
90.2(48.4)
Turton9.7
(15.9)
4.97(
6.07,
16.00)
MD4.97
higher
(6.077
lowerto
16.00
Verylow
Imprecision
Mean
difference/
RiskRatio
(95%CI)
Absolute
effect/
Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
BIT(totalscore)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Robertson1990 single
215
blind
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
BIT(totalscore)(6monthsfollowup)(Betterindicatedbyhighervalues)
2
RCTs
single
Nys2008
191
blind
Robertson1990
215
Serious
limitations
(c)
Noserious
inconsistency
Noserious
indirectness
BITconventional(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Fanthome
199580
Turton2010264
RCTs
single
blinded
Serious
limitations
(d)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
185
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Cognitive
rehabilitation
Frequency
(%)/mean
(SD)/median
(range)
Usualcare
Frequency
(%)/mean
(SD)/
median
(range)
Absolute
effect/
Mean
Mean
difference/ Differen
RiskRatio
ce(MD) Confidence
(95%CI)
(95%CI) (ineffect)
higher)
Serious
imprecision
(b)
Fanthome
97.6(27.9)
Nys
123.2(25.1)
Turton24.5
(15.7)
Fanthome
84(50.3)
Nys
116.5(36.5)
Turton21.8
(22.2)
4.11(
7.03,
15.25)
MD4.11
higher
(7.03
lowerto
15.25
higher)
Low
Noserious
imprecision
Fanthome
37.6(21.3)
Robertson
30.2(11.9)
Fanthome
42.9(29.3)
Robertson
31.2(11.9)
1.38(
8.43,
5.67)
MD1.38
lower
(8.43
lowerto
5.67
higher)
Moderate
Noserious
imprecision
Fanthome
45.1(19)
Robertson
30.1(11.5)
Fanthome
39(26)
Robertson
32.8(11.9)
1.76(
8.62,
5.09)
MD1.76
lower
(8.62
lowerto
5.09
higher)
Moderate
Serious
30.1(13.2)
33.5(12.6)
3.40( MD3.4
Imprecision
BITconventional(12monthsfollowup)(Betterindicatedbyhighervalues)
3
Fanthome
199580
Nys2008191
Turton2010264
RCTs
single
blinded
Serious
limitations
(d)
Noserious
inconsistency
Noserious
indirectness
BITbehavioural(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Fanthome
199580
Robertson
2002216
RCTs
single
blinded
Serious
limitations
(d)
Noserious
inconsistency
Noserious
indirectness
BITbehavioural(23monthsfollowup)(Betterindicatedbyhighervalues)
2
Fanthome
199580
Robertson
2002216
RCTs
single
blinded
Serious
limitations
(d)
Noserious
inconsistency
Noserious
indirectness
BITbehavioural(6monthsfollowup)(Betterindicatedbyhighervalues)
1
RCT
Veryserious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
186
Verylow
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
Noofstudies
Robertson
2002216
Design
single
blinded
Effect
Limitations
limitations
(g,h)
Inconsistency
inconsistency
Indirectness
indirectness
Cognitive
rehabilitation
Frequency
(%)/mean
(SD)/median
(range)
Usualcare
Frequency
(%)/mean
(SD)/
median
(range)
Absolute
effect/
Mean
Mean
difference/ Differen
RiskRatio
ce(MD) Confidence
(95%CI)
(95%CI) (ineffect)
11.42,
lower
4.62)
(11.42
lowerto
4.62
higher)
Serious
imprecision
(b)
Nys
2.6(2.8)
Tsang
0.76(1.6)
Nys
2.5(2.5)
Tsang
0.02(2.46)
0.56(
1.79,
0.68)
MD0.56
lower
(1.79
lowerto
0.68
higher)
Low
Noserious
imprecision
Nys
6.1(3.4)
Rossi
0.68(0.85)
Nys
5.2(3.1)
Rossi
2.2(2.29)
1.29(
2.29,
0.29)
MD1.29
(2.29to
0.29
lower)
Low
Nys
21.5(13.1)
Tsang
8.65(13.15)
Nys
20.7(19)
Tsang
1.88(5.02)
5.99(
0.25,
12.23)
MD5.99
higher
(0.25
lowerto
12.23
higher)
Low
Imprecision
imprecision
(b)
Linebisection(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Nys2008191
Tsang2009
263
RCT
single
blinded
Serious
limitations
(i)
Noserious
inconsistency
Noserious
indirectness
Linebisectiontest(1monthfollowup)(Betterindicatedbyhighervalues)
2
Nys2008191
Rossi1990
RCTs
single
blinded/
unblinded
Veryserious
limitations
(j)
Noserious
inconsistency
Noserious
indirectness
StarCancellationtest(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Nys2008191
Tsang2009263
RCTs
single
blinded
Serious
limitations
(i)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
NationalClinicalGuidelineCentre,2013.
187
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Cognitive
rehabilitation
Frequency
(%)/mean
(SD)/median
(range)
Usualcare
Frequency
(%)/mean
(SD)/
median
(range)
Veryserious
imprecision
(e,f)
43.1(13.7)
42.3(16.4)
0.80(
14.83,
16.43)
MD0.8
higher
(14.83
lowerto
16.43
higher)
Verylow
Serious
imprecision
(b)
224.32
(55.38)
199.44
(64.87)
24.88(
8.55,
58.31)
MD24.88
higher
(8.55
lowerto
58.31
higher)
Low
Kalra
37.19(13.1)
Kalra
30.12
(18.45)
7.07(
1.80
15.94)
MD7.07
higher
(1.80
lowerto
15.94
higher)
Low
13.19(1.47)
9.72(1.33)
3.47
(2.69,
4.25)
MD3.47
higher
(2.69to
Moderate
Imprecision
Mean
difference/
RiskRatio
(95%CI)
Absolute
effect/
Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
StarCancellationtest(1monthfollowup)(Betterindicatedbyhighervalues)
1
Nys2008191
RCT
single
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
RPAB(totalscore)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Kalra1997130
RCT
single
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
RPAB(cancellationsubtest)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Kalra1997130
RCTs
single
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
RPAB(bodyimagesubtest)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Kalra1997130
RCTs
single
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
NationalClinicalGuidelineCentre,2013.
188
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Cognitive
rehabilitation
Frequency
(%)/mean
(SD)/median
(range)
Usualcare
Frequency
(%)/mean
(SD)/
median
(range)
Absolute
effect/
Mean
Mean
difference/ Differen
RiskRatio
ce(MD) Confidence
(95%CI)
(95%CI) (ineffect)
4.25
higher)
Robertson
43.4(30.4)
Tsang
10(12.12)
Robertson
43.2(28.3)
Tsang
2.65(6.52)
6.61
(0.41,
12.80)
MD6.61
higher
(0.41to
12.80
higher
Moderate
20(16.4)
23.1(14.5)
3.10(
13.21,
7.01)
MD3.10
lower
(13.21
lowerto
7.01
higher)
Low
15/18(83.3%) 7/21
(33.3%)
RR2.50
(1.32to
4.74)
500more
per1000
(from185
moreto
625more)
Low
2.4(4.24)
7.40(
11.78,
3.02)
MD7.4
lower
(11.78to
Low
Lettercancellationtest(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Robertson
1990215
Tsang2009263
RCTs
single
blinded
Serious
limitations
(c)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
LetterCancellationtest(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Robertson
1990215
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
Tangentscreenexamination(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Rossi1990222
RCT
unblinded
Veryserious
limitations
(o)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
LineCancellationtest(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Rossi1990222
RCT
unblinded
Veryserious
limitations
(o)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
9.8(9.17)
NationalClinicalGuidelineCentre,2013.
189
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Cognitive
rehabilitation
Frequency
(%)/mean
(SD)/median
(range)
Usualcare
Frequency
(%)/mean
(SD)/
median
(range)
Absolute
effect/
Mean
Mean
difference/ Differen
RiskRatio
ce(MD) Confidence
(95%CI)
(95%CI) (ineffect)
3.02
lower)
Nys
0.8(0.8)
Tsang
0.18(1.19)
Nys
1(0.9)
Tsang
0.18(0.88)
0.08(
0.63,
0.47)
MD0.08
lower
(0.63
lowerto
0.47
higher)
Verylow
1.6(1)
2.3(0.5)
0.70(
1.44,
0.04)
MD0.7
lower
(1.44
lowerto
0.04
higher)
Low
Representationaldrawingtest(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Nys2008191
Tsang2009263
RCTs
single
blinded
Serious
limitations
(i)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(p)
Representationaldrawingtest(1monthfollowup)(Betterindicatedbyhighervalues)
1
Nys2008191
RCT
single
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b,e)
(a)
Partialrandomizationandunclearallocationconcealment
ConfidenceintervalcrossedoneendofdefaultMID.
(c)
Onehadpartialrandomization(Robertson,1990),onestudyhadunclearrandomization(Nys,2008)andbothstudieshadunclearallocationconcealment.
(d)
Unclearrandomizationprocessandallocationconcealment.
(e)
Smallsamplesize,eitherarm<10participants(Nys2008;Fanthome1995).
(f)
ConfidenceintervalcrossedbothendsofdefaultMID.
(g)
Unclearrandomizationandallocationconcealment.
(h)
Dropoutrate20%ineacharm(Robertson2002).
(i)
Unclearrandomizationandallocationconcealment(Nys,2008)
(j)
Unblinded(Rossi1990)withunclearrandomizationandallocationconcealment.
(k)
Unclearrandomizationandallocationconcealment.
(l)
Inadequaterandomizationandunclearallocationconcealment.
(b)
NationalClinicalGuidelineCentre,2013.
190
StrokeRehabilitation
Cognitivefunctioning
(m)
Imprecisioncouldnotbeassessedasresultswerepresentedonlyinmedians(range).
Nomeanorstandarddeviationwasreportedinthestudy,socouldnotbemetaanalysedandunabletocalculaterelativeandabsoluteeffect.
(o)
Unblindedstudywithunclearrandomizationandallocationconcealment(Rossi1990).
(p)
ConfidenceintervalcrossedbothendsofdefaultMID.
(n)
Narrativesummary
ThefollowingstudiesaresummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Onerandomisedcontrolstudy166comprising29participants,whohadtestedpositiveforvisualneglect,reportedimprovementsforboththeexperimental
(prismaticlensesof5degreespluspointingtask)andcontrolgroup(shamlensespluspointingtask).However,participantswearingprismaticlensesdid
notimprovesignificantlymorethanthecontrolparticipants(VERYLOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
191
StrokeRehabilitation
Cognitivefunctioning
8.1.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingcognitiverehabilitationinterventionswithusualcareto
improvespatialawarenessand/orvisualneglectwereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Thestudiesidentifiedintheclinicalreviewusedavarietyofdifferentinterventions.TheGDG
consideredthatatypicalcostcouldbeestimatedbasedontheresourcesreportedintheRCTby
Turtonetal(2010)264thatlookedatusingprismglassesintensessionswithanoccupational
therapist.Theauthorwascontactedforinformationonresourcesusedinthetrial.Inthetrial,prism
glasseswerecomparedwithplainglassesandtherewasnodifferenceinpersonneluse.However,for
purposesofcosting,theresourceuseintheinterventionarmwasusedandassumedtobeontopof
usualcare.TheresourceuseandcostsaresummarisedinTable42below.
Table42: Interventioncostsprisminterventionforspatialawarenessand/orvisualneglect
Resources
Frequency
Unitcosts
Costperpatient
(c)
10sessionswithan
occupationaltherapist(a)
30minutespersession
45perhour
225
Prismsglasses(b)
n/a
44.95excludingVAT
44.95
Total
270
(a) Assessmentresourcescouldalsoberequired,suchasneuropsychologicalandfunctionaltests.
234
(b) Prismglassescost:Manufacturerwebsite .Assumedthateachpatientwoulduseonepairofglasses.Ifglassesarereused,costs
wouldbelower.
(c) EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitscostsofhealthandsocialcarereportand
51
Agendaforchangehospitalsalaryband6 (typicalsalarybandidentifiedbyclinicalGDGmembers).
8.1.1.3
Evidencestatements
Clinicalevidencestatements
Onestudy215of36participantsfoundthattherewasnosignificantdifferenceintotalBITscore
betweenthosewhoreceivedcomputerisedscanningandattentiontrainingandthosewhoreceived
usualcareattheendofinterventionperiod(posttreatment)(LOWCONFIDENCEINEFFECT).
Twostudies191,215of52participantsfoundthattherewasnosignificantdifferenceintotalBITscore
betweenthosewhoreceivedcomputerisedscanningandattentiontrainingorrepetitiveprismand
thosewhoreceivedusualcareat6monthsfollowup(LOWCONFIDENCEINEFFECT).
Twostudies80,264of54participantsfoundthattherewasnosignificantdifferenceinBITconventional
scorebetweenthosewhoreceivedfeedbackglassesandthosewhoreceivedusualcareattheendof
interventionperiod(posttreatment)(VERYLOWCONFIDENCEINEFFECT).
Threestudies80,191,264of70participantsfoundthattherewasnosignificantdifferenceinBIT
conventionalscorebetweenthosewhoreceivedfeedbackglassesorrepetitiveprismsandthosewho
receivedusualcareatupto1monthfollowup(LOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
192
StrokeRehabilitation
Cognitivefunctioning
Twostudies80,216of58participantsfoundthattherewasnosignificantdifferenceinBITbehavioural
scorebetweenthosewhoreceivedfeedbackglassesorlimbactivationtreatmentwithperceptual
trainingandthosewhoreceivedusualcare,eitherattheendofinterventionperiod(posttreatment
)(MODERATECONFIDENCEINEFFECT)orat23monthsfollowup(MODERATECONFIDENCEIN
EFFECT).
Onestudy216of40participantsfoundthattherewasnosignificantdifferenceinBITbehavioural
scorebetweenthosewhoreceivedlimbactivationtreatmentwithperceptualtrainingandthosewho
receivedusualcareattheendof6monthsfollowup(VERYLOWCONFIDENCEINEFFECT).
Twostudies191,263of50participantsfoundthattherewasnosignificantdifferenceinlinebisection
scoreandstarcancellationbetweenthosewhoreceivedrepetitiveprismsorrighthalffieldeye
patchingandthosewhoreceivedusualcareattheendofinterventionperiod(posttreatment)(LOW
CONFIDENCEINEFFECT).
Twostudies191,222of55participantsfoundthatwearingrepetitiveprismwasassociatedwitha
statisticallysignificantgreaterimprovementinlinebisection,comparedtothosereceivingusualcare
attheendoffollowup(LOWCONFIDENCEINEFFECT).
Onestudy191of16participantsfoundnosignificantdifferenceinstarcancellationbetweenthose
whowerewearingrepetitiveprismsandthosereceivingusualcareattheendof1monthfollowup
(VERYLOWCONFIDENCEINEFFECT).
Onestudy130of50participantsfoundnosignificantdifferenceinoverallRivermeadPerceptual
AssessmentBattery(RPAB)scorebetweenspatiomotorcueingandusualcareattheendofthetrial
(LOWCONFIDENCEINEFFECT).
Onestudy130of50participantsfoundnosignificantdifferenceinspatiomotorcueingandperceptual
trainingthecancellationsubtestfromRivermeadPerceptualassessmentBattery(RPAB),compared
tousualcareattheendofinterventionperiod(posttreatment)(LOWCONFIDENCEINEFFECT).
Onestudy130of50participantsfoundthatspatiomotorcueingandperceptualtrainingwas
associatedwithastatisticallysignificantgreaterimprovementinthebodyimagesubtestfrom
RivermeadPerceptualAssessmentBattery(RPAB),comparedtousualcareattheendofthe
interventionperiod(posttreatment)(MODERATECONFIDENCEINEFFECT).
Twostudies215,263of70participantsfoundthatcomputerbasedattentiontrainingorrighthalffield
eyepatchingwasassociatedwithastatisticallysignificantgreaterimprovementinlettercancellation,
comparedtousualcareattheendofinterventionperiod(posttreatment)(LOWCONFIDENCEIN
EFFECT).
Onestudy215of36participantsfoundthattherewasnosignificantdifferenceinlettercancellation
betweenparticipantsreceivingcomputerbasedattentiontrainingandthosereceivingusualcareat
theendof6monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy222of39participantsfoundthatprismtrainingwasassociatedwithastatistically
significantgreaterimprovementcomparedtousualcareattheendofinterventionperiod(post
treatment)onthefollowingoutcomes:
Tangentscreenexamination(LOWCONFIDENCEINEFFECT)
Linecancellation(LOWCONFIDENCEINEFFECT).
Twostudies191,263of50participantsfoundthattherewasnosignificantdifferencein
representationaldrawingtestbetweenparticipantswhoreceivedrepetitiveprismstrainingorright
halffieldeyepatchingandthosewhoreceivedusualcareattheendofinterventionperiod(post
treatment)(VERYLOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
193
StrokeRehabilitation
Cognitivefunctioning
Onestudy191of16participantsfoundthattherewasnosignificantdifferenceinrepresentational
drawingtestbetweenthosewhoreceivedrepetitiveprismandusualcareattheendof1month
followup(VERYLOWCONFIDENCEINEFFECT).
Onerandomisedcontrolstudy166comprising29participants,whohadtestedpositiveforvisual
neglect,reportedimprovementsforboththeexperimental(prismaticlensesof5degreesplus
pointingtask)andcontrolgroup(shamlensespluspointingtask).However,participantswearing
prismaticlensesdidnotimprovesignificantlymorethanthecontrolparticipants(VERYLOW
CONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
8.1.2
Recommendationsandlinktoevidence
40.Screenpeopleafterstrokeforcognitivedeficits.Whereacognitive
deficitisidentified,carryoutadetailedassessmentusingvalid,reliable
andresponsivetoolsbeforedesigningatreatmentprogramme.
41.Provideeducationandsupportforpeoplewithstrokeandtheir
familiesandcarerstohelpthemunderstandtheextentandimpactof
cognitivedeficitsafterstroke,recognisingthatthesemayvaryover
timeandindifferentsettings.
42.Assesstheeffectofvisualneglectafterstrokeonfunctionaltaskssuch
asmobility,dressing,eatingandusingawheelchair,usingstandardised
assessmentsandbehaviouralobservation.
43.Useinterventionsforvisualneglectafterstrokethatfocusonthe
relevantfunctionaltasks,takingintoaccounttheunderlying
impairment.Forexample:
interventionstohelppeoplescantotheneglectedside,suchas
brightlycolouredlinesorhighlighterontheedgeofthepage
alertingtechniquessuchasauditorycues
repetitivetaskperformancesuchasdressing
alteringtheperceptualinputusingprismglasses.
Recommendations
Relativevalueplaced
ontheoutcomes
considered
TheGDGconsideredthatinterventionswhichweredesignedtoaddressthe
underlyingimpairmentmightbeevaluatedusingmeasuresoftheextentofthe
impairmentsuchaslinebisectionorcancellationtests.However,theGDGalsofelt
thatitwasimportanttoassesstheimpactofinterventionsonfunctionalactivity,
andthatstudiesshouldreportonfunctionalperformanceaswellasimpairment
levelmeasures.
Qualityofevidence
Alltheincludedstudiesforthisquestionlookedatimprovingvisualneglect.
TheGDGnotedthatallthestudiesweresmallandhadlimitationsintermsofstudy
design.Confidenceintheeffectsshownrangedfrommoderatetoverylowforall
outcomes.Theincludedstudiesuseddifferentinterventionsincludingfeedback
/prismaticglasses,computerisedscanningandattentiontraining,orperceptual
trainingpluslimbactivatingdeviceperceptiontraining,attentionalmotor
integrationandprisms.
NationalClinicalGuidelineCentre,2013.
194
StrokeRehabilitation
Cognitivefunctioning
Somebenefitwasfoundforprismsandcomputerisedscanning191,215,222,263as
measuredbyletterorlinecancellationandlinebisectiontestoutcomes.TheGDG
notedthatuniqueinterventiondeliveredinthestudybyFanthome80andthatthis
hasnotbeenreproducedbyanyotherresearchstudy.
Tradeoffbetween
clinicalbenefitsand
harms
TheGDGagreedthatprismsofferedsmallbenefitsattheimpairmentlevelwithno
evidenceoffunctionalbenefit.Althoughlittleevidenceofclinicallyimportant
benefitwasfound,therewasnoevidenceofharmsassociatedwiththe
interventionseither.TheGDGagreedthatgiventhelimitedevidenceavailableand
thelimitationsofthestudiesarecommendationforassessmentwouldbemore
appropriate.AlthoughnoparticularinterventioncouldberecommendedtheGDG
wereofaviewthatitwasimportanttooffertherapiesthataddressedthe
individualscognitiveimpairmentinordertomaximiseanindividualsabilityto
engageineverydayactivities,andthatthiswasbestdonebyaddressingboth
impairmentsandactivitylimitations,forexamplebyencouragingscanningduring
theperformanceofadressingtask.
Economic
considerations
Nocosteffectivenessstudieswerefoundforthisquestion.Thetypicalcostper
patientfordeliveringaninterventionthataddressesneglectwasestimatedbased
onthestudybyTurtonandcolleagues264at270.TheGDGconsideredthatthe
costofprovidingthisorotherinterventionswaslikelytobeoffsetbythepotential
benefitstopatientsintermsoftheirabilitytoengageineverydayactivities,and
thusimprovedqualityoflife.
Otherconsiderations
TheGDGacknowledgedthatpeopleoftenhavemultipleinteractingcognitive
difficulties.Theresearchtendstofocusonthesedifficultiesinisolationbutinreal
lifetreatmentmodalitiesshouldrecognisethecomplexityoftheindividuals
difficulties.TheGDGconsideredtheresearchpresentedontheindividualcognitive
deficitsbuthavealsomaderecommendationsbasedonthereallifeproblems
patientsexperience.
Identificationofcognitivedeficitsisoftendonebyformalneuropsychometric
screeninginthesestudies.TheGDGagreedtheassessmentofoutcomeis
extremelycomplex,andtheuseofindividualpsychometrictestsasanoutcome
shouldbeusedandinterpretedwithcaution,becausetheyareassessments,while
theoutcomesusedtomeasurecognitiveperformancearealsotypically
multifacetedaddressingattention,memoryandperceptualissues.Analternative
wayofconsideringoutcomeistoconsidergoalachievement,buttheGDGagreed
therearedifferingviewsonwhetherthisisanappropriateoutcometouse.Itwas
acknowledgedthatstandardassessmentsareusedalongwithbehavioural
observationtoassesstheeffectofvisualneglectonusualfunctionalactivities.
TheGDGacknowledgedthestokequalitystandardtoscreenforcognitive
impairment189andagreedthatitwasimportanttomakeageneral
recommendationaboutit.TheGDGalsohighlightedtheneedforhealth
professionalstoprovideinformationandsupporttopatientsandtheircarerson
theimpactthatcognitiveimpairmentmayhave.
TheGDGagreedthatthiswasapotentialtopicforfurtherresearch.
8.2 Memoryfunction
Memoryistheabilitytoencode,storeandretrieveinformation.Memoryproblemsareacommon
cognitivecomplaintfollowingstroke.Memoryrehabilitationprogrammeseitherattempttoretrain
impairedmemoryfunctions,orteachpatientsstrategiestocopewiththem.Factorsthatcan
contributetomemorydifficultiesincludeattentionandexecutivefunction.Inaddition,thepresence
oflowmoodand/orapathyalsoneedstobeassessedasbothoftheseareassociatedwithstrokeand
canpresentwithmemoryproblems.
NationalClinicalGuidelineCentre,2013.
195
StrokeRehabilitation
Cognitivefunctioning
Acomprehensiveassessmentofmemorywillexaminerecognitionandrecallmemoryinverbaland
nonverbaldomainsaswellasnewlearningofinformation.Rememberingtodosomethinginthe
future(prospectivememory)needstobedistinguishedfromrememberinginformationfromthepast
(retrospectivememory).
Differenttypesofmemoryimpairmentimpactonfunctioninvariousways.Forexample,theimpact
ofmemoryimpairmentsmaybeseenasdifficultiesinrememberingrecentinformationsuchasa
therapistsname,thecauseofstroke,orwhenarelativelastvisited.
Difficultywithprospectivememorymayresultinforgettingtoperformtaskssuchastakingtablets,
orpracticinganexerciseprogramme.Bothofthesememorydeficitsimpactonrehabilitation.Other
formsofdeficitsmayimpactmoresignificantlyonfamiliesandcarers.Autobiographicaland
semanticknowledgeaccumulatedduringlifethroughreadingorverbalcommunicationand
experiencesisusuallyrelativelywellpreservedalthoughdetailedexaminationmayrevealpatchy
loss.Impairednonverbalmemorymayresultinpeoplewithstrokebecominglostinparticular
situationssuchaswhentheyareoutinthecommunity.
8.2.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
memorystrategiesversususualcaretoimprovememory
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention:
Comparison:
Usualcare
Outcomes:
Mnemonicstrategiesassociationandorganisation,
drillandpractice,
memoryaidsinternal,
externalorboth,
errorlesslearning.
Interventionshavebeenseparatedintothreegroups:
Compensatorystrategies,RestorativestrategiesandRehearsal
drillandpracticestrategies.
WechslerMemoryScale
Rivermeadbehaviouralmemoryassessment,
CognitiveFailuresQuestionnaire
DysexecutiveQuestionnaire
EverydayMemoryQuestionnaire
8.2.1.1
Clinicalevidence
SearcheswereconductedforsystematicandRCTscomparingtheeffectivenessofmemorystrategies
withusualcareforadultsandyoungpeople16orolderwhohavehadastroke.Onlystudieswitha
minimumsamplesizeof10participants(5ineacharm)andincludingatleast50%ofparticipants
withstrokewereselected.Tworandomisedcontrolledtrials(RCTs)wereidentified.Table43below
summarisesthepopulation,interventionandoutcomesforeachofthestudies.
Table43: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixF.
STUDY
POPULATION
INTERVENTION
Doornhein
Firsttimestroke
Memorytraining:Twicea Pseudotraining:
NationalClinicalGuidelineCentre,2013.
196
COMPARISON
OUTCOMES
Fortarget
StrokeRehabilitation
Cognitivefunctioning
STUDY
199870
POPULATION
patientsadmitted
toarehabilitation
centrewith
cognitive/memory
andsensorymotor
deficits
INTERVENTION
weekfor4weeks.
Mnemonicstrategies
includingassociation
andorganisation.
Homeworkbookswere
alsoused.(N=6)
COMPARISON
Drillandpractice
exercisesincluding
spendingmore
timerepeating
material
(N=6)
OUTCOMES
memorytasks:
NameFace
PairedAssociated
MemoryTest,
StylusMazetest.
ForControl
memorytask:15
WordsTest,
OxfordRecurring
facesTest,
Subjective
Memory
Questionnaire.
Aben2,119
Patientswhohave
hadastrokeif18
monthsormore
hadelapsedsince
theirfirstandonly
stroke.Subjective
memorycomplaints
wereassessed
usingasemi
structured
telephone
interview.Patients
whoreported
memoryproblems
butnevertheless
wereableto
adequatelydeal
withthesedeficits
byusingmemory
aidswereexcluded.
Memoryselfefficacy
trainingtrainingin
memorystrategiesin9
twiceweeklysessions.
Therewere4parts:(1)
informationonmemory
andstroke(2)trainingin
internalandexternal
memorystrategies
(visualisation,diaryuse
andtakingnotes)(3)
psychoeducation(4)
realisticgoalsetting
regardingmemory
demandingtasks.
Peersupport
groupsin9twice
weeklysessionsin
whichgeneral
educationon
causesand
consequencesof
strokewas
provided.
MemorySelf
efficacy(MSE)
Delayedrecall
fromtheauditory
verballearning
task(AVLT)
Delayedrecall
fromthe
Rivermead
Behavioural
MemoryTest
(RBMT)
Qualityoflife
score(EQ5D)
NationalClinicalGuidelineCentre,2013.
197
StrokeRehabilitation
Cognitivefunctioning
Comparison:Cognitiverehabilitation(memorystrategies)forimprovingmemoryversususualcare
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskof
bias
Inconsistency
Indirectness
Imprecisio
n
Memoryself
efficacy
training
mean
unadjusted
changescore
(SE)
Memoryselfefficacyscore(followup10days;Betterindicatedbyhighervalues)
1
Aben
20122
0.48(0.14)
randomised serious(
trials
a)
noserious
inconsistency
noserious
indirectness
serious(b)
DelayedrecallAVLT(followup10days;Betterindicatedbyhighervalues)
1
Aben
20122
randomised serious(
trials
a)
noserious
inconsistency
noserious
indirectness
Control
(peer
support)
mean
unadjusted
change
score(SE)
Effect
0.12(0.12)
0.40
(0.07
to
0.73)
beta0.40higher
(0.07higherto0.73
higher)
Low
1.22(0.29)
0.11
(0.93
to
0.71)
beta0.11lower
(0.93lowerto0.71
higher)
Low
Mean
Differ
ence
(95%
CI)
Baselineadjusted
betavalue*
Confid
ence
(in
effect)
serious(b)
1.01(0.26)
DelayedrecallRBMT(followup10days;Betterindicatedbyhighervalues)
1
Aben
20122
serious(b)
0.01(0.49)
0.97(0.46)
0.63
(2.02
to
0.76)
beta0.63lower
(2.02lowerto0.76
higher)
Low
noserious
imprecisio
n
0.02(0.02)
0.00(0.02)
0.02
(0.04
to
0.08)
beta0.02lower
(0.04lowerto0.08
higher)
Moder
ate
randomised serious(
trials
a)
noserious
inconsistency
noserious
indirectness
QualityofLifeEQ5D(followup10days;Betterindicatedbyhighervalues)
1
Aben
20122
randomised serious(
trials
a)
noserious
inconsistency
noserious
indirectness
NationalClinicalGuidelineCentre,2013.
198
StrokeRehabilitation
Cognitivefunctioning
*Note.Apositive(ornegative)numbermeansthattheinterventiongroupscored(higher(orlower)thanthecontrolgroupatfollowupadjustedforbaseline.Thebetavalueisan
indicatoroftheinfluencethatgroupinghasonthechangefrombaselinethehigherthisvaluethelargerthebetweengroupdifference.
(a)
Thestudywasdowngradedforunclearrandomisationsequencegeneration.
(b)
TheconfidenceintervalcrossesonedefaultMID(0.5ofStandardmeandifference)
Narrativesummary
ThefollowingstudyissummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Oneunblindedstudy70of12patientsreportedthatmnemonicstrategytreatmentshowedasignificantimprovementinthetrainedmemoryskills,but
therewasnoimprovementoncontrolmemorytasks.Subjectiveratingsofeverydaymemoryfunctioningdidnotdifferbetweenthetwogroups.
NationalClinicalGuidelineCentre,2013.
199
StrokeRehabilitation
Cognitivefunctioning
8.2.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingcognitiverehabilitationmemorystrategieswithusual
caretoimprovememorywereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Inpracticemostcognitiverehabilitationwouldbebasedoncompensatorystrategiesor
environmentalmanipulation,andnottheinterventionswithinthetrialsconsidered.Typicalcostsof
deliveringaninterventionaimedatimprovingmemoryinpatientswhohavehadastrokewas
thereforeestimatedbasedonresourceuseestimatesprovidedbyclinicalmembersoftheGDG.
ThesecostsaresummarisedinTable28. Inaddition,ifcomputerprogramsareused,additionalcosts
wouldbeincurred.
Table44: Interventioncostscognitiverehabilitationformemory
Resources
Frequency
Unitcosts(a)
Costperpatient
Initialassessmentbya
psychologist
2hours
136perhour
272
Goalsettingwithmulti
disciplinaryteam
1hour,with15
minutesallocatedto
memorygoals
136perhourpsychologist
35perhournurse
45perhourphysiotherapist
45perhouroccupational
therapist
132perhourmedical
consultant
98
Interventionifinpatient:
45minutesper
136perhourpsychologist
occupationaltherapistand session,twiceaweek 45perhouroccupational
psychologistsessions
for6weeks
therapist
1629
Interventionifinthe
community:occupational
therapistandpsychologist
sessions
45minutesper
session,onceaweek
for6weeks
136perhourpsychologist
45perhouroccupational
therapist
815
Totalpersonnelcost
(incrementaloverusual
care)
Inpatient:1999
Community:1184
a)EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcarereportandthe
51
followingAgendaforChangesalarybandspsychologist(band8),physiotherapistandoccupationaltherapist(band6),nurse(band5)
(typicalsalarybandsidentifiedbyclinicalGDGmembers).
NationalClinicalGuidelineCentre,2013.
200
StrokeRehabilitation
Cognitivefunctioning
8.2.1.3
Evidencestatements
Clinicalevidencestatements
Onestudy2of153participantsfoundthattherewasnosignificantdifferenceindelayedrecallAVLT
betweentheparticipantswhoreceivedmemoryselfefficiencytrainingandthosewhoreceivedusual
care(peersupport)(LOWCONFIDENCEINEFFECT).
Onestudy2of153participantsfoundthattherewasnosignificantdifferenceindelayedrecallRBMT
betweentheparticipantswhoreceivedmemoryselfefficiencytrainingandthosewhoreceivedusual
care(peersupport)(LOWCONFIDENCEINEFFECT).
Onestudy2of153participantsfoundthattherewasnosignificantdifferenceinQualityofLife
(EQ5D)betweentheparticipantswhoreceivedmemoryselfefficiencytrainingandthosewho
receivedusualcare(peersupport)(MODERATECONFIDENCEINEFFECT).
Onestudy2of153participantsfoundthatasignificantimprovementininmemoryselfefficacy
scoresbetweentheparticipantswhoreceivedmemoryselfefficacytrainingandthosewhoreceived
usualcare(peersupport)(LOWCONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
8.2.2
Recommendationsandlinktoevidence
44.Assessmemoryandotherrelevantdomainsofcognitivefunctioning
(suchasexecutivefunctions)inpeopleafterstroke,particularlywhere
impairmentsinmemoryaffecteverydayactivity.
45.Useinterventionsformemoryandcognitivefunctionsafterstrokethat
focusontherelevantfunctionaltasks,takingintoaccountthe
underlyingimpairment.Interventionscouldinclude:
increasingawarenessofthememorydeficit
enhancinglearningusingerrorlesslearningandelaborative
techniques(makingassociations,useofmnemonics,internal
strategiesrelatedtoencodinginformationsuchaspreview,
question,read,state,test)
externalaids(forexample,diaries,lists,calendarsandalarms)
environmentalstrategies(routinesandenvironmentalprompts).
Recommendations:
Relativevalue
placedonthe
outcomes
considered
TheGDGconsideredthatrecallinginformationinthememoryofstrokepatients
afteradelaywasthemostimportantoutcomeforthisrecommendation.Theyalso
thoughtthatbeingabletoreflectbackonthingsthathappenedpreviouslywould
benefitgeneralwellbeingandthereforepositivelyaffectqualityoflifewhichwas
anotherreportedoutcome.
Tradeoffbetween
clinicalbenefitsand
harms
TheGDGagreedthatrehabilitationisaboutacquiringskillsregardlessofthetime
periodbetweentheonsetofstrokeandintroductionofanintervention.Memory
problemsmayhavelongtermimpactonavarietyoftasks,soassessmentsshould
reflectthisandinterventionsneedtobetailoredanddeliveredaccordingly.
NationalClinicalGuidelineCentre,2013.
201
StrokeRehabilitation
Cognitivefunctioning
TheGDGnotedthatmemoryselfefficiencytraining(includingtrainingonstrategies
toaidretentionofinformation)didnotprovideconclusiveevidenceforageneral
memoryimprovement,whichconflictedwithexperiencefromclinicalpractice,
Economic
considerations
Nocosteffectivenessstudieswerefound.Personnelcostfordeliveringamemory
interventionprogrammewasestimatedat1999(inpatient)/1184(community)
basedonGDGestimatesoftheresourceuseinvolved.TheGDGconsideredthatvery
fewrehabilitationunitswouldhavecomputersoftwareavailablecurrently;therefore
thesewouldincuradditionalcosts.TheGDGconsideredthattheadditionalcosts
wouldpotentiallybeoffsetbythelongtermbenefittopatientsintermsofimproved
qualityoflife.
Qualityofevidence
TheGDGnotedthatoneofthetwostudies70consideredwasverysmallandhad
limitationsintermsofstudydesignandimprecisionaroundtheestimateofeffect.
Theotherstudy2wasmethodologicallybetterconductedandincludedovera
hundredparticipantswhohavehadastroke.However,itusedaparticular
frameworkwiththeaimtoincreasememoryefficiencyratherthanmemorycapacity
orabilitytousememoryineverydaysituations.
TheDoornhein70studyfoundthatteachingmnemonicstrategiesofassociationand
organisationwaslinkedtoimprovedperformanceinspecifictrainedmemorytasks,
butdidnottransfertoothertasks.TheAbenstudy(2012)2didfindanimprovement
inmemoryefficiency,butnogeneralimprovementindelayedrecall.Sincethe
interventionwasmemoryselfefficiencytrainingtheGDGfeltthatanimprovement
inthisabilityonitsownwasnotaveryconvincingresult.TheGDGconsideredthat
thetypeofmemorydomainsaddressedinthestudiesdidnotaddresstherangeof
memorydifficultiesthatmaybefacedbypatients.Rotelearninganddelayedrecallis
notnecessarilydirectlytranslatableintoimprovementsindailyfunctionalabilities.
Otherconsiderations TheGDGconsideredwellestablishedresearchonsimilarmemoryproblemsinother
neurologicalconditions,andinthesestudiesitwasfoundthatpatientsdobenefit
fromtheuseofsomecompensatorystrategies,suchastheuseofmnemonics,
diaries,lists,alarmsandemployingenvironmentalpromptsorfollowingacertain
routinetohelpwithmemorydeficit.Similarstrategiesshouldbetaughttopeople
whohavehadastrokewhereappropriate.Itisimportantinthisrespectthatthe
strategiesareadaptedtotheindividualslearningstyleandparticularimpairment
ratherthanhavingonegeneraltrainingscheduletofitall.
TheGDGagreedthatfurtherresearchisrequired.Thegroupagreedthatmemory
needstobeassessedandwherememoryimpactsoneverydayactivityinterventions
shouldbetargetedatthatactivity,takingintoaccounttheunderlyingmemory
problems.TheGDGnotedthatthesuccessofotherrehabilitativeinterventionsmay
becontingentonmemoryandthereforetheimpactofmemoryonfunctionis
importantandshouldnotbeunderestimated.
8.3 Attentionfunction
Attentionproblemscanoccurfollowingstrokeandarecommoninpeoplewithdamagetotheright
sideoftheirbrain.Itisbestdescribedasthesustainedfocusonsalientinformationwhilefilteringor
ignoringextraneousinformation.Attentionisaverybasicfunctionthatoftenisaprecursortoall
otherneurological/cognitivefunctions.Fivedifferenttypesofattentionhavebeendescribed
Focusedattention:Theabilitytoresponddiscretelytospecificvisual,auditoryortactile
stimuli.
Sustainedattention:Theabilitytomaintainaconsistentbehaviouralresponseduring
continuousandrepetitiveactivity.
Selectiveattention:Theabilitytomaintainabehaviouralorcognitivesetinthefaceof
distractingorcompetingstimuli.
NationalClinicalGuidelineCentre,2013.
202
StrokeRehabilitation
Cognitivefunctioning
Alternatingattention:Theabilityofmentalflexibilitythatallowsindividualstoshifttheir
focusofattentionandmovebetweentaskshavingdifferentcognitiverequirements.
Dividedattention:Thisisthehighestlevelofattentionanditreferstotheabilitytorespond
simultaneouslytomultipletasksormultipletaskdemands.
Althoughthereissomespontaneousrecoveryofattentioninsomepatients,somesymptomsmay
persistforyears.Cognitiverehabilitationtrainingaimsatmanagingdifferentaspectsofattentionand
canimprovepeople'sabilitytoparticipateindailyactivity.
Workingmemoryandattentionarecloselyrelated.Workingmemoryisessentialindetermining
whereattentionshouldbedirected,filteringinformationandtheabilitytoinhibitcompetingstimuli;
thiscanbedescribedascontrolofattention.
peopleafterstrokewhatistheclinicalandcosteffectivenessofsustainedattentiontrainingversus
usualcaretoimproveattention?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention
Computerisedtrainingprogrammeusingreactiontimesand
patternrecognition.
Comparison
Usualcare
Outcomes
Testofeverydayattention,
CognitivefailuresQuestionnaire
DisexecutiveQuestionnaire
EverydayMemoryQuestionnaire
8.3.1.1
Clinicalevidence
SearcheswereconductedforsystematicreviewsorRCTsthatcomparedsustainedattentiontraining
versususualcaretoimproveattentioninadultsoryoungpeople16orolderwhohavehadastroke.
Onlystudieswithaminimumsamplesizeof10participants(5ineacharm)andincludingatleast
50%ofparticipantswithstrokewereselected.TwoRCTswereidentified.Table45belowsummarises
thepopulation,interventionandoutcomesfortheincludedstudies.
Table45: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixF.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Barker
Collo
200916
Acutestroke
survivorsadmitted
toNewZealand
hospitalswho
experiencedan
attentiondeficit
within2weekspost
stroke
Attentionprocesstraining
(APT):sustained,selective,
alternating,anddivided
attentiontraining(forexample
numbercancellationwith
visualdistractor,sustained
attentioninnoiseusingaudio
CDs,flexibleshape
cancellation,setdependent
alternatingattentiontasks)
administeredbyaregistered
clinicalneuropsychologist.
Standardcare
(notspecifiedin
thepaper).
(N=40)
IntegratedVisual
Auditory
Continuous
Performancetest
(IVACPT)
Fullattention,
Auditory
attention,
Visual
attention.
NationalClinicalGuidelineCentre,2013.
203
StrokeRehabilitation
Cognitivefunctioning
STUDY
POPULATION
INTERVENTION
Participantsreceivedupto30
hoursofindividualAPTfor1
houronweekdaysfor4weeks
(mean13.5hours).(N=38)
COMPARISON
OUTCOMES
Westerber
g2007284
Participantsaged
3465ofvocational
activitywhohad
experiencedstroke
1236monthsago
andhadself
reporteddeficitsin
attention.
Computerisedworking
memorytraining:was
implementedwithacomputer
softwareproductusedat
homeforabout40minutes
/day,5days/weekfor5
weeks.Tasksinvolved
reproducingalightsequence
inavisuospatialgrid,
indicatingnumbersinreverse
order,identifyingletter
positionsinasequence,
identifyingalettersequencein
pseudowords,finding
mismatchedletters,etc.
Participantsreportedtheir
dailyresultsviainternettoa
serveratthehospital.
Feedbackfromapsychologist
providedviatelephoneoncea
week.
(N=9)
Usualcare:no
memorytraining
andnocontact
witha
psychologist.
(N=9)
WechslerAdult
Intelligence
Scale:
Spanboard
(measures
visuospatial
WM),
Digitspan
(measures
auditoryWM)
Strooptime
(sec)
NationalClinicalGuidelineCentre,2013.
204
Stroopraw
score
Cognitive
failure
questionnaire
scores
StrokeRehabilitation
Cognitivefunctioning
Comparison:Cognitiverehabilitation(Sustainedattentiontraining)versususualcare
Table46: SustainedattentiontrainingversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
No.of
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Sustained
attention
training
Imprecision Mean(SD)
Usual
care
Mean
(SD)
Mean
differenc
e(95%
CI)
Mean
Differen
ce(MD)
(95%CI)
(a)
(a)
2.76
(1.31,
4.21)
MD2.76
higher
(1.31to
4.21
higher)
High
(a)
(a)
2.49
(1.24,
3.74)
MD2.49
higher
(1.24to
3.74
higher)
High
(a)
(a)
1.96
(0.49,
3.43)
MD1.96
higher
(0.49to
3.43
higher)
High
(a)
(a)
0.83(
0.46,
2.12)
MD0.83
higher
(0.46
lowerto
Moderate
Confidence
(ineffect)
IVACPT(fullattention)changes(5weeksfollowup)(Betterindicatedbyhighervalues)
1
RCTsingle
16
Barker2009 blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
IVACPT(fullattention)changes(6monthsfollowup)(Betterindicatedbyhighervalues)
1
RCTsingle
Barker200916 blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
IVACPT(auditoryattention)changes(5weeksfollowup)(Betterindicatedbyhighervalues)
1
RCTsingle
Barker200916 blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
IVACPT(auditoryattention)changes(6monthsfollowup)(Betterindicatedbyhighervalues)
1
RCTsingle
16
Barker2009 blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
NationalClinicalGuidelineCentre,2013.
205
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
No.of
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Sustained
attention
training
Imprecision Mean(SD)
Usual
care
Mean
(SD)
Mean
differenc
e(95%
CI)
Mean
Differen
ce(MD)
(95%CI)
2.12
higher)
(a)
(a)
1.56
(0.03,
3.09)
MD1.56
higher
(0.03to
3.09
higher)
High
(a)
(a)
1.41
(0.04,
2.78)
MD1.41
higher
(0.04to
2.78
higher)
High
Confidence
(ineffect)
IVACPT(visualattention)changes(5weeksfollowup)(Betterindicatedbyhighervalues)
1
RCTsingle
16
Barker2009 blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
IVACPT(visualattention)changes(6monthsfollowup)(Betterindicatedbyhighervalues)
1
RCTsingle
16
Barker2009 blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
(a)
Mean(SD)changesarenotgiveninthestudybygrouponlymeandifferenceswerereported.
ConfidenceintervalcrossedoneendofdefaultMID.
(b)
NationalClinicalGuidelineCentre,2013.
206
StrokeRehabilitation
Cognitivefunctioning
Table47:ComputerizedworkingmemorytrainingversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Computerise
dworking
memory
training
Mean(SD)
Effect
Usual
care
Mean
(SD)
Mean
difference
(95%CI)
Mean
differenc
e(MD)
(95%CI)
Confidence(
ineffect)
WechslerAdultintelligenceScaleRevisedSpanboard(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Westerberg
2007284
RCT
unclear
blinding
Veryserious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
6.2(1.0)
5.7
(1.8)
0.50(0.85
to1.85)
MD0.50
higher
(0.85low
erto
1.85
higher)
Verylow
WechslerAdultIntelligenceScaleRevisedDigitSpan(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Westerberg
2007284
RCT
unclear
blinding
Veryserious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
7.3(1.0)
5.7
(1.3)
1.60(0.53
to2.67)
MD1.60
higher
(0.53to
2.67
higher)
Low
Strooptime(sec)(posttreatmenteffect)(Betterindicatedbylowervalues)
1
Westerberg
2007284
RCT
unclear
blinding
Veryserious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
93(19)
124
(48)
31(64.73
to2.73)
MD31
lower
(64.73
lowerto
2.73
higher)
Verylow
Serious
91.1(1.27)
97.8
1.30(0.47
MD1.30 Verylow
Strooprawscore(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Veryserious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
207
StrokeRehabilitation
Cognitivefunctioning
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Computerise
dworking
memory
training
Mean(SD)
Effect
Usual
care
Mean
(SD)
WechslerAdultintelligenceScaleRevisedSpanboard(posttreatmenteffect)(Betterindicatedbyhighervalues)
unclear limitations
inconsistency
indirectness
imprecision
(2.4)
Westerberg
blinding (a)
(b)
2007284
Mean
difference
(95%CI)
Mean
differenc
e(MD)
(95%CI)
Confidence(
ineffect)
to3.07)
lower
(0.47
lowerto
3.07
higher)
13.8(
25.79to
1.81)
MD13.8 Verylow
lower
(25.79
lowerto
1.81
lower)
CognitiveFailureQuestionnaire(CFQscalerangingfrom0100,posttreatmenteffect)(betterindicatedbylowervalues)
1
Westerberg
2007284
RCT
unclear
blinding
Veryserious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
29.2(12.1)
43
(13.8)
(a)
Nodetailsonrandomisation.Unclearallocationconcealmentandblinding.
ConfidenceintervalcrossedoneendofdefaultMID(0.5ofthestandardmeandifference).
(b)
NationalClinicalGuidelineCentre,2013.
208
StrokeRehabilitation
Cognitivefunctioning
8.3.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingcognitiverehabilitationsustainedattentiontraining
withusualcaretoimproveattentionwereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
TheGDGadvisedtoestimateinterventioncostsbasedontheresourcesdescribedinBarker,200916
TheestimatedcostofthesoftwaretoperformanunlimitednumberofIntegratedVisualAuditory
ContinuousPerformanceTests(IVACPT)was1244kexcludingVAT(obtainedfromwww.bio
medical.com26).Personnelcosts,incrementaloverusualcare,areoutlinedinTable48.
Table48: InterventioncostspersonnelcostsassociatedwithIVACPT
Resources
Frequency
Unitcosts
Costperpatient
(b)
Baseline
neuropsychological
assessment(a)
2.5hours
repeatedat5
weeksand6
months
136perhour
1,020
IndividualAttention
ProcessTraining
(APT)sessions(a)
30hours
136perhour(b)
4,080
Totalpersonnelcost
(incrementalover
usualcare)
5,100
(a) Deliveredbyaneuropsychologist
(b) Clinicalpsychologistcostsusedascostsforaneuropsychologistcouldnotbeobtained.Estimatedbasedondataandmethodsfrom
51
PersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcarereportandAgendaforChangesalaryband8 (typical
salarybandidentifiedbyclinicalGDGmembers).
8.3.1.3
Evidencestatements
Clinicalevidencestatements
Onestudy16of78participantsfoundthosewhoreceivedthesustainedattentiontraining
experiencedastatisticallysignificantimprovementinfullattentionmeasuredbytheIntegrated
VisualAuditoryContinuousPerformancetest(IVACPT)at5weeksand6monthsfollowup
comparedtothosewhoreceivedusualcare(HIGHCONFIDENCEINEFFECT).
Onestudy16of78participantsfoundthosewhoreceivedthesustainedattentiontraining
experiencedastatisticallysignificantimprovementinauditoryattentionmeasuredbytheIntegrated
VisualAuditoryContinuousPerformancetest(IVACPT)at5weeksfollowupcomparedtothosewho
receivedusualcare(HIGHCONFIDENCEINEFFECT).
k US$1895(2011)convertedtoUKpounds(2010)usingpurchasingpowerparities194
NationalClinicalGuidelineCentre,2013.
209
StrokeRehabilitation
Cognitivefunctioning
Onestudy16of78participantsfoundtherewasnosignificantdifferenceontheauditoryattention
measuredbytheIntegratedVisualAuditoryContinuousPerformancetest(IVACPT)at6months
betweenparticipantswhoreceivedthesustainedattentiontrainingandthosewhoreceivedusual
care(MODERATECONFIDENCEINEFFECT).
Onestudy16of78participantsfoundthosewhoreceivedthesustainedattentiontraining
experiencedastatisticallysignificantimprovementinvisualattentionmeasuredbytheIntegrated
VisualAuditoryContinuousPerformancetest(IVACPT)at5weeksand6monthsfollowup
comparedtothosewhoreceivedusualcare(HIGHCONFIDENCEINEFFECT).
Onestudy284of18participantsfoundthattherewasnosignificantdifferenceinWechslerAdult
intelligenceScaleRevisedWAISRSpanBoardtestbetweentheparticipantswhoreceived
computerisedmemorytrainingandthosewhoreceivedusualcare(VERYLOWCONFIDENCEIN
EFFECT).
Onestudy284of18participantsfoundthatthosewhoreceivedcomputerisedmemorytraininghada
statisticallysignificantimprovementinWechslerAdultintelligenceScaleRevisedWAISRdigitspan
test,comparedwiththeparticipantswhoreceivedusualcare.(LOWCONFIDENCEINEFFECT)
Onestudy284of18participantsfoundthattherewasnosignificantdifferencethetimetakento
completetheSTROOPtask(sec)betweentheparticipantswhoreceivedcomputerisedmemory
trainingandthosewhoreceivedusualcare(VERYLOWCONFIDENCEINEFFECT).
Onestudy284of18participantsfoundthattherewasnosignificantdifferencenumberofSTROOP
itemscorrectlynamedbetweentheparticipantswhoreceivedcomputerisedmemorytrainingand
thosewhoreceivedusualcare(VERYLOWCONFIDENCEINEFFECT).
Onestudy284of18participantsfoundthatthosewhoreceivedcomputerisedmemorytraininghada
statisticallysignificantimprovementinthescoreofaCognitiveFailureQuestionnaire(CFQ),
comparedwiththeparticipantswhoreceivedusualcare.(VERYLOWCONFIDENCEINEFFECT)
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
8.3.2
Recommendationsandlinktoevidence
46.Assessattentionandcognitivefunctionsinpeopleafterstrokeusing
standardisedassessments.Usebehaviouralobservationtoevaluate
theimpactoftheimpairmentonfunctionaltasks.
47.Considerattentiontrainingforpeoplewithattentiondeficitsafter
stroke.
48.Useinterventionsforattentionandcognitivefunctionsafterstroke
thatfocusontherelevantfunctionaltasks.Forexample,usegeneric
techniquessuchasmanagingtheenvironmentandprovidingprompts
relevanttothefunctionaltask.
Recommendations
Relativevalueplaced
ontheoutcomes
considered
TheoutcomesincludedinthereviewweretheIntegratedVisualAuditory
ContinuousPerformancetest(IVACPT):fullattention,auditoryattention,visual
attentionandWechslerAdultIntelligenceScaleofdigitspanaswellas
performanceontheSTROOPtaskandscoresonthecognitivefailurequestionnaire.
NationalClinicalGuidelineCentre,2013.
210
StrokeRehabilitation
Cognitivefunctioning
Thetestsusedwereneuropsychologicalmeasuresofattentionandeventhough
thecognitivefailuresquestionnaireisaimedattestingmorefunctionalabilities
(suchaseverydaysituationsattendingtonames,andfocusingontasks)thevalidity
ofthismeasurewasquestioned.TheSTROOPtaskwasseenasagoodmeasureof
attentionalcapabilitiessinceitrequiresparticipantstofocusonparticularfeatures
whilstdisregardingotheraspects.Thisisanabilitythatcanbedirectlytranslatedto
morefunctionalperformancesuchasfocusingonataskinlightofother
distractions
Tradeoffbetween
clinicalbenefitsand
harms
AttentionimpairmentwasdefinedinthestudybyBarker16asperformanceless
than1standarddeviationbelowthenormativemeanonanytest.Inthesecond
includedstudy284attention/memoryproblemswerebasedonselfreportonlyand
thestudymethodologywaspoorlydescribed.ItwasagreedbytheGDGthat
assessmentsofcognitiveimpairmentsmaynotbeclinicallyrelevant,andattention
basedinterventionsshouldbeprovidedwhenthepersonwithstrokeortheir
carersidentifydifficultiesattributabletoattentiondifficulties.Thegroupagreed
thatarecommendationforspecificinterventionscouldnotbemadebasedon
thesetwostudies,butrecognisedthatinclinicalpracticetimewouldbespentwith
patientstoimproveattentiondeficits.Furtherresearchisrequired.
Economic
considerations
Nocosteffectivenessstudieswerefound.ThecostoftheIVACPTsoftwareto
delivertheinterventionusedinthestudyidentifiedfortheclinicalreviewwas
estimatedat1244.Inaddition,personnelcostswereestimatedat5,100per
personbasedonresourcesusedintheBarker,200916study.Giventhehighcostof
thisspecificinterventionandthelimitedevidenceofitsclinicaleffectiveness,the
evidencewasconsideredinsufficienttoconcludethatitwouldbecosteffective.
Qualityofevidence
Onewellconductedrandomisedcontrolledtrial16foundthatattentionprocess
trainingwasassociatedwithgreaterimprovementinattentionasmeasuredonthe
IVACPTvisualscaleandFullScaleAttentionQuotientat5weeksand6months
followup.
TheWesterberg284studyfoundthatacomputerisedworkingmemorytraining
programmeimprovedworkingmemorywhenmeasuredwiththeWechslerAdult
IntelligenceScaleRevisedDigitSpan(anauditorytestofworkingmemory),but
hadnoeffectwhenmeasuredbytheWechslerAdultIntelligenceScaleRevised
Spanboard(avisuospatialtestofworkingmemory).Bothofthesetestsrequire
attentionalcognitiveresources.Itwasnotedthatthetimescaleaddressedinthe
Westerbergstudyis20monthsaftertheonsetofstrokeandtheimpactofthe
interventionmaydifferifundertakeninthesubacutestage.Noimprovementwas
seenonSTROOPperformance.However,theinterventionresultedinfewer
cognitivefailuresasmeasuredbyselfreport.Thequalityoftheevidenceforthese
outcomeswaslowtoverylowandthereforewedonothaveconfidenceinthe
effectsreportedbythisstudy.
TheGDGagreedthatusualcarewouldnormallyconsistofabaselineassessment
(similartotheonereportedbyBarker,2009)butofshorterweekday
neuropsychologicalsessionsthanthosereportedinthestudywhichwere
describedasonehourperweekdayforfourweeks.TheGDGnotedthatall
interventionshaveabaselineassessmentbutthecontentoftheassessmentvaries.
TheGDGfeltthatcurrentlycomputerbasedrehabilitationmaynotbeavailable
everywhere,howeverwithincreasingfamiliarityandaccessofthepopulationto
personalcomputersthiswouldchange.Thelimitedevidencefoundindicatesthe
needforfurtherresearchtobeundertaken.
Otherconsiderations
Eventhoughevidencefromonestudycanberatedashigh,theGDGconsidered
thatthiswouldneedtobereplicatedtodemonstratearobusteffect.However,
problemswithattentionhaveseriouseffectsonselfesteemandqualityoflifein
generalandarequiteupsettingforpersonswhohavehadastrokeandtheircarers
/family.Inparticularitaffectsfunctionalperformanceineverydaysituations,such
NationalClinicalGuidelineCentre,2013.
211
StrokeRehabilitation
Cognitivefunctioning
asthefocusontasksthatareimportantwhilstignoringdistractionfromother
environmentalfactors.Itwasthereforeseenasimportanttoassesspossible
impairmentsandadoptanindividualisedapproachtohelpthepersonto
participatemoreconfidentlyinactivitiesofdailyliving.
NationalClinicalGuidelineCentre,2013.
212
StrokeRehabilitation
Emotionalfunctioning
9 Emotionalfunctioning
9.1 Psychologicaltherapies
Psychologicaltherapiesmaybecharacterisedasanapproachwhichinvolvesaconfidingrelationship,
thattakesplacewithinatherapeuticsetting,withatheoreticalbasis(anunderstandingofmodelsof
normalandabnormalbehaviour)involvingatherapeuticprocesswhichagainhasanunderpinning
theoreticalmodel.Therapycanbedeliveredtoanindividual,acouple,afamilyoragroup.
Manypeoplewhohavehadastrokeexperiencedistresswhichcanimpactnegativelyonfunctional
outcome.Inaddition,notonlyarethephysicalconsequencesofstrokeassociatedwithemotional
disorders,thecognitiveaspectsofstrokemayalsoimpactontheirabilitytodealwiththeemotional
consequencesofthestroke.
Psychologicaltherapiesmaybeusefulforindividualswithstroke.Theseinterventionsemphasisethe
individual'sownresidualstrengths,clarifythepatient'sconcernsandteachnewstrategiesforcoping
effectivelyandmanagingdistress.Itisoftenusefultodrawuponavarietyofpsychologicalmodels
(forexampleBehaviourtherapy,cognitivebehaviourtherapyandalsotheoriesofchange)depending
ontheindividual'spresentation.
Psychologicaltherapiesmayhelptheindividualandtheircarerswithpoststrokeemotionaldisorders
andrelationshipissues.Psychologicalinterventionsofthistypemayalsobeneededtofacilitatean
individualandcarersunderstandingandadjustmenttocognitiveimpairments,communication
impairmentsortophysicaldisabilities.Itiscriticaltonotethattheimpactofphysical,cognitiveand
emotionaldifficultiesarelikelytooverlap;thereforethedeliveryofanystandardintervention(for
example,cognitivebehaviourtherapy)islikelytoneedadaptationtosuitanindividualscognitive
and/orphysicalpresentation.WithintheNHS,psychologicaltherapiesareprovidedbymembersof
differentprofessionaldisciplines,includingclinicalneuropsychologists,clinicalpsychologists,
speciallytrainedmentalhealthnurses,occupationaltherapistsandcounsellors.Inthecontextof
stroke,wherepatientshaveimpairmentswhichimpactontheirabilitytoparticipateinpsychological
treatments,itisimportantthatthetherapistunderstandsthenatureandimpactoftheimpairments
andhowtheyinteract.
9.1.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
psychologicaltherapiesprovidedtothefamily(includingthepatient)?
ClinicalMethodologicalIntroduction
Population:
Familycarers(familymemberorrelative,orotherunpaidcarer
support)ofpeoplewithstroketoincludeadultsandyoung
peopleover16withstroke
Intervention
FamilyTherapy
CognitiveBehaviourTherapy
Relationshipcounselling(toincludeCouplestherapy)
(allinterventionsmayincludesomeformofinformation)
Comparison:
Usualcare(usuallynothing)
Outcomes:
QualityofLife(forbothcarerandpatient)
AnyQOLanddepressionoutcomesincludingthefollowing:
strokeimpactscale,EuroQoL,caregiverburdenscale,caregiver
strainindex,carerstrainindex,burdenofstrokescale,Stroke
andaphasiaqualityoflifescale,ASCOTscale.
Occurrenceofdepression/anxiety/moodincarers
NationalClinicalGuidelineCentre,2013.
213
StrokeRehabilitation
Emotionalfunctioning
ClinicalMethodologicalIntroduction
9.1.1.1
BeckDepressionInventory,BeckDepressionInventory2,
GeriatricDepressionScale,neuropsychiatryinventory,Hospital
AnxietyandDepressionScale(HADS),Generalhealth
questionnaire,VisualAnalogueMoodScale,SADQ.
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTsorsystematicreviewsofobservational
studiescomparingpsychologicaltherapieswithusualcaretoimprovequalityoflifeforbothcarer
andstrokepatientsolderthan16yearsold.OneRCTwasidentifiedthatmettheprespecified
protocol.Table40summarisesthestudycharacteristicsoftheincludedstudy.
Table49: Summaryofthestudyincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
242
Smith2012
POPULATION
INTERVENTION
COMPARISON
Participantswere
marriedcouples:female
caregiver(CGs)andher
husbandwhohashada
stroke(PWS).Eitherthe
caregiverorher
husbandhadtoscore
fiveormoreonthe
PersonalHealth
Questionnaire(PHQ9:
atleastmild
depression).
Participantswere
excludediftheywere
medicallyunstableor
terminallyillandifthey
werecognitivelyunable
toparticipate.
Theinterventionwas
basedontheStress
ProcessModel.It
consistedofan
onlinesupport
programoffive
components
designedtoprovide
thecaregiverswith
knowledgeresources
andskillstohelp
boththemselvesand
theirpartnerto
reducetheir
personaldistressand
toprovideoptimal
emotionalcareto
thePWS.Attempts
wererepeatedly
madeto
acknowledgethe
positiveandnegative
feelingsofboth
membersoftheCG
PWSdyad,aswellas
toillustratehow
theywere
intertwined.CGs
wereencouragedto
interactwithPWSin
waystoenhance
theirmutualwell
being.
CGPWSdyads Centrefor
hadaccesstothe
Epidemiologic
onlineresource
alStudies
centre,buthad
Depression
noexposureto
scale(CESD)
thekey
MasteryScale
intervention
(ameasureto
components.
assesscoping
ability)
SelfEsteem
scale,
Medical
Outcomes
Study(MOS)
SocialSupport
survey
(measuring
amountof
emotional,
informational
and
affectionate
support).
NationalClinicalGuidelineCentre,2013.
214
OUTCOMES
StrokeRehabilitation
Emotionalfunctioning
Comparisonpsychologicaltherapiesvs.controltoimprovemoodincaregiversandpersonswithstroke
Table50: Webbasedpsychologicaltherapyvs.controloutcomesforCAREGIVERS(WIVES)only(baselineadjustedmeans(sd))
SummaryofFindings
Qualityassessment
Effect
Noof
studies
Relati
ve
Risk/
Mean
differe
nce/
(95%
CI)
Design
Riskof
bias
Inconsistency
Indirectness
Imprecisio
n
Webbased
psychological
intervention
Frequency(%)/mean
(SD)/median(range)
Control
Frequency(%)/
mean(SD)/
median(range)
Absoluteeffector
meandifference
(MD)
(95%CI)
Confid
ence
(in
effect)
CentreforEpidemiologicalStudiesDepressionscale(CESD)Posttest(followup11weeks;rangeofscores:060;Betterindicatedbylowervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
13.9(7.7)
19.7(7.4)
5.8(
11.07
to
0.53)
MD5.8lower
(11.07to0.53
lower)
LOW
CentreforEpidemiologicalStudiesDepressionscale(CESD)1monthfollowup(rangeofscores:060;Betterindicatedbylowervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
13.4(6.2)
16.6(6.2)
3.2(
7.5to
1.1)
MD3.2lower(7.5
lowerto1.1
higher)
LOW
23.6(2.5)
0.6(
1.21
to
2.41)
MD0.6higher
(1.21lowerto
2.41higher)
LOW
24.4(2.1)
0.3(
1.69
to
MD0.3lower
(1.69lowerto
1.09higher)
VERY
LOW
MasteryScalePosttest(followup11weeks;rangeofscores:936;Betterindicatedbyhighervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
24.2(2.7)
MasteryScale1monthfollowup(rangeofscores:936;Betterindicatedbyhighervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(b)
24.1(1.9)
NationalClinicalGuidelineCentre,2013.
215
StrokeRehabilitation
Emotionalfunctioning
SummaryofFindings
Qualityassessment
Noof
studies
Design
Effect
Riskof
bias
Inconsistency
Indirectness
Imprecisio
n
Webbased
psychological
intervention
Frequency(%)/mean
(SD)/median(range)
Control
Frequency(%)/
mean(SD)/
median(range)
Relati
ve
Risk/
Mean
differe
nce/
(95%
CI)
1.09)
Absoluteeffector
meandifference
(MD)
(95%CI)
Confid
ence
(in
effect)
SelfesteemscalePosttest(followup11weeks;rangeofscores:1040;Betterindicatedbyhighervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(c)
31.6(2.3)
31.9(2.5)
0.3(
1.96
to
1.36)
MD0.3lower
(1.96lowerto
1.36higher)
VERY
LOW
32.6(2.9)
1.5(
3.44
to
0.44)
MD1.5lower
(3.44lowerto
0.44higher)
LOW
Selfesteemscale1monthfollowup(rangeofscores:1040;Betterindicatedbyhighervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
31.1(2.7)
MedicalOutcomesStudy(MOS)SocialSupportsurveyPosttest(followup11weeks;rangeofscores:1155;Betterindicatedbyhighervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(c)
37(6.6)
37(6.6)
0(
4.58
to
4.58)
MD0higher(4.58
lowerto4.58
higher)
VERY
LOW
MD2.5lower(6.8
lowerto1.8
higher)
VERY
LOW
MedicalOutcomesStudy(MOS)SocialSupportsurvey1monthfollowup(rangeofscores:1155;Betterindicatedbyhighervalues)
1
Smith
2012242
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(c)
33.8(6.2)
36.3(6.2)
2.5(
6.8to
1.8)
(a)
Allocationconcealmentisunclearandtherearebaselinedifferences(whichwouldunderestimateeffectsofintervention)andnoparticipantblinding.
(b)
TheconfidenceintervalcrossesonedefaultMID
NationalClinicalGuidelineCentre,2013.
216
StrokeRehabilitation
Emotionalfunctioning
(c)
TheconfidenceintervaloftheoveralleffectcrossesthedefaultMIDfavouringtheinterventionandthedefaultMIDfavouringthecontrolgroup
NationalClinicalGuidelineCentre,2013.
217
StrokeRehabilitation
Emotionalfunctioning
Table51: Webbasedpsychologicaltherapyvs.controloutcomesforPERSONSWITHSTROKE(HUSBANDS)only(baselineadjustedmeans(sd))
SummaryofFindings
Qualityassessment
Noof
studie
s
Design
Effect
Riskof
bias
Inconsistency
Indirectness
Imprecisio
n
Webbasedpsychological
interventionFrequency(%)/
mean(SD)/median(range)
control
Frequency
(%)/mean
(SD)/
median
(range)
Absolute
effector
RelativeRisk mean
/Mean
difference
difference/
(MD)(95%
(95%CI)
CI)
Confid
ence
in
effect
CentreforEpidemiologicalStudiesDepressionscale(CESD)Posttest(followup11weeks;rangeofscores:060;Betterindicatedbylowervalues)
1
Smith
201224
2
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(b)
19.5(8.5)
20.4(8.7)
0.9(6.86to MD0.9lower
5.06)
(6.86lower
VERY
to5.06
LOW
higher)
CentreforEpidemiologicalStudiesDepressionscale(CESD)1monthfollowup(rangeofscores:060;Betterindicatedbylowervalues)
1
Smith
201224
2
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(c)
14(8.1)
17.9(7.8)
3.9(9.45to MD3.9lower
1.65)
(9.45lower
LOW
to1.65
higher)
22.8(4.9)
1.2(4.53to MD1.2lower
2.13)
(4.53lower
LOW
to2.13
higher)
24.4(4.5)
0.2(2.85to MD0.2
3.25)
higher(2.85
lowerto3.25
higher)
VERY
LOW
27.7(3.7)
1(3.64to
MasteryScalePosttest(followup11weeks;rangeofscores:936;Betterindicatedbyhighervalues)
1
Smith
201224
2
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(c)
21.6(4.6)
MasteryScale1monthfollowup(rangeofscores:936;Betterindicatedbyhighervalues)
1
Smith
201224
2
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(b)
24.6(4.3)
SelfesteemscalePosttest(followup11weeks;rangeofscores:1040;Betterindicatedbyhighervalues)
1
randomise
serious
noserious
noserious
serious(c)
26.7(3.9)
NationalClinicalGuidelineCentre,2013.
218
MD1lower
StrokeRehabilitation
Emotionalfunctioning
SummaryofFindings
Qualityassessment
Noof
studie
s
Smith
201224
2
Design
dtrials
Effect
Riskof
bias
(a)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecisio
n
Webbasedpsychological
interventionFrequency(%)/
mean(SD)/median(range)
control
Frequency
(%)/mean
(SD)/
median
(range)
Absolute
effector
RelativeRisk mean
/Mean
difference
difference/
(MD)(95%
(95%CI)
CI)
1.64)
(3.64lower
to1.64
higher)
Confid
ence
in
effect
LOW
27.2(4.9)
1.3(2.03to
4.63)
MD1.3
higher(2.03
lowerto4.63
higher)
LOW
MD0.5
higher(4.08
lowerto5.08
higher)
VERY
LOW
Selfesteemscale1monthfollowup(rangeofscores:1040;Betterindicatedbyhighervalues)
1
Smith
201224
2
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(c)
28.5(4.6)
MedicalOutcomesStudy(MOS)SocialSupportsurveyPosttest(followup11weeks;rangeofscores:1155;Betterindicatedbyhighervalues)
1
Smith
201224
2
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(b)
41.5(6.6)
41(6.6)
0.5(4.08to
5.08)
MedicalOutcomesStudy(MOS)SocialSupportsurvey1monthfollowup(rangeofscores:1155;Betterindicatedbyhighervalues)
1
Smith
201224
2
(a)
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(b)
43.2(4.6)
44(4.9)
0.8(4.13to MD0.8lower
2.53)
(4.13lower
VERY
to2.53
LOW
higher)
Allocationconcealmentisunclearandtherearebaselinedifferences(whichwouldunderestimateeffectsofintervention)andnoparticipantblinding.
(b)
TheconfidenceintervaloftheoveralleffectcrossesthedefaultMIDfavouringtheinterventionandthedefaultMIDfavouringthecontrolgroup
(c)
TheconfidenceintervalcrossesonedefaultMID
NationalClinicalGuidelineCentre,2013.
219
StrokeRehabilitation
Emotionalfunctioning
9.1.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingpsychologicaltherapiesforthefamilywithusualcare
wereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Beforedeliveringthepsychologicaltherapyformoodtothepatient,somescreeningswouldbe
requirestoassessthepatientsneuropsychologicalprofile.Abriefscreeningofcognitionand
communicationcoulddetermineifthereareanygrossissueswhichmightformbarrierstoeffective
participationintherapy.Iftherearenone,psychologicaltherapiescouldbestartedrightaway.
Whereissuesareindicatedwithcognition,communicationorbehaviour,atalevelthatmayaffect
deliveryofandengagementtotherapy,additionalneuropsychologicalassessmentwouldberequired
toassessthepatientscapabilitiesandcapacityforparticipatinginpsychologicaltherapyandto
provideabasisforanypossibleadaptationstotherapy.
Theinitialneuropsychologicalassessmenttestkitvariesandwouldcomeinatapproximatelyoneoff
costof2000to3000dependingonwhichtestsarepurchasedasthisisusuallydowntothe
cliniciansdiscretionandthisisusuallyupdatedannuallyforaoneoffcostof500.Assessmentsare
carriedoutbyaBand8clinicianandcouldtakebetween24hours.Theestimatedcostperhourof
clientcontactforacommunitybasedclinicalpsychologist(Band8a)is136l;thereforethe
assessmentcostwouldbebetween272and544perpatient.
Thepsychologicaltherapyformoodforthepatientusuallyinvolvesanindirectconsultationwiththe
multidisciplinaryteamandfamilyanddirectclinicalconsultationwiththepersonwhohashada
stroke.(Ifaclinicalconsultationisneeded,thepackageoftherapywouldbenegotiatedwiththe
patient)
Bothassessmentandtherapyinputvaryaccordingtothepatientsneed.Asaballparkfigureupto12
sessionsofpsychologicaltherapymaybeofferedtostrokepatientsfordepression;usuallyone
sessioniscarriedoutperweekandeachsessionwouldtakebetween45minutesandonehour.The
estimatedcostperhourofclientcontactforclinicalpsychologist(band8a)is136m.Thetotal
averagecostoftherapywouldbe1,224perpatient.
Intheeventthatthepsychologistidentifiesmoresignificantmooddisorderforwhichpsychological
therapyisnotappropriate,pharmacologicaltreatmentandtheneuropsychiatricinputwouldbe
requiredforbothassessmentandprescriptionofpharmacologicaltherapy.
9.1.1.3
Evidencestatements
Researchintopsychologicaltherapyforthisgroupofpatientsiswellknowntobeverydifficultandin
itsrelativeinfancyandthereforetheevidencepoolisverylimited.
l EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandAgendaforChangesalaryband8a51(typicalsalarybandidentifiedbyclinicalGDGmembers).
m EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandAgendaforChangesalaryband8a51(typicalsalarybandidentifiedbyclinicalGDGmembers).
NationalClinicalGuidelineCentre,2013.
220
StrokeRehabilitation
Emotionalfunctioning
Clinicalevidencestatements
Forcaregivers:
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthatcaregivers
depression,asmeasuredbytheCentreforEpidemiologicalStudiesDepressionscale(CESD),
statisticallyimprovedwhentheyreceivedanonlinepsychologicalinterventioncomparedtothosein
thecontrolconditionattheendofthe11weekinterventionperiod(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiversdepression,asmeasuredbytheCentreforEpidemiological
StudiesDepressionscale(CESD),whentheyreceivedanonlinepsychologicalinterventioncompared
tothoseinthecontrolconditionat1monthfollowupaftertheendofintervention(LOW
CONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiverslevelofcoping,asmeasuredbytheMasteryScale,whenthey
receivedanonlinepsychologicalinterventioncomparedtothoseinthecontrolconditionattheend
ofthe11weekinterventionperiod(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiverslevelofcoping,asmeasuredbytheMasteryScale,whenthey
receivedanonlinepsychologicalinterventioncomparedtothoseinthecontrolconditionat1month
followupaftertheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiverslevelofselfesteem,whentheyreceivedanonline
psychologicalintervention,comparedtothoseinthecontrolconditionattheendofthe11week
interventionperiod(VERYLOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiverslevelofselfesteem,whentheyreceivedanonline
psychologicalinterventioncomparedtothoseinthecontrolconditionat1monthfollowupafterthe
endofintervention(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiverslevelofsocialsupportskills,whentheyreceivedanonline
psychologicalintervention,comparedtothoseinthecontrolconditionattheendofthe11week
interventionperiod(VERYLOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementincaregiverslevelofsocialsupportskills,whentheyreceivedanonline
psychologicalinterventioncomparedtothoseinthecontrolconditionat1monthfollowupafter
theendofintervention(LOWCONFIDENCEINEFFECT)
Forthepersonwhohashadastroke:
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofdepressionofthepersonwhohashadastroke,asmeasured
bytheCentreforEpidemiologicalStudiesDepressionscale(CESD)whentheyreceivedanonline
psychologicalinterventioncomparedtothoseinthecontrolconditionattheendofthe11week
interventionperiod(VERYLOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofdepressionofthepersonwhohashadastroke,asmeasured
bytheCentreforEpidemiologicalStudiesDepressionscale(CESD),whentheyreceivedanonline
NationalClinicalGuidelineCentre,2013.
221
StrokeRehabilitation
Emotionalfunctioning
psychologicalinterventioncomparedtothoseinthecontrolconditionat1monthfollowupafterthe
endofintervention(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofcopingofthepersonwhohashadastroke,asmeasuredby
theMasteryScale,whentheyreceivedanonlinepsychologicalinterventioncomparedtothoseinthe
controlconditionattheendofthe11weekinterventionperiod(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofcopingofthepersonwhohashadastroke,asmeasuredby
theMasteryScale,whentheyreceivedanonlinepsychologicalinterventioncomparedtothoseinthe
controlconditionat1monthfollowupaftertheendofintervention(VERYLOWCONFIDENCEIN
EFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofselfesteemofthepersonwhohashadastroke,whenthey
receivedanonlinepsychologicalintervention,comparedtothoseinthecontrolconditionattheend
ofthe11weekinterventionperiod(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofselfesteemofthepersonwhohashadastroke,whenthey
receivedanonlinepsychologicalinterventioncomparedtothoseinthecontrolconditionat1month
followupaftertheendofintervention(LOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofsocialsupportskillsofthepersonwhohashadastroke,when
theyreceivedanonlinepsychologicalintervention,comparedtothoseinthecontrolconditionatthe
endofthe11weekinterventionperiod(VERYLOWCONFIDENCEINEFFECT)
Onestudy242comprising32pairsoffemalecaregiversandtheirhusbandsfoundthattherewasno
significantimprovementinthelevelofsocialsupportskillsofthepersonwhohashadastroke,when
theyreceivedanonlinepsychologicalinterventioncomparedtothoseinthecontrolconditionat1
monthfollowupaftertheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
9.1.2
Recommendationsandlinktoevidence
49.Assessemotionalfunctioninginthecontextofcognitivedifficulties
inpeopleafterstroke.Anyinterventionchosenshouldtakeinto
considerationthetypeorcomplexityofthepersons
neuropsychologicalpresentationandrelevantpersonalhistory.
50.Supportandeducatepeopleafterstrokeandtheirfamiliesand
carers,inrelationtoemotionaladjustmenttostroke,recognising
thatpsychologicalneedsmaychangeovertimeandindifferent
settings.
51.Whenneworpersistingemotionaldifficultiesareidentifiedatthe
persons6monthorannualstrokereviews,referthemto
appropriateservicesfordetailedassessmentandtreatment.
Recommendations
52.Managedepressionoranxietyinpeopleafterstrokewhohaveno
NationalClinicalGuidelineCentre,2013.
222
StrokeRehabilitation
Emotionalfunctioning
cognitiveimpairmentinlinewithrecommendationsinDepression
inadultswithachronicphysicalhealthproblem(NICEclinical
guideline91)andGeneralisedanxietydisorder(NICEclinical
guideline113).
Relativevaluesofdifferent Anyqualityoflifeordepressionoutcomewasincludedintheclinicalreview.
outcomes
Thestudyreporteddepression,abilitytocope,selfesteemandemotional
support.
Tradeoffbetweenclinical
benefitsandharms
Havingastrokecanaffectrelationshipsbetweentheindividualwithstroke
andaspouseorpartner.Transitiontonewrolesandadaptationtodisability
arechallenging,andhavebeenlikenedtonavigatingunchartedterritory.The
NationalServiceFrameworkforPeoplewithLongTerm(Neurological)
Conditionshighlightedtheneedforlifelongcareandsupportforpeoplewith
longtermneurologicalconditions,theirfamiliesandcarers.Inviewofthisthe
GDGwishedtospecificallyexaminetheevidencearoundsupportingfamilies
andcouples.Onlyonestudywasfoundinthisarea.Thisparticular
interventionusedanonlinesupportprogrammetoprovidecaregiverswith
knowledgeandskillstoreducetheirownpersonalstressandprovide
emotionalsupporttothepersontheywerecaringfor.WhiletheGDGagreedit
wasimportanttoprovidestrategiestohelppeoplecopeeffectivelyand
managedistressandanumberofdifferentpsychologicalapproachescouldbe
drawnupon,ofwhichonecouldbetheuseofcomputerbasedtherapies,it
wasquestionedwhetherthiswouldbethemostsuitableformatforsome
people,inparticularanoldergeneration.
TheGDGnotedthatprevalenceofdepressioninstrokesurvivorshasbeen
estimatedat1520%.Theynotedthatdepressioninstrokesurvivorsimpacts
onfamilymembers.TheGDGagreedthatanassessmentofemotional
functioningshouldbeconductedinallpatientswithstroke.TheGDG
consideredthatwhilstitwasnotpossibletoprovidedetailonwhatthe
assessmentshouldcomprise,ageneralrecommendationshouldbemadeas
thepersonsmoodwouldhaveamajorimpactonthequalityoflifeofboththe
patientandtheircarers.TheGDGbasedtheirrecommendationonconsensus
opinion.
Economicconsiderations
Nocosteffectivenessstudieswereidentifiedforthisquestion.
Bothassessmentandtherapyvaryaccordingtothepatientsneed.
AssessmentsarecarriedoutbyaBand8Aclinicianandcouldtakebetween2
4hours.Theestimatedcostperhourofclientcontactforacommunitybased
clinicalpsychologist(band8a)is136nthereforetheassessmentcostwouldbe
between272and544perpatient.
Upto12sessionsofpsychologicaltherapymaybeofferedtostrokepatients
fordepression;usuallyonesessioniscarriedoutperweekandeachsession
wouldtakearoundonehour.Thetotalaveragecostoftherapywouldbe
1,224perpatient.
Intheeventthatthepsychologistidentifiesmoresignificantemotional
difficultiesforwhichpsychologicaltherapyisnotappropriate,pharmacological
treatmentandtheneuropsychiatricinputwouldberequiredatthisstagefor
bothassessmentandprescriptionofpharmacologicaltherapy.
TheGDGconsideredthesecoststobelikelyoffsetbythebenefitsandthe
improvementsinthepatientsqualityoflifegeneratedbythepsychological
therapy.
Qualityofevidence
Thestudydemonstratedthatcaregiversdepressionstatisticallyimproved
n EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandAgendaforChangesalaryband8a51(typicalsalarybandidentifiedbyclinicalGDGmembers).
NationalClinicalGuidelineCentre,2013.
223
StrokeRehabilitation
Emotionalfunctioning
whentheyreceivedanonlinepsychologicalinterventioncomparedtothe
depressionscoresofthoseinthecontrolcondition.Thiswastheonlyoutcome
thatdemonstratedasignificantresult.Howeverconfidenceintheresultsfor
thisoutcomewaslowduetothestudybeingdowngradedforseriousriskof
biasandimprecision.Itwasnotclearatwhattimepointafterstrokethestudy
wasconducted.Theinterventionwasdeliveredtopeopleathome,andresults
werereportedattheendofintervention(11weeks)andatonemonthfollow
up.TheGDGagreedthatittakestimeforthepersonafterstroketobeready
forrehabilitationduetothepsychologicaladjustmentrequired,andthiswould
includecarerstoo.Thereforehavingonlyonemonthfollowupmaybea
limitingfactorofthisstudy.
Thegroupagreedthatprovisionofpsychologicalinterventionswasimportant
forfamiliesandcarersofpeopleafterstrokeandfurtherresearchneedstobe
conductedinthisarea.
Otherconsiderations
TheGDGagreedwiththerecommendationsmadeintheNICEguidanceonthe
managementofdepressioninadultswithachronichealthproblem(CG91),
Depression:thetreatmentandmanagementofdepressioninadults(CG90),
andGeneralisedanxietydisorderandpanicdisorder(withorwithout
agoraphobia)inadults(CG113).Theybelievedthesetobeapplicabletopeople
afterstrokewiththeaddedobservationthatduetocognitiveandlanguage
difficultiespsychologicaltherapiesneedtobedeliveredbyanappropriately
trainedandsupervisedprofessionalwhohasanunderstandingofthenatureof
thecognitiveandphysicaldifficultiesandtheirimpact.TheGDGagreedthat
peoplepresentingwithemotionaldifficultiesattheir6monthorannual
reviewsshouldbereferredfordetailedassessment.
NationalClinicalGuidelineCentre,2013.
224
StrokeRehabilitation
Vision
10 Vision
Visionmaybeaffectedafterstrokeinanumberofways.Peoplewithstrokemaybeawareof
difficultieswithperipheralvisionasaresultofavisualfielddefect,doublevisionasaresultof
impairedeyedmovementsorpoorcoordinationofeyemovements,andproblemsarisingasaresult
ofdifficultieswithvisualprocessing.Thischapterfocusesonthetreatmentofhemianopiaand
doublevision.
10.1 Eyemovementtherapy
Hemianopiasareestimatedtoaffectbetween8and25%ofpeoplewithstroke17,92.Thisvision
defectischaracterisedbylowvisionorblindnessincorrespondinghalvesofthefieldofvision.
Peoplesufferingfromhemianopiaorquadrantanopiamayrunintoobjects,triporfall,knockthings
over,andlosetheirplacewhenreading,orbesurprisedbypeopleorobjectsthatseemtoappear
suddenlyoutofnowhere.Somepeoplemaynotbeawareofthedeficit,especiallythosewith
associatedneglect.Eyemovementtherapyencouragesscanningintotheaffectedvisualfieldandis
atechniqueusedwithpatientswithahemianopiapoststroke.
10.1.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessofeye
movementtherapyforvisualfieldlossversususualcare?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention:
Eyemovementtherapyincluding:
Visualsearchtherapy
Visualscanning
Scanningcompensatorytraining
Comparison:
Usualcare(usuallynothing)
Shamvisualrehabilitation
Outcomes:
Reading(speedandaccuracy)
Eyemovementtasks
Scanning
LetterCancellationTest
10.1.1.1
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingeyemovementtherapyasan
interventionforvisualfieldlossinpeopleafterstroke.Onlystudieswithaminimumsamplesizeof
10participants(5ineacharm)andincludingatleast50%ofparticipantswithstrokewereselected.
ThreeRCTswereidentified.Table52summarisesthepopulation,intervention,comparisonand
outcomesforeachofthestudies.
Table52: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
37
Carter,1983
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Acutestroke
patientsina
hospitalsetting
withouttumours
Cognitiveskill
retraininginvolving
visualscanningand
Routinestroke
program.(N=17)
Lettercancellation
test
Visualspatialtasks
NationalClinicalGuidelineCentre,2013.
225
StrokeRehabilitation
Vision
STUDY
POPULATION
INTERVENTION
orextensive
visualspatialtasks.
bilateraldamage. (N=16)
COMPARISON
OUTCOMES
matchingtosample
(Identifyingobjects)
Spitzyna,2007 Patientswitha
246
rightsided
homonymous
hemianopiathat
interferedwith
reading(69%of
themhad
experienced
stroke).
Readingmovingtext
Shamvisual
(MovingText)that
rehabilitation
scrolledfromrightto therapy.(N=8)
left,dailyfortwofour
weekblocks.(N=11)
Readingspeed
Textreadingspeed
Singlewordreading
speed
Eyemovementtask
Visualfieldperimetry
Modden
2012174
Thereweretwo
different
interventions:(1)
Restitutiontraining:A
computerbased
therapyintegrated
perimeterprogram
whichcreatedthe
exactmeasurement
oftheindividual
visualfieldborder.
Targetstimuli
appearedinthe
hemianopicborder
zonetowhichthe
participanthadto
respond(intervention
basedontheprincipal
ofcovertattention
shift.(N=15)
(2)Compensatory
therapy:Acomputer
basedtherapywhich
wasadapted
individuallyaccording
tothesideofthe
hemianopia.The
therapywasusing
visualscanningand
theparticipanthadto
respondtoatarget
icon.(N=15)
Visualfield
expansion(TAP*
visualfield
omissions)
Visualsearch
(cancellationtaskof
theBIT)
Readingperformance
(standardisedtexts
oftheWechsler
MemoryTest)
Attention(alertness
test(TAPPhasic
Alertness)
Visualconjunction
search(TAP,visual
scanning)
BarthelIndex
Patientswith
homonymous
hemianopiawitha
posteriorcerebral
arterystroke.
Patientswere
excludedifthey
hadvisualneglect
eyemovement
disorders,
neuropsychological
disorderslike
aphasia,
dysexecutive
syndromes,
memorydeficits,
orhigherorder
motor
impairmentslike
apraxia.
Occupational
therapy
consistingof
individually
adapted
stimulationof
dailyactivity
tasksto
compensatevia
eye,head,and
body
movements.
(N=15)
*Note.TAP=TestbatteriezurAufmerksamkeitsprfung(Attentiontestbattery)Zimmermann&Fimm(2002)
NationalClinicalGuidelineCentre,2013.
226
StrokeRehabilitation
Vision
Comparison:EyeMovementTherapy(EMT)forvisualfieldlossversususualcare/shamvisualrehabilitation
Table53:EyeMovementTherapyversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Author
Design
Effect
Limitations
Inconsistency
Indirectness
Eye
Movement
Therapy
Mean(SD)
Mean
Usualcare difference
Mean(SD) (95%ci)
Noserious
imprecision
35.9(21.3)
3.8(13.2)
32.10
(15.96,
48.24)
MD32.1
higher
(15.96to
48.24
higher)
Moderate
Noserious
imprecision
31(22.8)
3.3(18)
34.30
(19.28,
49.32)
MD34.3
higher
(19.28to
49.32
higher)
Moderate
Imprecision
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
Visualscanninglettercancellationtest(Betterindicatedbyhighervalues)
Carter,
198337
RCT
Serious
Noserious
limitations(a) inconsistency
Noserious
indirectness
Visualspatialtasksmatchingtosample(Betterindicatedbyhighervalues)
Carter,
198337
(a)
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Randomization,blindingandallocationconcealmentnotclear.
NationalClinicalGuidelineCentre,2013.
227
StrokeRehabilitation
Vision
Table54: Restitutionaltraining/compensatorytreatmentvs.usualcare(occupationaltherapy)
Qualityassessment
Noof
studies
Design
SummaryofFindings
Riskof
bias
Inconsistency
Indirectness
Imprecision Restitutionaltraining
/compensatory
therapy
Mean(SD)
Usual
care
(OT)
Mean
(SD)
Effect
Mean
difference
(95%CI)
Meandifference
(MD)(95%CI)
Confidence
ineffect
Visualfieldenlargement(TAP,VisualFieldAssessment)RestitutionalTraining(followup15days;measuredwith:meanchangescores;Betterindicatedbyhigher
values)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
3.9(4.9)
1.3
(4.7)
2.60(
0.84,6.04)
MD2.6higher
(0.84lowerto
6.04higher)
LOW
Visualfieldenlargement(TAP,VisualFieldAssessment)CompensatoryTreatment(followup15days;measuredwith:meanchangescores;Betterindicatedbyhigher
values)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
2.9(4.0)
1.3
(4.7)
1.60(
1.52,4.72)
MD1.6higher
(1.52lowerto
4.72higher)
LOW
BITcancellationtaskRestitutionalTraining(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
5.3(10.5)
2.3
(5.0)
3(2.89,
8.89)
MD3higher
(2.89lowerto
8.89higher)
LOW
BITcancellationtaskCompensatoryTreatment(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
5.4(5.2)
2.3
(5.0)
3.1(0.55,
6.75)
MD3.1higher
(0.55lowerto
6.75higher)
LOW
Readingperformance(WechslerMemoryTest)RestitutionalTraining(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
very
serious(c)
0.9(2.4)
0.7
(1.0)
0.2(1.12,
1.52)
MD0.2higher
(1.12lowerto
1.52higher)
VERYLOW
Readingperformance(WechslerMemoryTest)CompensatoryTreatment(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
very
serious(c)
0.9(1.1)
0.7
(1.0)
0.2(0.55,
0.95)
NationalClinicalGuidelineCentre,2013.
228
MD0.2higher
(0.55lowerto
VERYLOW
StrokeRehabilitation
Vision
Qualityassessment
Noof
studies
Design
SummaryofFindings
Riskof
bias
Inconsistency
Indirectness
Imprecision Restitutionaltraining
/compensatory
therapy
Mean(SD)
Usual
care
(OT)
Mean
(SD)
Effect
Mean
difference
(95%CI)
2012174
Meandifference
(MD)(95%CI)
Confidence
ineffect
0.95higher)
Attention(TAP,PhasicAlertness)RestitutionalTraining(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
28.5(56.9)
13.3
(112.7)
41.8
(22.09,
105.69)
MD41.8higher
(22.09lowerto
105.69higher)
LOW
Attention(TAP,PhasicAlertness)CompensatoryTreatment(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
77.8(112.9)
13.3
(112.7)
91.1
(10.37,
171.83)
MD91.1higher
(10.37to171.83
higher)
LOW
Visualconjunctionsearch(TAPvisualscanning)RestitutionalTraining(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
very
serious(c)
2.7(5.1)
3.5
(6.8)
0.8(5.10, MD0.8lower
3.50)
(5.1lowerto3.5
higher)
VERYLOW
Visualconjunctionsearch(TAPvisualscanning)CompensatoryTreatment(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
7.0(5.0)
3.5
(6.8)
3.5(77,
7.77)
MD3.5higher
(0.77lowerto
7.77higher)
LOW
ExtendedBarthelIndexRestitutionalTraining(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(d)
1.5(2.8)
1.8
(2.0)
0.3(2.04, MD0.3lower
3.50)
(2.04lowerto
1.44higher)
LOW
ExtendedBarthelIndexCompensatoryTreatment(followup15days;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Modde
n
2012174
randomised Serious
trials
(a)
noserious
inconsistency
noserious
indirectness
serious(d)
3.3(3.6)
1.8
(2.0)
1.5(0.58,
3.58)
NationalClinicalGuidelineCentre,2013.
229
MD1.5higher
(0.58lowerto
3.58higher)
LOW
StrokeRehabilitation
Vision
(a)
Therewasnoallocationconcealmentandonlyparticipantswereblinded.
(b)
Theconfidenceintervalofthetotaleffectrangesfromimprovementassociatedwiththeinterventiontonoeffect(crossingonedefaultMID)
(c)
Theconfidenceintervalofthetotaleffectrangesfromappreciablebenefittoappreciableharm(crossedtwodefaultMIDs)
(d)
Theconfidenceintervalofthetotaleffectrangesfromappreciableimprovementassociatedwithusualcaretonoeffect(crossingagreedMID1.85)
(e)
Theconfidenceintervalofthetotaleffectrangesfromappreciableimprovementassociatedwiththeinterventiontonoeffect(crossingagreedMID1.85)
NarrativeSummary
ThefollowingstudyissummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Spitzynaetal,2007246comparedreadingmovingtexttoashamvisualrehabilitationinhemianopicpatients(mainlystrokepatients).Readingmovingtext
inducedsmallfieldoptokineticnystagmus(OKN)andpreferentiallyaffectedreadingsaccadesintotheblindfield.Theoutcomesreportedwere:reading
speeds,eyemovementsandvisualfieldperimetry.Authors246reportedasignificantimprovementinthereadingspeedsandassociatedeyemovements
withparticipantsinthereadingmovingtextgroupcomparedwiththeshamvisualrehabilitationgroupbuttherewasnochangewiththevisualfield
perimetryacrossthegroups.
NationalClinicalGuidelineCentre,2013.
230
StrokeRehabilitation
Vision
10.1.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingeyemovementtherapyforvisualfieldlosswithusual
carewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
TheGDGconsideredthateyemovementtherapyforvisualfieldlosswouldmostlikelybedelivered
byanorthoptistoranoccupationaltherapistintheNHSandwouldtypicallyconsistofaninitial60
minuteassessmentwitha30minutefollowupappointmenteverythreeweeksandfollowupwould
berequiredonaveragefor6months.Theestimatedcostperhourofclientcontactforaband7
orthoptistis59o51(typicalsalarybandidentifiedbyclinicalGDGmembers).Thisequatestoan
estimatedtotalcostperpatientof285.
10.1.1.3
Evidencestatements
Clinicalevidencestatement(s)
Onestudy37comprisingof33participantsshowedastatisticallysignificantimprovementinthevisual
scanning(lettercancellation)testintheeyemovementtherapygroupcomparedtotheusualcare
group(MODERATECONFIDENCEINEFFECT).
Onestudy37comprisingof33participantsshowedastatisticallysignificantimprovementinvisual
spatialtasksforparticipantswhoreceivedeyemovementtherapycomparedtotheusualcare
group(MODERATECONFIDENCEINEFFECT).
Visualfieldenlargement
Restitutionaltraining
Onestudy174comprisingof15participantsintherestitutionaltraininggroupand15control
participantsshowednosignificantimprovementinvisualfieldenlargement(asassessedby
anattentionvisualfieldassessment)betweenrestitutionaltrainingandcontrolgroups(LOW
CONFIDENCEINEFFECT).
Compensatorytherapy
Onestudy174comprisingof15participantsinthecompensatorytreatmentgroupand15
controlparticipantsshowednosignificantimprovementinvisualfieldenlargement(as
assessedbyanattentionvisualfieldassessment)betweencompensatorytreatmentand
controlgroups(LOWCONFIDENCEINEFFECT).
Visualsearch(BITcancellationtest)
Restitutionaltraining
o EstimatedbasedondataandmethodsfromthePersonalSocialServicesResearchUnitUnitcostsofhealthandsocial
carereportandAgendaforChangesalaryband7.Assumedthatanorthoptistiscostedsimilartootheralliedhealth
professionals.
NationalClinicalGuidelineCentre,2013.
231
StrokeRehabilitation
Vision
Onestudy174comprisingof15participantsintherestitutionaltraininggroupand15control
participantsshowednosignificantimprovementinvisualsearchability(asassessedbythe
BITcancellationtask)betweenrestitutionaltrainingandcontrolgroups(LOWCONFIDENCE
INEFFECT).
Compensatorytherapy
Onestudy174comprisingof15participantsinthecompensatorytreatmentgroupand15
controlparticipantsshowednoimprovementinvisualsearchability(asassessedbytheBIT
cancellationtask)betweencompensatorytreatmentandcontrolgroups(LOWCONFIDENCE
INEFFECT).
Readingperformance(readingtextfromWechslerMemoryTest)
Restitutionaltraining
Onestudy174comprisingof15participantsintherestitutionaltraininggroupand15control
participantsshowednosignificantimprovementinreadingperformance(readingtextfrom
theWechslerMemoryTest)betweenrestitutionaltrainingandcontrolgroups(VERYLOW
CONFIDENCEINEFFECT).
Compensatorytherapy
Onestudy174comprisingof15participantsinthecompensatorytreatmentgroupand15
controlparticipantsshowednoimprovementinreadingperformance(readingtextfromthe
WechslerMemoryTest)betweencompensatorytreatmentandcontrolgroups(VERYLOW
CONFIDENCEINEFFECT).
Attention(PhasicAlertness)
Restitutionaltraining
Onestudy174comprisingof15participantsintherestitutionaltraininggroupand15control
participantsshowednosignificantimprovementinattentioncontrolbetweenrestitutional
trainingandcontrolgroups(LOWCONFIDENCEINEFFECT).
Compensatorytherapy
Onestudy174comprisingof15participantsinthecompensatorytreatmentgroupand15
controlparticipantsshowedastatisticallysignificantimprovementinattentioncontrol
associatedwithcompensatorytreatmentcomparedtousualcare(LOWCONFIDENCEIN
EFFECT).
Visualconjunctionsearch(visualscanningtest)
Restitutionaltraining
Onestudy174comprisingof15participantsintherestitutionaltraininggroupand15control
participantsshowednosignificantimprovementinvisualconjunctionsearchskills(assessed
byavisualscanningtest)betweenrestitutionaltrainingandcontrolgroups(VERYLOW
CONFIDENCEINEFFECT).
Compensatorytherapy
Onestudy174comprisingof15participantsinthecompensatorytreatmentgroupand15
controlparticipantsshowednoimprovementinvisualconjunctionsearchskills(assessedby
avisualscanningtest)betweencompensatorytreatmentandcontrolgroups(LOW
CONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
232
StrokeRehabilitation
Vision
Activitiesofdailyliving(BarthelIndex)
Restitutionaltraining
Onestudy174comprisingof15participantsintherestitutionaltraininggroupand15control
participantsshowednosignificantimprovementinperformanceofactivitiesofdailyliving
(assessedbytheBarthelIndex)betweenrestitutionaltrainingandcontrolgroups(VERYLOW
CONFIDENCEINEFFECT).
Compensatorytherapy
Onestudy174comprisingof15participantsinthecompensatorytreatmentgroupand15
controlparticipantsshowednoimprovementinperformanceofactivitiesofdailyliving
(assessedbytheBarthelIndex)betweencompensatorytreatmentandcontrolgroups(LOW
CONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
10.1.2
Recommendationsandlinktoevidence
53.Screenpeopleafterstrokeforvisualdifficulties.
54.Offereyemovementtherapytopeoplewhohavepersisting
hemianopiaafterstrokeandwhoareawareofthecondition.
55.Whenadvisingpeoplewithvisualproblemsafterstrokeaboutdriving,
consulttheDriverandVehicleLicensingAgency(DVLA)regulations.
Recommendation
Relativevaluesof
differentoutcomes
Theoutcomesofinterestincludedinthereviewwere:readingspeedand
accuracy,eyemovementscanningandlettercancellation.TheGDGconsidered
thattheoutcomesmeasuresincludedinthereviewwereofequalvalue,although
readingspeedandaccuracyrepresentsareallifetaskwhereasscanningandletter
cancellationareimpairmentlevelmeasures.
Tradeoffbetween
clinicalbenefitsand
harms
Homonymoushemianopiacanimpactonarangeofactivitiesofdailyliving
including,reading,driving,navigation,eating,hygienerelatedactivities,andsocial
interaction.Therearesignificantsafetyissuesassociatedwithmisseddiagnosis
includingfalls,injuriesandmotorvehicleaccidents.
Thereisabenefittoadiagnosisofpersistenthomonymoushemianopiainterms
ofaccesstoregistrationofvisualimpairmentandsubsequentaccesstosensory
rehabilitationteams.
Aproportionofpatientsdospontaneouslyadapttotheimpairment,butthe
numbersarepresentlyunknown.Persistent(nonrecovered)homonymous
hemianopiacanhaveasignificantimpactonqualityoflife.Thegroupconsidered
thattreatingthisconditionwouldprovidemajorbenefitsintermsofimproving
qualityoflifefortheindividualpatient.TheGDGalsobelievedthebenefitsofthe
interventionaresignificantgiventherisksofleavingtheconditionuntreated.
Patientswithhomonymoushemianopiamustnotdrivewithinoneyearoftheir
strokeonset.TheymaybeabletoreapplytotheDVLAafteroneyeariftheycan
provethattheyhavelearnedtocompensateforthedefect.(Medicalpractitioners
AtaGlanceGuidetotheCurrentMedicalStandardsofFitnesstoDrive,DVLA,
201171).
NationalClinicalGuidelineCentre,2013.
233
StrokeRehabilitation
Vision
Economic
considerations
Nocosteffectivenessstudieswereidentified.Deliveringeyemovementtherapy
forvisualfieldlosswouldinvolvesomeadditionalcostsintermsofanorthoptist
oroccupationaltherapistassessmentandfollowuptime.TheGDGconsidered
thattheadditionalcostswouldpotentiallybeoffsetbythelongtermbenefitto
patientsintermsofimprovedqualityoflife.
Qualityofevidence
OnesmallstudybyCarter,198337examiningamixedpopulationofpatientswith
neglectandhomonymoushemianopia,demonstratedanimprovementinvisual
scanningstrategiesafterinterventionasmeasuredbylettercancellationand
visualspatialtestandtheconfidenceintheseeffectswasgradedasmoderate.
Asecondsmallstudy(Spitzyna,2007)examinedpatients(75%ofwhomhada
stroke)withpersistenthomonymoushemianopia,usinganovelinterventionof
movingtext.Theauthorsreportedasignificantimprovementinreadingspeed
andeyemovementsbutresultswerenotpresentedinnumericaldatathatcould
beincludedwithintheGRADEanalysis.
TheGDGnotedthattheCarterStudywaspoorlydefinedintermsofpatient
recruitmentandthatitwasunclearifthepatientshadhemianopaorvisual
neglect,orboth,butthesameinterventionwasusedforbotheffectively.
TheGDGconsideredthattherewasinsufficientevidencetoreachgeneralised
conclusionsregardingefficacyrelatedtoactivitiesofdailyliving,althoughthereis
someevidenceregardingeffectivenessforreading.TheGDGconsideredthatit
wasimportantforpeoplewhohavehadastroketobeassessedforvisualfield
defectsandbecauseoftheimpactthisimpairmenthasonthequalityofthe
personslifeandtheserioussafetyissuesinleavingthisuntreated,TheGDG
agreedthatastronglywordedrecommendationneededtobemadetoreflect
theseconcernseventhoughtheevidencewaslimitedtoonesmallstudy.
Itwasnotedthatfurtherresearchinthisareaisrequired.
Otherconsiderations
TheGDGwereuncertainabouttheprevalenceofhomonymoushemianopiawithin
astrokepopulationandrequestedthatanadditionalliteraturesearchbe
conducted.Sixstudies17;45;79;99;92;261wereidentifiedwhichaddressedprevalence,
thesewereofvaryingquality,oftenexaminingaselectedpopulationwithina
hospitalsetting.Onthebasisofthesestudies,theGDGfeltaprevalenceof
persistenthomonymoushemianopiainthecommunitywaslikelytobebetween8
and25%.
Halfofthepatientswithinthepapersreviewedwerenotawarethattheywere
sufferingfromhomonymoushemianopia.Itwasnotedthatroutinescreeningfor
visualfielddefectswasnotcurrentlyuniversalandthereforepotentialpatients
werenotidentifiedorreferredfortherapy.Attentionshouldbepaidatstroke
onsettoelicitingvisualfielddefects.Thegroupconsideredthatperforming
screeningassessmentisgoodpracticeandshouldbeundertaken.
10.2 Diplopiaorotherongoingvisualsymptomsafterstroke
Astrokemayleadtoproblemswitheyemovementswhichresultinbotheyesnotworkingtogether
asapair.Thiscanmakeitdifficulttofocusonspecificthingsbecauseofblurredvisionaswellas
diplopia(ordoublevision)whichimpactsonreading,walkingandperformingeverydayactivities.
Treatmentcaninvolveprisms,exercisesandocclusion.
Asearchforsystematicreviewswascarriedoutforevidenceonthemanagementofdiplopiaand
ongoingvisualsymptomsinpeopleafterstroke.Noreviewswereidentifiedandtherefore
recommendationsinthissectionwerebasedonmodifiedDelphiconsensusstatementswhichwere
basedonrecommendationsfrompublishednationalandinternationalguidelines.Belowweprovide
tablesofstatementsthatreachedconsensusandstatementsthatdidnotreachconsensusandgivea
summaryofhowtheywereusedtodrawuptherecommendations.Fordetailsontheprocessand
methodologyusedforthemodifiedDelphisurveyseeAppendixF.ThissectionoftheDelphisurvey
NationalClinicalGuidelineCentre,2013.
234
StrokeRehabilitation
Vision
wasaimedatDelphipanelmemberswiththerelevantexperiencetocommentonvisualimpairments
instroke.Othermemberscouldoptoutofthissection.Thereforetheresponseratewaslower.
10.2.1
10.2.2
Evidencereview:Howshouldpeoplewithvisualimpairmentsincludingdiplopiabebest
managedafterastroke?
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Components
Continuedmonitoringandreaccessintorehab
Longtermsupport/careathome
Socialparticipationactivities
Carer/familysupport&education
Outcomes
Patientandcarersatisfaction
Qualityoflife
optimisedstrategiestominimiseimpairmentandmaximiseactivity/participation
Delphistatementswhereconsensuswasachieved
Table55: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
Statement
10.2.3
Peoplewhohavepersistingdouble
visionafterstrokerequireaformal
orthopticassessment.
Results
%
70.8
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
1/24(4%)panelmember
commented
Thepersonwhocommentedthought
thatallotherformsofvisual
impairmentwouldalsorequire
orthopticassessment.
Delphistatementwhereconsensuswasnotreached
Table56: Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
1.
Statement
Allpeoplewhohaveimpairedacuity,
doublevisionoravisualfielddefect
followingastrokerequireaformal
ophthalmologyassessment.
Results
%
Amountandcontentofpanel
commentsorthemes
23.8
Inround27/24(29%)panel
memberscommented;7/21(33%)in
round3
Itwaspointedoutthatdifferent
aspectsinthestatementrequire
differentactions(Impairedacuity
anddoublevisionbothrequirean
ophthalmologicaldiagnosis.Visual
fielddefectafterstrokeisless
problematic,andthediagnosisis
usuallyknowninsuchcases
adaptivetreatmentsandeducation
NationalClinicalGuidelineCentre,2013.
235
StrokeRehabilitation
Vision
Number
Statement
Results
%
Amountandcontentofpanel
commentsorthemes
arethepriority.).
Othercommentsalsohighlighted
thatthisisnotalwaysneeded.
2.
Peoplewhohaveongoingvisual
symptomsafterastroke,shouldbe
providedwithinformationon
compensatorystrategiesfrom:
Ophthalmologyservices
Orthopticservices
Occupationaltherapyservices
15.7
50.0
31.5
Inround26/23(26%)panel
memberscommented;9/20(45%)in
round3
Itwashighlightedthatitdependson
availabilityandontheneed
(OccupationalTherapistsaremost
likelytoadvisererehabilitationand
applicationtodailylifewhereas
orthoptistscanadviseonvision
strategies.OphthalmologywillAx
andRxeyeproblemsbutperhapsnot
somuchadviseonstrategies.).
Onepanelmemberwasinvolvedin
thedevelopmentofwebbased
therapiesthatworkbyinducing
compensatoryeyemovements
3.
Peoplewhohavehadastrokeand
havevisualimpairmentsshouldbe
providedwithcontactdetailsforthe
RNIBorStrokeAssociationforfurther
informationonvisualimpairments
afterstroke.
38.1
Inround24/23(17%)panel
memberscommented;1/21(5%)in
round3
Peoplewhohavepersistingdouble
visionafterstrokerequireaformal
orthopticassessment.
Itwaspointedoutthatthisshouldbe
doneifsymptomspersistandnot
givenroutinelytoeverybody.
4.
Assessmentandinformationfor
registeringassightimpairedor
severelysightimpairedshouldbe
providedbyreferraltoan
ophthalmologist.
47.6
NationalClinicalGuidelineCentre,2013.
236
Inround22/24(8%)panel
memberscommented;5/21(24%)in
round3
Itwascommentedthat:
Allinvolvedinstrokecareshould
realisethatonlyophthalmologists
cansignthecertificationofvisual
impairmentform.
Othersqueriedwhetheran
orthoptistcouldalsodothis.
StrokeRehabilitation
Vision
10.2.4
RecommendationsandlinkstoDelphiconsensussurvey
Statements
27.Peoplewhohavepersistingdoublevisionafterstrokerequireaformal
orthopticassessment.
56.Referpeoplewithpersistingdoublevisionafterstrokeforformal
orthopticassessment.
Economicconsiderations Therearecostsassociatedwithaformalorthopticassessment.The
estimatedcostperhourofclientcontactforaband7orthoptistis59p51
(typicalsalarybandidentifiedbyclinicalGDGmembers).Thereis
currentlyalackofconvincingevidenceinfavourofanyinterventionfor
thetreatmentofdiplopiaafterstroke.However,theGDGthoughtthata
formalorthopticassessmentmightindicateunderlyingindividualcauses
thatmayleadtopossibletreatmentactivities.Forthisreason,theGDG
consideredthecostsassociatedwithorthopticassessmentlikelytobe
offsetbyitsbenefits.
Otherconsiderations
TheGDGinterpretedthelackofconsensusasindicatingnoconclusive
agreementcouldbedrawnfromtheDelphipanelonwhatisbeneficial
fordiplopia.TheGDGtookintoaccountthatthisisaconditionthat
wouldseriouslyaffectanindividualsqualityoflifeandthatitistherefore
importantthatthisisformallyassessed.
Eventhoughthereisnotenoughrobustevidencetosupportone
treatmentoveranotherfordiplopiaatpresent,theGDGthoughtthatthe
resultsmayindicateapathoftreatmentoptionsbasedonindividual
need.Itisalsopossiblethataformalorthopticassessmentmightindicate
underlyingindividualcausesthatmayleadtopossibletreatment
activities,suchasprismsorpatching.
TheGDGalsoconsideredthatthattheprovisionofinformationtothe
personwhoexperiencesdiplopiapoststrokeandtheircarer/familyis
centralinthisprocess(includingavailabletreatmentoptions).However,
theGDGstressedthatitisimportantforclinicianstokeepinmindthat
thereiscurrentlyalackofconvincingevidenceinfavourofany
intervention.Itisthereforenecessaryindiscussionswiththepatientand
theircarers/familytobesensitiveandsetrealisticgoals.
p EstimatedbasedondataandmethodsfromthePersonalSocialServicesResearchUnitUnitcostsofhealthandsocial
carereportandAgendaforChangesalaryband7.Assumedthatanorthoptistiscostedsimilartootheralliedhealth
professionals.
NationalClinicalGuidelineCentre,2013.
237
StrokeRehabilitation
Swallowing
11 Swallowing
Dysphagia(difficultyswallowing)iscommonfollowingstroke,occurringinupto67%ofstroke
patients.Strokepatientswithdysphagiahavehigherratesofchestinfection,aspirationpneumonia,
dehydrationandmalnutritionthanstrokepatientswithoutdysphagia.Thepresenceofdysphagiais
alsoassociatedwithasignificantlyincreasedriskofdeath,disability,lengthofhospitalstay,and
institutionalcare.
Symptomsandsignswhichmayindicatethepresenceofdysphagiainclude:
Afeelingthatfoodorliquidisstickinginthethroat;
Asensationofaforeignbodyor"lump"inthethroat;
Aneedtomodifyorrestrictcertainfoodtypes
Drooling;
Difficultyinitiatingaswallow
Nasalregurgitationoffoodordrinkduringswallowing
Coughingorchokingduringeatinganddrinking
Gurglyorwetvoiceafterswallowing
Unexplainedweightloss
Respiratorysymptomsincludingincreasingrespiratoryrateandshortnessofbreath.
Dysphagiarehabilitationprogrammesuseacombinationofapproachesaimedateither
improvingorcompensatingfortheunderlyingdisorder.Programmesmayfocuson
strengtheningmusclesoronusingdifferentgroupsofmusclestoassumethefunctionof
thedamagedmuscles.Generaldysphagiamanagementprogrammesthatincorporate
earlyidentificationofswallowingdifficultiesthroughscreeningorassessmentand
modificationoforalintakehavebeenassociatedwithareducedriskofpneumoniainthe
acutestageofstroke.
11.1.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
interventionsforswallowingversusalternativeinterventions/usualcaretoimprove
difficultyswallowing(dysphagia)?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehada
stroke
Intervention:
Carbonatedwater
Frazierfreewaterprotocol
Swallowingexercises:
a.effortfulswallowingtechnique
b.headpositioning
c.tongueexercises
d.thickenedfluids/texturemodification
NationalClinicalGuidelineCentre,2013.
238
StrokeRehabilitation
Swallowing
ClinicalMethodologicalIntroduction
e.Mendelssohnsmanoeuvre
11.1.1.1
Comparison:
Usualcare
Thickeningfluids
NilbymouthAlternativeinterventions
Nasogastricfeeding
Outcomes:
Occurrenceofaspirationpneumonia
Occurrenceofchestinfections
Reductioninhospitalstay
Reductioninreadmission
Returntonormaldiet
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparinginterventionstoimprove
swallowingforreducingdysphagiainpatientswithstroke.Onlystudieswithaminimumsamplesize
of20participants(10ineacharm)andincludingatleast50%ofparticipantswithstrokewere
selected.Three(3)RCTswereidentified.
Table57:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
Carnaby
etal,
200636
Patientswithstroke Standardlowintensity
withintheprevious
swallowingtherapy:
7dayswitha
appropriatedietary
clinicaldiagnosisof
modificationand
swallowing
swallowingcompensation
difficultyandno
strategies,mainly
historyof
environmental
swallowing
modifications(for
treatmentor
exampleupright
surgeryofthehead
positioningforfeeding),
orneck.
safeswallowingadvice
(forexamplereducedrate
ofeating)for3timesper
weekforamonthor
duringthehospitalstay(if
lessthanamonth)
(N=102)
Standardhighintensity
swallowingtherapy:
dietarymodificationand
directswallowing
exercises(forexample,
effortfulswallowing,
supraglotticswallow
technique)everyworking
dayforamonthordaily
forthedurationof
hospitalstay(iflessthana
NationalClinicalGuidelineCentre,2013.
239
OUTCOMES
Usualcare:
Returntopre
strokedietin6
Physiciansreferred
months.
theirpatientsto
thespeechand
Occurrenceof
languagetherapists
severechest
iftheyconsideredit
infection
tobeappropriate.
Treatment,if
offered,consisted
mainlyof
supervisionfor
feedingand
precautionsforsafe
swallowing(for
example,
positioning,slowed
rateoffeeding).
(N=102)
StrokeRehabilitation
Swallowing
STUDY
POPULATION
INTERVENTION
month).(N=102)
COMPARISON
OUTCOMES
DePippo
etal,
199463
Acutestroke
patientsover20
yearsoldwith
strokeina
rehabilitationunit
withnoknown
historyof
significantoralor
pharyngeal
anomaly.
Dietcontrolanddaily
reinforcementof
compensatoryswallowing
techniquesaddedtoformal
dysphagiatreatmentsession
(dietprescription,dietand
compensatoryswallowing
technique
recommendations)bya
dysphagiatherapist.(N=39)
Formaldysphagia
treatmentsession
(dietprescription,
dietand
compensatory
swallowing
technique
recommendations)
byadysphagia
therapist(N=38).
Occurrenceof
pneumonia
Patientshad
thickenedfluids
only(withmealsor
asrequested);no
compensatory
swallow
techniques,direct
orindirectswallow
therapyorcues
given.(N=10)
Occurrenceof
aspiration
pneumonia
NationalClinicalGuidelineCentre,2013.
240
StrokeRehabilitation
Swallowing
Comparisonofbehaviouralinterventionsfordysphagiaversususualcare
Table58:StandardlowintensityswallowingtherapyfordysphagiaversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Standard
low
intensity
swallowing
therapy
Frequencies
(%)
Effect
Usualcare
Frequencies
(%)
Relative
Risk
(95%CI)
Absolute
effect(95% Confidence
CI)
(ineffect)
Inconsistency
Indirectness
Imprecision
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
65/102
(63.7%))
57/102
(55.9%)
1.14
(0.91to
1.43)
78more
per1000
(from50
fewerto
240more)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
26/102
(25.5%)
48/102
(47.1%)
0.54
(0.37to
0.8)
216fewer Moderate
per1000
(from94
fewerto
296fewer)
Returntoprestrokedietin6months
1
Carnaby
200636
RCT
Noserious
single
limitations
blinded
Moderate
Occurrenceofseverechestinfection
1
Carnaby
200636
(a)
(b)
RCT
Noserious
single
limitations
blinded
ConfidenceintervalcrossedoneendofdefaultMID(1.25).
ConfidenceintervalcrossedoneendofdefaultMID(0.75).
NationalClinicalGuidelineCentre,2013.
241
StrokeRehabilitation
Swallowing
Table59:StandardhighintensityswallowingtherapyfordysphagiaversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Standard
high
intensity
swallowing
therapy
Frequencies
(%)
Effect
Usual
care
Frequenc
ies(%)
Relative
Risk(95%
CI)
Absolute
effect(95%
CI)
Confidence
(ineffect)
Inconsistency Indirectness
Imprecision
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
71/102
(69.6%)
57/102
(55.9%)
1.25(1to
1.54)
140moreper
1000(from0
moreto302
more)
Moderate
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
28/102
(27.5%)
48/102
(47.1%)
0.58(0.4to
0.85)
198fewerper
1000(from71
fewerto282
fewer)
Moderate
Returntoprestrokedietin6months
1
Carnaby
200636
RCTsingle Noserious
blinded
limitations
Occurrenceofseverechestinfection
1
Carnaby
200636
(a)
(b)
RCTsingle Noserious
blinded
limitations
ConfidenceintervalcrossedoneendofMID(1.25)
ConfidenceintervalcrossedoneendofMID(0.75)
NationalClinicalGuidelineCentre,2013.
242
StrokeRehabilitation
Swallowing
Table60:ReinforcementofswallowingposturesversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency Indirectness
Imprecision
Reinforcement
ofswallowing
postures
(frequencies%)
Usual
care
(frequen
cies%)
Effect
Relative
Risk(95%
CI)
Absolute
effect(95%
CI)
Occurrenceofpneumonia
1
DePippo
199463
(a)
(b)
RCT
Serious
limitations
(a)
Confidence
(ineffect)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(b)
2/39(5.1%)
5/38
(13.2%)
Noallocationconcealment,unclearblinding
ConfidenceintervalcrossedbothendsofMID(0.75,1.25)
NationalClinicalGuidelineCentre,2013.
243
0.39(0.08
to1.89)
80fewer
per1000
(from121
fewerto
117more)
Verylow
StrokeRehabilitation
Swallowing
Comparisonofunlimitedoralintakeofwaterinadditiontothickenedliquidsversusthickenedliquidsonly
Table61:UnlimitedoralintakeofwaterinadditiontothickenedliquidsversusthickenedliquidsonlyClinicalstudycharacteristicsandclinical
summaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency Indirectness
Imprecision
Unlimited
oralintake
ofwater
(plus
thickened
liquids)
(frequenci
es%)
Thickened
liquids
only
(frequenci
es%)
Effect
Relative
Risk(95%
CI)
Absolute
effect(95%
CI)
Occurrenceofaspirationpneumonia
1Garon
199789
RCT
unblinded
Very
serious
limitations(
a)
Confidence
(ineffect)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
0/10
0/10
(b)
(a) Unblindedstudy;randomisationandallocationconcealmentunclear.
(b) Noeventsexperiencedinanygroup(intervention,control)sonorelativeandabsoluteeffectcouldbeestimated.
NationalClinicalGuidelineCentre,2013.
244
(b)
Low
StrokeRehabilitation
Swallowing
11.1.2
EconomicLiteraturereview
Onestudywasincludedthatincludedtherelevantcomparison.169Thisissummarisedinthe
economicevidenceprofilebelow(Table62andTable63).Seealsothefullstudyevidencetablesin
AppendixI.
Table62: StandardlowintensityswallowingtherapyversususualcareEconomicstudy
characteristics
Study
169
Marsh2010
(UK)
Limitations
Applicability
Othercomments
Minorlimitations(a)
Directlyapplicable
(b)
Theauthorslabelledthisstudyasacost
benefitanalysis;howeveritisinfacta
costanalysistakingintoaccountinitial
andfuturecosts.
(a) EffectivenessdataisbasedononeRCTsodoesnotreflectalltheevidenceinthisareaastheclinicalreviewincluded
morestudies.
(b) Theprobabilityofrequiringhospitalorcommunitycareforchestinfectionwasbasedondatanotspecifictopeoplewith
dysphagia.
Table63: StandardlowintensityswallowingtherapyversususualcareEconomicsummaryof
findings
Study
169
Marsh2010
(UK)
Incremental
cost
Incremental
effects
NetBenefit
Uncertainty
saves213
(a)
NR
NR
Thresholdanalysis:standardlow
intensityswallowingtherapyiscost
savingaslongasprobabilityof
developingachestinfectionwith
standardtherapyisbelow38%.
Thecostofchestinfectionrequiring
hospitaladmissionwasvariedbetween
1,800and5,100.Standardlow
intensityswallowingtherapyiscost
savingaslongasthecostofchest
infectionrequiringhospitaladmissionis
above2,000.
(a) Costsofstafftimefortheinitialstrategy,futurecostsoftreatingchestinfectionsinhospitalandcommunity.Standard
lowintensityswallowingtherapyismorecostlyinitiallywhencomparedtousualcare(219versus59)butitis
associatedwithlowerratesofchestinfectionsandlowercostoftreatingchestinfections(659versus872).
11.1.3
Evidencestatements
Clinicalevidencestatement(s)
Standardlowintensityswallowingtherapyfordysphagiaversususualcare
Onestudy(Carnaby200636)comprising204peoplewhohavehadastrokeshowedthatpeoplewho
receivedastandardlowintensityswallowingtherapywerenomorelikelytoreturntotheirpre
strokedietafter6monthsthanthosewhoreceivedusualcare(moderateconfidenceintheeffect).
Onestudy(Carnaby200636)comprising204peoplewhohavehadastrokeshowedthattherewere
significantlyfewerpeopleofthosewhoreceivedastandardlowintensityswallowingtherapy
NationalClinicalGuidelineCentre,2013.
245
StrokeRehabilitation
Swallowing
experiencingchestinfectionscomparedtothosewhoreceivedusualcare(moderateconfidencein
theeffect).
Standardhighintensityswallowingtherapyfordysphagiaversususualcare
Onestudy(Carnaby200636)comprising204peoplewhohavehadastrokeshowedthatpeoplewho
receivedastandardhighintensityswallowingtherapyweresignificantlymorelikelytoreturntotheir
prestrokedietafter6monthsthanthosewhoreceivedusualcare(moderateconfidenceinthe
effect).
Onestudy(Carnaby200636)comprising204peoplewhohavehadastrokeshowedthattherewere
significantlyfewerpeopleofthosewhoreceivedastandardhighintensityswallowingtherapy
experiencingchestinfectionscomparedtothosewhoreceivedusualcare(moderateconfidencein
theeffect).
Reinforcementofswallowingposturesversususualcare
Onestudy(DePippo199463)comprising77peoplewhohavehadastrokeshowedthatpeoplewho
receivedreinforcementofswallowingposturesdidnothavehigherratesofpneumoniacomparedto
thosewhoreceivedusualcare(verylowconfidenceintheeffect).
Unlimitedoralintakeofwaterinadditiontothickenedliquidsversusthickenedliquidsonly
Onestudy(Garon199789)comprising20peoplewhohavehadastrokeshowednocaseofaspiration
pneumoniaineitherthegroupwhoreceivedunlimitedoralintakeofwaterinadditiontothickened
liquidsorthethickenedliquidsonlygroup(verylowconfidenceintheeffect)
Healtheconomicevidencestatement(s)
OnedirectlyapplicablestudywithminorlimitationsshowedthatlowintensitySLTsavesaround
213perpatientcomparedtousualcarewheninitialcostsandcostoftreatingchestinfections
areincluded.
NationalClinicalGuidelineCentre,2013.
246
11.1.4
Recommendationsandlinktoevidence
57.Assessswallowinginpeopleafterstrokeinlinewith
recommendationsinStroke(NICEclinicalguideline68).
58.Offerswallowingtherapyatleast3timesaweektopeople
withdysphagiaafterstrokewhoareabletoparticipate,foras
longastheycontinuetomakefunctionalgains.Swallowing
therapycouldincludecompensatorystrategies,exercisesand
posturaladvice.
59.Ensurethateffectivemouthcareisgiventopeoplewith
difficultyswallowingafterstroke,inordertodecreasethe
riskofaspirationpneumonia.
60.Healthcareprofessionalswithrelevantskillsandtrainingin
thediagnosis,assessmentandmanagementofswallowing
disordersshouldregularlymonitorandreassesspeoplewith
dysphagiaafterstrokewhoarehavingmodifiedfoodand
liquiduntiltheyarestable(thisrecommendationisfrom
Nutritionsupportinadults[NICEclinicalguideline32]).
61.Providenutritionsupporttopeoplewithdysphagiainline
withrecommendationsinNutritionsupportinadults(NICE
clinicalguideline32)andStroke(NICEclinicalguideline68).
Relativevaluesofdifferent
outcomes
Theoutcomesreportedinthestudiesincluded:returntonormaldiet,
occurrenceofchestinfectionandaspirationpneumonia.
Intheshorttermthepreventionofaspirationpneumoniaisacritical
outcome,butinthelongtermareturntoanormaldiethasasignificant
impactonqualityoflifeforbothpatientsandcarers.Dysphagiamay
resultinpercutaneousendoscopicgastrostomy(PEG)feeding,which
mayhaveasignificantnegativeimpactonqualityoflifeaswellas
significantlyincreasedcosts.
Tradeoffbetweenclinical
benefitsandharms
Untreateddysphagiacouldleadtoseriouscomplicationsincluding:
aspirationpneumonia,dehydrationanddeath.Normalswallowing
allowspeopletoenjoymealtimesandrelatedsocialinteractions,andis
thereforeconsideredtobelinkedtoanimprovementinqualityoflife.
Peoplewhoarehavingthickenedfoodmayneedassistancewithoral
hygieneandthisshouldbemonitored.TheGDGagreedthatgoodoral
hygienehasbeenlinkedwithareductioninaspirationpneumoniaand
shouldbeincorporatedintoanydysphagiamanagementplan.Thegroup
notedthatpeoplewithdysphagiahaveahigherriskofaspiration
pneumonia.
Economicconsiderations
Onedirectlyapplicablestudywithminorlimitationsshowedthatlow
intensityswallowingtherapysavesaround213perpatientcomparedto
usualcarewheninitialcostsandcostoftreatingchestinfectionsare
included.
TheGDGagreedthatthecostofprovidingswallowingtherapyfor
NationalClinicalGuidelineCentre,2013.
247
dysphagiacomparedtousualcarecouldpotentiallybeoffsetbycost
savingsduetoreductionsinchestinfectionsandimprovedoutcomesfor
patientsincludingreducedmortalityandimprovementinqualityoflife.
Qualityofevidence
TheGDGconsideredtheCarnabystudy36tobeawellconductedsingle
centredstudy,whichexaminedtheeffectsofhighandlowintensity
swallowingonreturntoprestrokedietat6monthsandaspiration
pneumoniacomparedtoacontrolgroup.Thecontrolgroupwasnot
typicalofcurrentUKpractice,wherephysiciansonlyreferredtheir
patientstothespeechandlanguagetherapistsiftheyconsideredittobe
appropriatewasnottypicalofcurrentUKpractice.Treatment,ifoffered,
consistedmainlyofsupervisionforfeedingandprecautionsforsafe
swallowing(forexample,positioning,slowedrateoffeeding).
Appraisalofthisstudygradedtheresultsfortheoutcomesreportedas
moderate.Theevidenceshowedthatasignificantlylowerproportionof
participantswhoreceivedtheswallowingtherapyexperiencedchest
infectionscomparedtousualcaregroup.Inadditionasignificantly
higherproportionofparticipantsreceivingthehighintensityswallowing
therapyreturnedtoprestrokedietat6monthscomparedtousualcare.
Thestudywasnotpoweredtocomparelowagainsthighintensity
therapy,buttherewasconsensusamongstthegroupthatthebenefitof
swallowingtherapyemployingafullrangeoftechniquesclearly
outweighedtheharmsandshouldbeofferedatleastthreetimesaweek
topatientswithdysphagia.
Itisnotpossibletorecommendthehighintensityinterventionfromthe
evidencereviewed,buttheGDGagreedthattherangeofswallowing
therapiesshouldbespecifiedandthattheminimumshouldbethelow
intensitytherapyofatleast3timesperweek,butinsomecircumstances
thehighintensitymaybemoreappropriateforthosepatientswhoare
medicallystable,abletotolerateanhouroftherapyeachdayandfollow
instructions/informationprovided.
OnesmallstudybyGaron89examinedtheeffectsofthickenedfluidsand
freeaccesstowaterontheoccurrenceofpneumoniabuttherewereno
episodesofpneumoniaineithergroup.Onthebasisofthisstudy,the
authorsreportedthattheyallowfreeaccesstowater.However,
membersoftheGDGwereawareofotherstudiesinvestigatingfree
accesstowaterbutnootherRCTdatawasavailableatpresent.The
GDGdidnotconsidertheresultsfromthisstudyweresufficientto
recommendfreeaccesstowater.Therewasuncertaintyamongstthe
GDGaboutwhethertheremaybepotentialharmsbutitwasagreedthis
wasanimportantareawhichrequiresfurtherresearch.
Otherconsiderations
Thegroupwereawareofagrowingevidencebaseofthebenefitsof
postoperativepatientsarefullyhydratedinreducinglengthofstayin
hospital.
TheGDGnotedthatpatientsshouldbeweighedregularlyandany
weightlossneedstobeexplainedandagreedthattheproblemofweight
lossmaybeduetodysphagia,butcouldalsobeattributedtoother
causessuchasdifficultiesfeedingduetoneglect,orupperlimb
weaknessordepression.
NationalClinicalGuidelineCentre,2013.
Draftforconsultation:22112012
248
12 Communication
12.1 Aphasia
Aphasiadescribesalanguagedisorderthatresultsfromdamagetoareasofthebrainresponsiblefor
differentaspectsoflanguage.Oneorseveralmodesofcommunicationincludingcomprehensionand
expressionwhichinvolvespeech,writingandgesture,maybeaffected.Beyondthedirect
impairment,aphasiaimpactsonmanyaspectsoftheindividualslifesuchasrelationships,social
engagementandindependence.Ithasbeenestimatedthatapproximatelyonethirdofstroke
survivorsareaffectedbyaphasia(DepartmentofHealth2007).
TheSpeechandLanguageTherapistsassessmentresultsinformtheaimsandobjectivesoftargeted
intervention.Thiswillhavebeennegotiatedwiththeindividualandasappropriatewiththeirfamily
orcarers.SpeechandLanguageTherapyisfocusedonimprovinganindividual'sabilityto
communicatethroughmultiplestrategiesbyaimingto:
helpthepersontouseandenhanceremainingabilities.
restorelanguageabilitiesasmuchaspossible.
compensateforlanguageproblemsbydevelopingstrategies.
learnothermethodsofcommunicating.
Coachothers(family,healthandsocialcarestaff)tolearneffectivecommunicationskillsto
maximisetheaphasicpatientscompetence.
AsearchforevidencefromsystematicreviewswascarriedoutandaCochranesystematicreview
(Bradyetal,201230)wasidentifiedforthemanagementofaphasiaanddysarthria.Thissystematic
reviewwasupdatedandrecommendationsweredrawnonthisevidence.Therewasalackofdirect
evidenceforinterventionsfordysphasia,dysarthriaandapraxiaofspeechandthereforemodified
Delphistatementsweredevelopedforthistopicareabasedonrecommendationsinpublished
nationalandinternationalguidelines(section12.3).
12.1.1
EvidenceReview:Inpeoplewhohaveaphasiaafterstrokeisspeechandlanguagetherapy
comparedtonospeechandlanguagetherapyorplacebo(socialsupportandstimulation)
effectiveinimprovinglanguage/communicationabilitiesand/orpsychologicalwellbeing?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohaveaphasiaafterstroke
andwhohavebeenassessedashavingaphasia.
Intervention
Speechandlanguagetherapy:
Comparison
Anyformoftargetedpracticetasksormethodologieswiththe
aimofimprovinglanguageorcommunicationabilitiesnot
necessarilyprovidedbyaprofessionalspeechandlanguage
therapist
Nospeechandlanguagetherapy
Placebo(socialsupportandcommunicativestimulation):
Emotional,psychologicalorcreativeinterventions(suchasart,
danceormusic),conversationorotherinformal,unstructured
communicativeinteractions.Thiscomparisondoesnotinclude
targetedtherapeuticinterventionsthataimtoresolve
participantsexpressiveorreceptivespeechandlanguage
impairments
NationalClinicalGuidelineCentre,2013.
249
ClinicalMethodologicalIntroduction
Outcomes
12.1.1.1
Functionalcommunication(languageorcommunicationskills
sufficienttopermitthetransmissionofmessageviaspoken,
writtenornonverbalmodalities,oracombinationofthese
channels)
Formalmeasuresofreceptivelanguageskills(language
understanding)
Formalmeasuresofexpressivelanguageskills(language
production)
Overalllevelofseverityofaphasiaasmeasuredbyspecialisttest
batteries(mayincludeWesternAphasiaBatteryorPorchIndexof
CommunicativeAbilities)
Psychologicalorsocialwellbeingincludingdepression,anxietyand
distress
Patientsatisfaction/carerandfamilyviews
Compliance/dropout
ClinicalEvidenceReview
AsearchwasconductedforsystematicreviewscomparingtheclinicaleffectivenessofSpeechand
LanguageTherapy(SLT)withnoSLTorplacebo(socialsupportandstimulation)toimprovelanguage,
communicationabilitiesand/orpsychologicalwellbeinginadultsandyoungpeople16orolderwho
havehadastroke.
OneCochranesystematicreview(Brady201230)thatassessedtheeffectivenessofSLTforaphasia
afterstrokewasidentified.TheCochranereviewincludedatotalof39trials(RCTs).Fromthesetrials,
weincluded12trialsmatchingourprotocol(Table64)andweinspectedanadditionaltrial107from
theCochranelistofexcludedtrials.Wedeemedthistrialsuitableforreinclusion.Weexcluded
studieswithChineselanguageoutcomemeasures(duetomajorlinguisticdifferences)andthose
withanacutestrokepopulation.
AfurtherupdatesearchwasconductedforanytrialpublishedsinceJuly2011whichwasthesearch
cutoffdateoftheincludedCochranereviewandonestudy(Palmer2012198)(Table66)was
identified.
IntheCochranesystematicreviewthefollowingstrategyofanalysiswasadopted:
TrialswereincludediftheyreportedacomparisonbetweenagroupthatreceivedSLT
intervention(providedeitherbyaspeechandlanguagetherapist,atrainedvolunteeror
computer)andagroupthatreceived:
o NoSLTintervention(Table64);or
o Socialsupportorstimulation(Table9)
SixFivetrials133241283157236randomisedparticipantsacrossthreeormoregroups(trialarms).For
thepurposeofmetaanalysis,datafromthesetrialswerepresentedandpooledwithinpaired
comparisons(seeGRADEtablesandforestplots)
Thereviewpresenteddatafromthesefivetrials133241283157236inpairedsubcomparisons.For
exampledatafromWertz1986weredividedintotwosubcomparisonsof(1)conventionalSLT
versusnoSLT(Wertz1986i283),(2)volunteeredfacilitatedSLTversusnoSLT(Wertz1986ii283)
(Table64).Otherexampleswere;Katz1997i133;Katz1997ii133;Smith1981i241;Smith1981ii241;
Wertz1986i283;Wertz1986ii283
Differentmeasurementtools(forexample,WesternAphasiaBattery(WAB),Amsterdam
NijmeganEverydayLanguageTest(ANELTA),PorchIndexofCommunicativeAbilities(PICA)
amongstothers)assessingasingleoutcomewerecombinedanddatapresentedinameta
NationalClinicalGuidelineCentre,2013.
250
analysisusingstandardisedmeandifferencesummarystatistic(seeGRADETablesandforest
plots)
ForSLTversusnoSLT,reportedfollowupassessmentsrangedfromtwomonths(Smania2006240)
to12months(MacKay1988)(seeGRADEtableandforestplots)
ForSLTversussocialsupportandstimulation,reportedfollowupassessmentsrangedfromfour
weeks(Rochon2005217)totenmonths(Hartman1987107)
Nonlanguageoutcomeswerealsoreported.Thesewereselfreportedanxiety,depressionand
hostility.
Inadditionnumberofdropoutsandnoncompliancewithtreatmentwerealsoanalysed
Forthisreview,wehaveincluded9trials(Table64)comparingSLTtonoSLTand5trials(Table9)
comparingSLTtoplacebo(socialsupportandstimulation).
Wehavealsoincludedanadditionaltrial(Hartman1987107)comparingconventionalSLTwith
emotionallysupportivecounsellingtherapy(placebo)(Table9).Itwasclassifiedasaquasi
randomisedstudybytheCochrane.Thisstudyhadseriousstudylimitationsduetopoorallocation
concealment,butweconcludedthatitwasarandomisedstudyandreincludeditintheanalysis.The
studylimitationswerethenconsideredintheGRADErating(seeGRADETable68).
Overallfunctionalcommunication,receptivelanguage,expressivelanguageandseverityof
impairmentacrossincludedtrialsaswellasthedifferentassessmenttoolsusedtomeasurethese
outcomeswereanalysed.Forthisreason,wehaveonerowrepresentingthetotaleffectand23
followingrowsforthedifferentassessmenttoolsused.
Theevidencestatementsalsoreflectthetotaleffectsaswellastheeffectsoftheassessmenttools
used.
PleaseseeAppendixMforexcludedtrials.
Table64: SpeechandLanguageTherapy(SLT)versusnoSLT
OverviewofincludedstudiesfromtheCochranesystematicreview
NUMBEROF
PARTICIPANTS
STUDIES
66
INTERVENTION
OUTCOMES
Doesborgh2004 ;
Katz1997i133;Katz1997ii
133
103participants
ComputermediatedSLT:
Improvenamingusing
computercueing
programme;computerised
languagetasksusingvisual
matchingandreading
comprehension.
Functional
communication
Receptivelanguage
skills
Expressivelanguage
skills
Severityofaphasia
Psychologicalor
socialwellbeing
includingdepression,
anxietyanddistress
Compliance/drop
out
Jufeng2005iiLincoln
1984156;Smania2006240;
Smith1981ii241;Wertz
1986i283
548participants
AllusedconventionalSLT:
Aschosenbyeachspeech
andlanguagetherapist.
NationalClinicalGuidelineCentre,2013.
251
NUMBEROF
PARTICIPANTS
STUDIES
163
INTERVENTION
MacKay1988 ;Wertz
1986ii283
179participants
VolunteerfacilitatedSLT:
SLTadministeredbytrained
volunteer(family
member/friend)withno
previoushealthcare
experience.
Smith1981i241
33participants
IntensiveSLT:
TypeofintensiveSLTnot
described,butintensivedue
tonumberandlengthof
sessionsperweek
OUTCOMES
Table65: SpeechandLanguageTherapy(SLT)versusplacebo(socialsupportandstimulation)
OverviewofincludedstudiesfromtheCochranesystematicreviewandoneadditionalstudy
(Hartman1987107)thatwasexcludedintheCochranereviewbutaddedtothisreview
NUMBEROF
PARTICIPANTS
STUDIES
56
INTERVENTION
David1982 ;Elman
199976;Hartman
1987107;Lincoln
1982iii157;Shewan
1984iii236
284participants
ConventionalSLT:
Aschosenbyeach
speechandlanguage
therapist
Rochon2005217
5participants
Sentencemapping
SLT:
Shewan1984ii236
53participants
Languageoriented
SLT:
SOCIALSUPPORT
AND
STIMULATION*
Untrained
volunteers
receiveddetails
aboutparticipants
aphasia,andwere
instructedto
stimulate
communicationto
thebestoftheir
ability.Theywere
notgiven
instructioninSLT
techniques;
participantsalso
attendedsocial
groupactivitiesof
theirchoice.
Unstructured
conversation
4levelsoftreatment: aboutcurrent
events;
active,subjectcleft,
participantswere
passive,objectcleft
givenanarrative
sentences
retellingtaskon
alternatesessions
Basedon
psycholinguistic
(psychologyof
language)principles
providedbyspeech
andlanguage
NationalClinicalGuidelineCentre,2013.
252
Basedon
stimulation
orientation,
providing
psychological
support,
communicationin
unstructured
settingscarried
OUTCOMES
Functional
communication
Receptive
languageskills
Expressive
languageskills
Severityof
aphasia
Psychologicalor
socialwellbeing
including
depression,
anxietyand
distress
Carers
perspectiveof
theparticipants
communication
Compliance/
dropout
NUMBEROF
PARTICIPANTS
STUDIES
Bowen201229
ACTNoW(Assessing
theeffectivenessof
Communication
TherapyintheNorth
West)
170participants
and135carers
INTERVENTION
therapists
Therapystarted2
weeksafterstoke
andinvolved22
contacts,for18
hours(mean),
deliveredover13
weeksinboth
hospitaland
communitysettings
byqualifiedNHSSL
therapists.
(N=85)
SOCIALSUPPORT
AND
STIMULATION*
outbynurses
OUTCOMES
19contacts,for15
hours(mean),
deliveredover13
weeksby
employedvisitors
withno
professional
experienceof
strokeorSL
therapy.Visitors
weretrainedto
deliversocial
attentionabsent
ofanyintuitive
formof
communication
therapyor
strategy.
(N=85)
*supportandcommunicativestimulation:Providedbyvolunteers.TheyweregivennoguidanceorinstructioninSLT
techniquesbutwereprovidedwithdetailedinformationontheirpatientscommunicationproblemsandwereinstructed
tostimulatecommunicationtothebestoftheirability.
Table66: OverviewofadditionalRCT(Palmer2012198)sincethesearchcutoffdateofthe
Cochranesystematicreview
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Palmer
2012198
Participantswith
strokeandaphasia
withwordfinding
difficultiesasoneof
thepredominant
features;abilityto
repeatspoken
words;nolonger
receivingspeech
andlanguage
therapy.
Usuallanguage
activitiesasforcontrol
groupplusspeechand
languagetherapy
deliveredthrough
independentuseof
computerprogram
(StepbyStep;libraryof
over13000language
exercises)supported
byavolunteer;work
throughexercisesfor
atleast20minutes3
daysaweekfor5
months
Participationin
activitiesthat
providegeneral
language
stimulation:
attendanceat
communication
supportgroupsand
conversation,
readingandwriting
activitiesthatare
partofdailylife
Percentage
improvementin
wordretrieval
ability(from
Objectand
ActionNaming
Battery)
NationalClinicalGuidelineCentre,2013.
253
Comparison:SpeechandLanguageTherapy(SLT)versusnoSLT
Table67: SpeechandLanguageTherapy(SLT)versusnoSLTStudyreferencesandsummaryoffindings
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
Noserious
indirectness
Serious
imprecision(
b)
Seesub
groupfor
means
Seesub
groupfor
means
0.28(0.03,
0.59)
SMD0.28
Low
higher(0.03
lowerto
0.59higher)
Serious
imprecision(
b)
Katz(i):
13.8(5.3)
Katz(ii):
13.8(5.3)
Katz(i):
13.7(5)
Katz(ii):
12.2(6.7)
0.14(0.40,
0.69)
SMD0.14
Low
higher(0.4
lowerto
0.69higher)
Serious
imprecision(
b)
34.3(8.4)
25.5(10.3)
0.88(0.10,
1.87)
SMD0.88
Low
higher(0.1
lowerto
1.87higher)
Wertz(i):
55.60
0.25(0.16;
0.66)
SMD0.25
Low
higher(0.16
Functionalcommunication(Betterindicatedbyhighervalues)
3
RCTsingle
blind
Seesub
groups
below(next
6rows)
Serious
limitations(a
)
Noserious
inconsistency
FunctionalcommunicationWAB(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
FunctionalcommunicationANELTA(Betterindicatedbyhighervalues)
1
Doesborgh
200466
RCTsingle
blind
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
FunctionalcommunicationFunctionalCommunicationProfile(Betterindicatedbyhighervalues)
1
Wertz
RCTsingle
blind
Serious
limitations(a
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
Wertz(i):
59.35
NationalClinicalGuidelineCentre,2013.
254
Qualityassessment
Summaryoffindings
Effect
Noof
studies
1986i283;
Wertz
1986ii283
Design
Limitations
)
Inconsistency
Indirectness
Imprecision
b)
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
(19.62)
(19.56)
Wertz(ii):
Wertz(ii):
62.05
55.60
(21.83)
(19.56)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
lowerto
0.66higher)
Seesub
groupfor
means
Seesub
groupfor
means
0.10(0.20,
0.39)
SMD0.1
Moderate
higher(0.2
lowerto
0.39higher)
Katz(i):
61.7(19.8)
Katz(ii):
61.7(19.8)
Katz(i):
58.7(25.3)
Katz(ii):
57.9(23.9)
0.15(0.40,
0.69)
SMD0.15
Low
higher(0.4
lowerto
0.69higher)
Smania:
18.2(7.65)
Wertz(i):
118.39
(41.95)
Wertz(ii):
119.89
Smania:
14.94
(10.23)
Wertz(i):
119.91
(38.48)
Wertz(ii):
0.08(0.27,
0.43)
SMD0.08
Moderate
higher(0.27
lowerto
0.43higher)
Receptivelanguage:auditorycomprehension(Betterindicatedbyhighervalues)
3
RCTsingle
blind
Seesub
groups
below(next
4rows)
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Receptivelanguage:auditorycomprehensionPICAsubtest(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
Receptivelanguage:auditorycomprehensionTokenTest(Betterindicatedbyhighervalues)
2
Smania
2006240;
Wertz
1986i283;
Wertz
1986ii283
RCTsingle
blind
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
NationalClinicalGuidelineCentre,2013.
255
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
(45.06)
119.91
(38.48)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
Seesub
groupfor
means
0.11(0.21,
0.44)
SMD0.11
Moderate
higher(0.21
lowerto
0.44higher)
Wertz(i):
75.03
(18.06)
Wertz(ii):
75.03
(18.06)
0.11(0.3,
0.52)
SMD0.11
Low
higher(0.3
lowerto
0.52higher)
Katz(i):
69.8(22.6)
Katz(ii):
69.8(22.6)
Katz(i):
69.3(20.2)
Katz(ii):
65.1(22.2)
0.12(0.42,
0.67)
SMD0.12
Low
higher(0.42
lowerto
0.67higher)
Katz(i):
79.8(14.1)
Katz(i):
66.3(21.9)
8.04(1.55,
14.52)
MD8.04
Low
higher(1.55
Receptivelanguage:readingcomprehension(Betterindicatedbyhighervalues)
2
RCTsingle
blind
Seesub
groups
below(next
4rows)
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Seesub
groupfor
means
Receptivelanguage:readingcomprehensionReadingComprehensionBatteryforAphasia(Betterindicatedbyhighervalues)
1
Wertz
1986i283;
Wertz
1986ii283
RCTsingle
blind
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
Wertz(i):
76.90
(16.97)
Wertz(ii):
77.24
(20.79)
Receptivelanguage:readingcomprehensionPICAreadingsubtest(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
Receptivelanguage:gestureusePICAGesturalsubtest(Betterindicatedbyhighervalues)
2
Katz1997i
RCTsingle
blind
Serious
limitations(a
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
NationalClinicalGuidelineCentre,2013.
256
Qualityassessment
Summaryoffindings
Effect
Noof
studies
133
;Katz
1997ii133;
Wertz
1986i283;
Wertz
1986ii283
Design
Limitations
)
Inconsistency
Indirectness
Imprecision
e)
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Katz(ii):
Katz(ii):
79.8(14.1) 68.30(23)
Wertz(i):
Wertz(i):
65.32
59.68
(19.03)
(20.98)
Wertz(ii):
Wertz(ii):
62.78
59.68
(25.67)
(20.98)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
to14.52
higher)
Receptivelanguage:gesturecomprehension(postintervention)Unnamedgesturecomprehensionassessmenttool(Betterindicatedbyhighervalues)
1
Smania
2006240
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
7.36(2.17)
8.28(1.36)
0.92(
2.19,0.35)
MD0.92
Low
lower(2.19
lowerto
0.35higher)
Receptivelanguage:gesturecomprehension(2monthsfollowup)Unnamedgesturecomprehensionassessmenttool(Betterindicatedbyhighervalues)
1
Smania
2006240
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
6.75(2.81)
7.89(1.17)
1.14(
3.23,0.95)
MD1.14
Low
lower(3.23
lowerto
0.95higher)
Noserious
indirectness
Serious
imprecision(
b)
Seesub
groupfor
means
Seesub
groupfor
means
0.2(0.27,
0.68)
SMD0.2
Low
higher(0.27
lowerto
0.68higher)
Expressivelanguage:naming(Betterindicatedbyhighervalues)
2
RCTsingle
blind
Seesub
groups
below(next
4rows)
Serious
limitations(a
)
Noserious
inconsistency
NationalClinicalGuidelineCentre,2013.
257
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
75.6(38.7)
75.7(36.7)
SMD0
Verylow
higher(0.93
lowerto
0.93higher)
Katz(i):7
(2.4)
Katz(ii):7
(2.4)
Katz(i):6.9 0.27(0.27,
(2.8)
0.82)
Katz(ii):5.5
(3.3)
Expressivelanguage:namingBostonNamingTest(Betterindicatedbyhighervalues)
1
Doesborgh
2004b66
RCTsingle
blind
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(f
)
0.00(0.93,
0.93)
Expressivelanguage:namingWABNamingsubtest(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133;
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
SMD0.27
Low
higher(0.27
lowerto
0.82higher)
Expressivelanguage:namingObjectandActionNamingBattery5monthsfollowup(Meandifferenceinchangefrombaselinebetterindicatedbyhighervalues)
1
Palmer
202198
RCTsingle
blind
Noserious
inconsistenc
y
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
N=15
N=13
N/A
Mean
difference
from
baseline
19.8(4.4
35.2)
Moderate
Expressivelanguage:namingObjectandActionNamingBattery8monthsfollowup(Meandifferenceinchangefrombaselinebetterindicatedbyhighervalues)
1
Palmer
202198
RCTsingle
blind
Noserious
inconsistenc
y
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
N=12
N=11
NationalClinicalGuidelineCentre,2013.
258
N/A
Mean
difference
from
baseline
Moderate
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
11.3(7.4
29.9)
Katz(i):
62.3(22.3)
Katz(ii):
62.3(22.3)
Wertz(i):
56.48
(18.29)
Wertz(ii):
57.41
(20.1)
Katz(i):
58.1(19.1)
Katz(ii):
50.6(24.5)
Wertz(i):
52.8
(19.48)
Wertz(ii):
52.8
(19.48)
5.28(1.33,
11.89)
MD5.28
Low
higher(1.33
lowerto
11.89
higher)
Serious
imprecision(
b)
Katz(i):
61.9(14.8)
Katz(ii):
61.9(14.8)
Katz(i):
60.4(19)
Katz(ii):
55.4(24.2)
3.88(5.75,
13.5)
MD3.88
Low
higher(5.75
lowerto
13.5higher)
Serious
imprecision(
b)
Seesub
groupfor
means
Seesub
groupfor
means
0.28(0.05,
0.61)
SMD0.28
Low
higher(0.05
lowerto
0.61higher)
Imprecision
Expressivelanguage:generalPICAVerbalsubtest(Betterindicatedbyhighervalues)
2
Katz1997i
133
;Katz
1997ii133;
Wertz
1986i283;
Wertz
1986ii283
RCTsingle
blind
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
Expressivelanguage:writtencopyingPICACopyingsubtest(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Expressivelanguage:written(Betterindicatedbyhighervalues)
2
RCTsingle
blind
Seesub
groups
below(next
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
259
Qualityassessment
Summaryoffindings
Effect
Noof
studies
4rows)
Design
Limitations
Inconsistency
Indirectness
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
Serious
imprecision(
b)
Katz(i):
66.9(23.2)
Katz(ii):
66.9(23.2)
Katz(i):
59.2(23.1)
Katz(ii):
57.9(25.3)
0.34(0.21,
0.89)
SMD0.34
Low
higher(0.21
lowerto
0.89higher)
Serious
imprecision(
b)
Wertz(i):
72.64
(16.6)
Wertz(ii):
74.86
(21.74)
Wertz(i):
68.57
(22.69)
Wertz(ii):
68.57
(22.69)
0.25(0.16,
0.66)
SMD0.16
Low
higher(0.16
lowerto
0.66higher)
Katz(i):7.3
(2.9)
Katz(ii):7.3
(2.9)
Katz(i):6.7 0.92(0.76,
(3.4)
2.61)
Katz(ii):6.1
(3.4)
MD0.92
Low
higher(0.76
lowerto
2.61higher)
Katz(i):
66.4(19.4)
Katz(ii):
Katz(i):
61.3(17.4)
Katz(ii):
SMD0.26
Low
higher(0.07
lowerto
Imprecision
Expressivelanguage:writtenPICAWritingsubtest(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Expressivelanguage:writtenPICAGraphic(Betterindicatedbyhighervalues)
1
Wertz
1986i283;
Wertz
1986ii283
RCTsingle
blind
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Expressivelanguage:repetitionWABRepetitionsubtest(Betterindicatedbyhighervalues)
1
Katz1997i
133
;Katz
1997ii133
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
Severityofimpairment:PorchIndexofCommunicativeAbility(Betterindicatedbylowervalues)
2
Katz1997i
133
;Katz
RCTsingle
blind
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
b)
NationalClinicalGuidelineCentre,2013.
260
0.26(0.07,
0.58)
Qualityassessment
Summaryoffindings
Effect
Noof
studies
1997ii133;
Wertz
1986i283;
Wertz
1986ii283
Design
Limitations
Inconsistency
Indirectness
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
66.4(19.4)
56.3(20.9)
Wertz(i):
Wertz(i):
65.65
61.66
(24.64)
(21.21)
Wertz(ii):
Wertz(ii):
67.19
61.66
(24.64)
(21.21)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
0.58higher)
3(3.2)
2.6(2.6)
0.4(0.57,
1.37)
MD0.4
High
higher(0.57
lowerto
1.37higher)
Noserious
imprecision
6.9(6.6)
6.2(5.8)
0.7(1.38,
2.78)
MD0.7
High
higher(1.38
lowerto
2.78higher)
Noserious
imprecision
2.7(2.7)
2.8(2.1)
0.1(0.9,
0.7)
MD0.1
High
lower(0.9
lowerto0.7
higher)
Imprecision
Psychosocial:MAACLAnxietyScale(MAACL)(Betterindicatedbylowervalues)
1
Lincoln
1984156
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Psychosocial:MAACLDepressionScale(MAACL)(Betterindicatedbylowervalues)
1
Lincoln
1984156
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Psychosocial:MAACLHostilityScale(MAACL)(Betterindicatedbylowervalues)
1
Lincoln
1984156
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Numberofdropouts(anyreason)
NationalClinicalGuidelineCentre,2013.
261
Qualityassessment
Summaryoffindings
Effect
Noof
studies
7
Doesborgh
200466;
Katz1997i
133
;Katz
1997ii133;
Lincoln
1984156;
Mackay
1988163;
Smania
2006240;
Smith
1981i241;
Smith
1981ii241;
Wertz
1986i283;
Wertz
1986ii283
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
Design
Limitations
Inconsistency
Indirectness
Imprecision
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
121/372
(32.50%)
122/342
(35.70%)
RR0.92
(0.76,1.11)
29fewer
per1000
(from86
fewerto39
more)
High
Noserious
indirectness
Veryserious
imprecision(
b)
7/20
(35%)
5/21
(23.8%)
RR1.47
(0.56to
3.88)
112more
per1000
(from105
fewerto
Verylow
Noncompliancewithallocatedintervention(anyreason)
1
Smania
2006240
RCTsingle
blind
Serious
limitations(c
)
Noserious
inconsistency
NationalClinicalGuidelineCentre,2013.
262
Qualityassessment
Summaryoffindings
Effect
Noof
studies
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
Design
Limitations
Inconsistency
Indirectness
Imprecision
Standard
Mean
Difference
SLT
NoSLT
(SMD)/
Mean(SD)/ Mean(SD)/ Mean
Frequency Frequency difference
(%)
(%)
(95%CI)
Unclearrandomisation;unclearallocationconcealment.Limitationswereconsideredbystudyweightsinthemetaanalysis
ConfidenceIntervalcrossesoneendofdefaultMID(0.5)
Unclearallocationconcealment
Outcomeassessorsnotblinded
ConfidenceIntervalcrossesMID(10.72)
ConfidenceIntervalcrossesbothendsofdefaultMID(0.5)
ConfidenceIntervalcrossesMID(9.74)
Heterogeneity=82%
NationalClinicalGuidelineCentre,2013.
263
Absolute
effect/
Standard
Mean
Difference
(SMD)(95% Confidence
CI)
(ineffect)
686more)
Comparison:SpeechandLanguageTherapy(SLT)versusplacebo(socialsupportandstimulation)
Table68: SpeechandLanguageTherapy(SLT)versusplacebo(socialsupportandstimulation)Studyreferencesandsummaryoffindings
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
Seesub
groupfor
means
Seesub
groupfor
means
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
FunctionalcommunicationFunctionalCommunication(Betterindicatedbyhighervalues)
2
4
rows)Davi
d1982
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
0.04(0.22, SMD0.04
0.29)
higher
(0.22lower
to0.29
higher)
High
0.10(
0.50,0.30)
SMD0.10
lower(0.50
lowerto
0.30
higher)
High
0.13(
0.20,0.47)
SMD0.13
lower(0.20
lowerto
High
FunctionalcommunicationFunctionalCommunicationProfile(postintervention)(Betterindicatedbyhighervalues)
1
David
198256
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
67(20.3)
69.2(22.4)
FunctionalcommunicationTherapyoutcomemeasuresTOMs(postintervention)(Betterindicatedbyhighervalues)
1
Bowen
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
67(20.3)
69.2(22.4)
NationalClinicalGuidelineCentre,2013.
264
Qualityassessment
Summaryoffindings
Effect
Noof
studies
201229
ACTNoW
Design
Limitations
Inconsistency
Indirectness
Imprecision
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
0.47
higher)
Confidence
(ineffect)
FunctionalcommunicationFCP(3monthfollowup)(Betterindicatedbyhighervalues)
1
David
198256
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
a)
70.4(19.1) 69(21.8)
1.4(8.01,
10.81)
MD1.4
higher
(8.01lower
to10.81
higher)
Moderate
69.3(19.6) 68(21.2)
1.3(8.07,
10.67)
MD1.3
higher
(8.07lower
to10.67
higher)
Moderate
6(12.94,
24.94)
MD6
higher
(12.94
lowerto
24.94
higher)
Verylow
FunctionalcommunicationFCP(6monthfollowup)(Betterindicatedbyhighervalues)
1
David
198256
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
a)
Receptivelanguage:auditorycomprehensionSentenceComprehensionTest(PCB)(Betterindicatedbyhighervalues)
1
Rochon
2005217
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
72(16)
66(4)
NationalClinicalGuidelineCentre,2013.
265
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
Receptivelanguage:auditorycomprehensionPictureComprehensionTest(PCB)(Betterindicatedbyhighervalues)
1
Rochon
2005217
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
78(16)
70(4)
MD8(
10.94,
26.94)
MD8
higher
(10.94
lowerto
26.94
higher)
Verylow
59(13.93)
62.83(16.13) 3.83(
18.95,
11.29)
MD3.83
lower
(18.95
lowerto
11.29
higher)
Verylow
0.87(1.7,
0.04)
MD0.87
lower(1.7
to0.04
lower)
Low
7(11.67,
MD7lower Low
Receptivelanguage:auditorycomprehensionTokenTest(Betterindicatedbyhighervalues)
1
157
Rochon
2005
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
Receptivelanguage:auditoryandwrittencomprehensionPICAGesturalsubtest(Betterindicatedbyhighervalues)
1
Lincoln
1982iii157
RCTsingle
blind
Serious
Noserious
limitation(d) inconsistency
Noserious
indirectness
Serious
imprecision(
a)
12.14(0.8) 13.01(0.87)
Expressivelanguage:singlewordsObjectNamingTest(ONT)(Betterindicatedbyhighervalues)
1
RCTsingle
Serious
Noserious
Noserious
Serious
9.83(6.32) 16.83(3.76)
NationalClinicalGuidelineCentre,2013.
266
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Lincoln
1982iii157
Design
blind
Limitations
limitation(d)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision(
a)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
(11.67to
2.33lower)
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
10(5.98)
24(6.72)
14(
20.35,
7.65)
MD14
lower
(20.35to
7.65lower)
Moderate
7(2)
5(3)
2(2.73,
6.73)
MD2
higher
(2.73lower
to6.73
higher)
Verylow
3(0.5)
3(0.63,
5.37)
MD3
higher
(0.63to
5.37
higher)
Verylow
Standard
Mean
Difference
/Mean
difference
(95%CI)
2.33)
Confidence
(ineffect)
Expressivelanguage:singlewordsWordfluency(Betterindicatedbyhighervalues)
1
Lincoln
1982iii157
RCTsingle
blind
Serious
Noserious
limitation(d) inconsistency
Noserious
indirectness
Noserious
imprecision
Expressivelanguage:sentencesCaplan&HannaTest:total(Betterindicatedbyhighervalues)
1
Rochon
2005217
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
Expressivelanguage:sentencesCaplan&HannaTest:treated(Betterindicatedbyhighervalues)
1
Rochon
2005217
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
6(2)
Expressivelanguage:sentencesCaplan&HannaTest:untreated(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
267
Qualityassessment
Summaryoffindings
Effect
Noof
studies
1
Rochon
2005217
Design
Limitations
Inconsistency
Indirectness
Imprecision
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
1(1)
2(3)
1(5.31,
3.31)
Lincoln
(iii):33.67
(22)
Rochon:
34.67
(4.04)
Lincoln(iii):
30.67(7.87)
Rochon:27
(11.31)
0.26(0.62, SMD0.26
1.15)
higher
(0.62lower
to1.15
higher)
Standard
Mean
Difference
/Mean
difference
(95%CI)
Confidence
(ineffect)
MD1lower Verylow
(5.31lower
to3.31
higher)
Expressivelanguage:picturedescriptionPicturedescription(Betterindicatedbyhighervalues)
2
Lincoln
1982iii157;
Rochon
2005217
RCTsingle
blind
Serious
limitation(e)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
Verylow
Expressivelanguage:picturedescriptionPicturedescriptionwithstructuremodelling:treateditems(Betterindicatedbyhighervalues)
1
Rochon
2005217
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
16(2.65)
14(4.24)
0.45(1.44, SMD0.45
2.33)
higher
(1.44lower
to2.33
higher)
Verylow
Expressivelanguage:picturedescriptionPicturedescriptionwithstructuremodelling:untreateditems(Betterindicatedbyhighervalues)
1
Rochon
2005217
RCTsingle
blind
Veryserious Noserious
limitation(b) inconsistency
Noserious
indirectness
Veryserious
imprecision(
c)
18.67
(3.06)
16(7.07)
NationalClinicalGuidelineCentre,2013.
268
0.41(1.46, SMD0.41
2.28)
higher
(1.46lower
Verylow
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
to2.28
higher)
Confidence
(ineffect)
Expressivelanguage:overallspokenPICAverbalsubtest(Betterindicatedbyhighervalues)
1
Lincoln
1982iii157
RCTsingle
blind
Serious
Noserious
limitation(d) inconsistency
Noserious
indirectness
Noserious
imprecision
10.52(1.2) 12.08(0.74)
1.56(
MD1.56
2.46,0.66) lower(2.46
to0.66
lower)
Moderate
7.52(1.34) 8.91(1)
1.39(
MD1.39
2.49,0.29) lower(2.49
to0.29
lower)
Low
0.07(
0.74,0.60)
Verylow
Expressivelanguage:writtenPICAgraphicsubtests(Betterindicatedbyhighervalues)
1
Lincoln
1982iii157
RCTsingle
blind
Serious
Noserious
limitation(d) inconsistency
Noserious
indirectness
Serious
imprecision(
a)
Expressivelanguage:singlewordsPICA7monthfollowupchangefrombaseline(Betterindicatedbyhighervalues)
1
Hartman
1987
RCTsingle
blind
Veryserious
limitations(b
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
a)
9.341.73
(1.28)
9.111.8
(1.37)
MD0.07
lower(0.74
lowerto
0.60
higher)
Expressivelanguage:singlewordsPICA10monthsfollowupchangefrombaseline(Betterindicatedbyhighervalues)
1
RCTsingle
blind
Veryserious
limitations(b
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
11.221.88
(1.1)
10.861.75
(1.47)
NationalClinicalGuidelineCentre,2013.
269
0.13(0.59, MD0.13
0.85)
higher
Verylow
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Hartman
1987
Design
Limitations
)
Inconsistency
Indirectness
Imprecision
a)
Noserious
indirectness
Serious
imprecision(
a)
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
(0.59lower
to0.85
higher)
Confidence
(ineffect)
Severityofimpairment:PICA(Betterindicatedbylowervalues)
1
Lincoln
1982iii157
RCTsingle
blind
Serious
Noserious
limitation(d) inconsistency
10.3(1.01) 11.43(0.67)
1.13(
MD1.13
1.91,0.35) lower(1.91
to0.35
lower)
Low
Psychosocial:CommunicationOutcomesAfterStrokescale(COAST)(6monthfollowup)(Betterindicatedbyhighervalues)
1
Bowen
201229
Bowen
2012
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
71(18)
73(18)
2.00(
8.59,4.59)
MD2.00
lower(8.59
lowerto
4.59
higher)
High
62(21)
62(18)
0.00(6.73, MD0.00
6.73)
(6.73lower
to6.73
higher)
High
Psychosocial:CarerCOAST(6monthfollowup)(Betterindicatedbyhighervalues)
1
Bowen
201229
Bowen
2012
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
NationalClinicalGuidelineCentre,2013.
270
Qualityassessment
Summaryoffindings
Effect
Noof
studies
Design
Limitations
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
Inconsistency
Indirectness
Imprecision
SLT
Mean
(SD)/
Frequency
(%)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(f
)
47/221
(21.3%)
5870/206
(34%)
RR0.69
(0.52to
0.92)
105fewer
per1000
(from27
fewerto
163fewer)
Moderate
Noserious
indirectness
Noserious
imprecision
8/216
(3.7%)
33/193
(17.1%)
RR0.21
(0.10to
0.45)
135fewer
per1000
(from94
fewerto
154fewer)
High
Confidence
(ineffect)
Numberofdropoutsforanyreason
34
Bowen
201229
ACTNoW
David
198256;
Elman
199976;
Shewan
1984ii236;
Shewan
1984iii236
RCTsingle
blind
Noserious
limitation
Noncompliancewithallocatedintervention(anyreason)
4
Bowen
201229
ACTNoW
David
198256;
Elman
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
NationalClinicalGuidelineCentre,2013.
271
Qualityassessment
Summaryoffindings
Effect
Noof
studies
199976;
Shewan
1984ii236;
Shewan
1984iii236
(a)
(b)
(c)
(d)
(e)
(f)
Design
Limitations
Inconsistency
Indirectness
Imprecision
SLT
Mean
(SD)/
Frequency
(%)
Placebo
(social
supportand
stimulation)
Mean(SD)/
Frequency
(%)
ConfidenceIntervalcrossesoneendofdefaultMID(0.5)
Unclearrandomisation,allocationconcealmentandblinding
ConfidenceIntervalcrossesbothendsofdefaultMID(0.5)
Unclearallocationconcealment
Unclearallocationconcealment.Limitationswereconsideredbystudyweightsinthemetaanalysis
ConfidenceintervalcrossesoneendofdefaultMID(0.75)
NationalClinicalGuidelineCentre,2013.
272
Standard
Mean
Difference
/Mean
difference
(95%CI)
Absolute
effect/
Standard
Mean
Difference
(SMD)or
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
StrokeRehabilitation
Communication
12.1.1.2
Economicevidence
Onestudythatincludedtherelevantcomparisonwasfound.29Thisissummarisedintheeconomic
evidenceprofilebelow(Table69andTable70).Seealsothefullstudyevidencetablein
AppendixI.Table69:
Earlyspeechtherapy(speechtherapydeliveredinthe
hospitalandcontinuedafterdischargeinthecommunity)versusnospeechtherapy
(attentioncontrol)Economicstudycharacteristics
Study
29
Bowen
Limitations
Applicability
Othercomments
Potentiallyserious
limitations(a)
Directlyapplicable
Costeffectivenessanalysisbased
onastudyincludedinour
clinicalreview.
Followup:6months.
(a) Utilityscoreswereobtainedatfollowupbutnotatbaselineastheauthorsfeltitwasnotfeasibletodoso.Therewasa
differenceinstrokeseverityatbaselinebetweenthetwogroupsmeaningthatitcannotbeassumedthatthetwogroups
wouldhavesimilarutilityscores.QALYscouldnotbecalculatedandthehealthoutcomeisjustthedifferenceinutilityat
theendoffollowup.Thescoresdontshowhowhealthstatushaschangedovertimeasthevalueisfromonetimepoint
(attheendoffollowup).Therewasalotofmissingobservationsfortheresourceuseandhealthoutcomesinboth
groups.Thismeantthatusingavailablecasedatacouldbiastheresults.Theauthorsusedmultipleimputationsto
imputemissingvaluesforparticipantswhocompletedscheduledfollowupforatleastoneoftheoutcomemeasures.
Thiswasdonetoreducetheimpactofmissingobservations.
Table70: Earlyspeechtherapy(speechtherapydeliveredinthehospitalandcontinuedafter
dischargeinthecommunity)versusnospeechtherapy(attentioncontrol)Economic
summaryoffindings
Study
29
Bowen2012
UKNHS
Increment
alcostper
patient()
Incremental
effects
(QALY)
Incremental
cost
effectiveness
(/QALY)
110(a,b)
0.005(a,b,c)
22,000(d)
NationalClinicalGuidelineCentre,2013.
273
Uncertainty
Inthedeterministicanalysis,speechand
languagetherapyisdominatedby
attentioncontrolwhentheincremental
costsandutilitiesareadjustedfor
baselinecovariates.
TheprobabilitythatSLtherapyiscost
effectiveis48%atawillingnesstopay
thresholdof20,000.
Attentioncontrolwasmorecost
effectivewhenthefollowingvariables
wereanalysed:usingtrialspecificcosts
ratherthannationalcosts,usingonly
availablecasedata,usingalternative
outcomemeasuresratherthantheEQ
5Dscores.
SLtherapywasmorecosteffectivewhen
thefollowingvariableswereanalysed:
usinganalternativeregressionmodelto
estimateincrementalcostsand
outcomes;usingtheTOMmeasureof
communicationoutcomesmeasure;
usingtheCommunicationOutcomes
AfterStrokescale(COAST)in
combinationwiththeDiscreteChoice
ExperimentweightsratherthanEQ5D
StrokeRehabilitation
Communication
Study
Increment
alcostper
patient()
Incremental
effects
(QALY)
Incremental
cost
effectiveness
(/QALY)
Uncertainty
scores.
(a) Incrementalvaluesoversixmonthsasreportedinthestudybasedontheprobabilisticresultscalculatedbyconducting
10,000simulationsontheestimatesofincrementalcostsandoutcomes.
(b) Includesmultipleimputationvaluesandareadjustedforbaselinecovariates.
(c) BasedonEQ5Ddatawascollectedfromstudyparticipantsattheendoffollowup.
(d) CalculatedbyNCGCbasedonthecostsandutilitydata.
12.1.1.3
Evidencestatements
Clinicalevidencestatements
SpeechandLanguageTherapy(SLT)versusNoSLT
Functionalcommunication
Threestudiescomprising176participantsfoundnosignificantdifferenceinfunctional
communicationbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthose
thatdidnotreceivespeechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Onestudy133i133iicomprising55participantsfoundnosignificantdifferenceinfunctional
communication(usingtheWesternAphasiaBattery(WAB)assessmenttool)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceive
speechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Onestudy66comprising18participantsfoundnosignificantdifferenceinfunctional
communication(usingtheAmsterdamNijmegenEverydayLanguageTest(ANELTA))
betweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdid
notreceivespeechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Onestudy283i283iicomprising103participantsfoundnosignificantdifferenceinfunctional
communication(usingthefunctionalcommunicationprofileassessmenttool)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceive
speechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Receptivelanguage:auditorycomprehension
Threestudiescomprising191participantsfoundnosignificantdifferenceinauditorycomprehension
skillsbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdid
notreceivespeechandlanguagetherapy(NoSLT)(MODERATECONFIDENCEINEFFECT).
Onestudy133i133iicomprising55participantsfoundnosignificantdifferenceinauditory
comprehensionskills(usingthePorchIndexofCommunicativeAbilities(PICA))betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceive
speechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Twostudies283i283ii240comprising136participantsfoundnosignificantdifferenceinauditory
comprehensionskills(usingthetokentestassessmenttool)betweentheparticipantsthat
receivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceivespeechand
languagetherapy(NoSLT)(MODERATECONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
274
StrokeRehabilitation
Communication
Receptivelanguage:readingcomprehension
Twostudiescomprising158participantsfoundnosignificantdifferenceinreadingcomprehension
skillsbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdid
notreceivespeechandlanguagetherapy(NoSLT)(MODERATECONFIDENCEINEFFECT).
Onestudy283i283iicomprising103participantsfoundnosignificantdifferenceinreading
comprehensionskills(usingthereadingcomprehensionbatteryforaphasia)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceive
speechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Onestudy133i133iicomprising55participantsfoundnosignificantdifferenceinreading
comprehensionskills(usingthePorchIndexofCommunicativeAbilities(PICA)reading
subset)betweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthose
thatdidnotreceivespeechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Receptivelanguage:gestureuse
Twostudies133i133ii283i283iicomprising158participantsshowedasignificantdifferenceingestureuse
(usingthePICAgesturalsubtest)infavourofthegroupthatreceivedspeechandlanguagetherapy
comparedtothosethatdidnotreceivespeechandlanguagetherapy(NoSLT)(LOWCONFIDENCEIN
EFFECT)
Receptivelanguage:gesturecomprehension
Onestudy240comprising33participantsfoundnosignificantdifferenceingesturecomprehension
skillsbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdid
notreceivespeechandlanguagetherapy(NoSLT)postintervention(LOWCONFIDENCEINEFFECT).
Receptivelanguage:gesturecomprehension2monthfollowup
Onestudy240comprising17participantsfoundnosignificantdifferenceingesturecomprehension
skillsbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdid
notreceivespeechandlanguagetherapy(NoSLT)at2monthfollowup(LOWCONFIDENCEIN
EFFECT).
Expressivelanguage:naming
Twostudiescomprising73participantsfoundnosignificantdifferenceinnamingskillsbetweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceivespeech
andlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Onestudy66comprising18participantsfoundnosignificantdifferenceinnamingskills(using
theBostonnamingtest)betweentheparticipantsthatreceivedspeechandlanguage
therapy(SLT)andthosethatdidnotreceivespeechandlanguagetherapy(NoSLT)(VERY
LOWCONFIDENCEINEFFECT).
Onestudy133i133iicomprising55participantsfoundnosignificantdifferenceinnamingskills
(usingtheWesternAphasiaBattery(WAB)namingtest)betweentheparticipantsthat
receivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceivespeechand
languagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Expressivelanguage:namingat5and8monthsfollowup(ObjectandActionNamingBattery)
Onestudy198comprising28participantsfoundasignificantimprovementinnamingability(usingthe
ObjectandActionNamingBattery)favouringcomputerbasedlanguagetherapyoverusualcare
(MODERATECONFIDENCEINEFFECT)at5monthsbutthisimprovementwasnolongerobservedat
the8monthfollowup(MODERATECONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
275
StrokeRehabilitation
Communication
Expressivelanguage:general
Twostudies133i133ii283i283iicomprising158participantsfoundnosignificantdifferenceinexpressive
languageskills(usingthePICAverbalsubtest)betweentheparticipantsthatreceivedspeechand
languagetherapy(SLT)andthosethatdidnotreceivespeechandlanguagetherapy(NoSLT)(LOW
CONFIDENCEINEFFECT).
Expressivelanguage:written
Twostudiescomprising158participantsfoundnosignificantdifferenceinwrittenexpressive
languageskillsbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)
andthosethatdidnotreceivespeechandlanguagetherapy(NoSLT)(MODERATE
CONFIDENCEINEFFECT).Onestudy133i133iicomprising55participantsfoundnosignificant
differenceinwrittenskills(usingthePICAcopyingandwritingsubtest)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceive
speechandlanguagetherapy(NoSLT)(LOWCONFIDENCEINEFFECT).Onestudy283i283ii
comprising103participantsfoundnosignificantdifferenceinwrittenskills(usingthePICA
graphicsubtest)betweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)
andthosethatdidnotreceivespeechandlanguagetherapy(NoSLT)(LOWCONFIDENCEIN
EFFECT).
Expressivelanguage:repetition
Onestudy133i133iicomprising55participantsfoundnosignificantdifferenceinrepetitionskills(using
theWABrepetitionsubtest)betweentheparticipantsthatreceivedspeechandlanguagetherapy
(SLT)andthosethatdidnotreceivespeechandlanguagetherapy(NoSLT)(LOWCONFIDENCEIN
EFFECT).
Severityofimpairment
Twostudies133i133ii283i283iicomprising165participantsfoundnosignificantdifferenceintheseverity
ofaphasiaimpairment(usingthePorchIndexofCommunicativeAbility)betweentheparticipants
thatreceivedspeechandlanguagetherapy(SLT)andthosethatdidnotreceivespeechandlanguage
therapy(NoSLT)(LOWCONFIDENCEINEFFECT).
Psychosocial
Onestudy156comprising137participantsfoundnosignificantdifferenceinanxiety,depressionand
hostilityscalesbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthose
thatdidnotreceivespeechandlanguagetherapy(NoSLT)(HIGHCONFIDENCEINEFFECT).
Numberofdropouts(anyreason)
Sevenstudies66133i133ii156163240241i241ii283i283iicomprising714participantsfoundnosignificant
differenceinthenumberofdropoutsbetweentheparticipantsthatreceivedspeechandlanguage
therapy(SLT)andthosethatdidnotreceivedspeechandlanguagetherapy(NoSLT)(HIGH
CONFIDENCEINEFFECT).
Noncompliancewithallocatedintervention(anyreason)
Onestudy240comprising41participantsfoundnosignificantdifferenceinthenumberofparticipants
complyingwiththeallocatedinterventionbetweentheparticipantsthatreceivedspeechand
languagetherapy(SLT)andthosethatdidnotreceivedspeechandlanguagetherapy(NoSLT)(VERY
LOWCONFIDENCEINEFFECT).
SpeechandLanguageTherapy(SLT)versusPlacebo(socialsupportandstimulation)
NationalClinicalGuidelineCentre,2013.
276
StrokeRehabilitation
Communication
Functionalcommunication
Twostudies2956comprising249participantsfoundnosignificantdifferenceinfunctional
communication(usingthefunctionalcommunicationprofileassessmenttool)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatreceivedsocialsupport
andstimulation(HIGHCONFIDENCEINEFFECT)postintervention.
Onestudy56comprising96participantsfoundnosignificantdifferenceinfunctional
communication(usingthefunctionalcommunicationprofileassessmenttool)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatreceivedsocial
supportandstimulation(HIGHCONFIDENCEINEFFECT)postintervention.
Onestudy29comprising153participantsfoundnosignificantdifferenceinfunctional
communication(usingtheTherapyOutcomeMeasureSubscale(TOM))betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatreceivedsocial
supportandstimulation(MODERATECONFIDENCEINEFFECT)postintervention.
Onestudy56comprising73participantsfoundnosignificantdifferenceinfunctionalcommunication
(usingthefunctionalcommunicationprofileassessmenttool)betweentheparticipantsthatreceived
speechandlanguagetherapy(SLT)andthosethatreceivedsocialsupportandstimulation
(MODERATECONFIDENCEINEFFECT)at3and6monthfollowup.
Receptivelanguage:auditorycomprehension
Onestudy217comprising5participantsfoundnosignificantdifferenceinauditorycomprehension
skills(usingthePhiladelphiacomprehensionbatterysentenceandpicturesubtests)betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatreceivedsocialsupport
andstimulation(VERYLOWCONFIDENCEINEFFECT).
Onestudy157iiicomprising18participantsfoundnosignificantdifferenceinauditorycomprehension
skills(usingthetokentest)betweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)
andthosethatreceivedsocialsupportandstimulation(VERYLOWCONFIDENCEINEFFECT).
Receptivelanguage:other
Onestudy157iiicomprising18participantsshowedasignificantdifferenceinauditoryandwritten
comprehensionskills(usingthePICAgesturalsubtest)infavouroftheparticipantsthatreceived
socialsupportandstimulationcomparedtothosethatreceivedSLT(LOWCONFIDENCEINEFFECT).
Expressivelanguage:singlewords
Onestudy157iiicomprising18participantsshowedasignificantdifferenceinnamingskills(usingthe
ObjectNamingTest(ONT))infavouroftheparticipantsthatreceivedsocialsupportandstimulation
comparedtothosethatreceivedSLT(LOWCONFIDENCEINEFFECT).
Onestudy157iiicomprising18participantsshowedasignificantdifferenceinnamingskills(usingthe
wordfluencytest)infavouroftheparticipantsthatreceivedsocialsupportandstimulation
comparedtothosethatreceivedSLT(MODERATECONFIDENCEINEFFECT).
Expressivelanguage:singlewords(followupmeasuresat7and10months)
Onestudy107comprising60participantsfoundnosignificantdifferenceinexpressivelanguageskills
betweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatreceived
socialsupportandstimulationat7and10monthsfollowup(VERYLOWCONFIDENCEINEFFECT).
Expressivelanguage:sentenceproduction
NationalClinicalGuidelineCentre,2013.
277
StrokeRehabilitation
Communication
Onestudy217comprising5participantsfoundnosignificantdifferenceinoverallsentenceproduction
(usingtheCaplan&Hannatest)betweentheparticipantsthatreceivedspeechandlanguagetherapy
(SLT)andthosethatreceivedsocialsupportandstimulation(VERYLOWCONFIDENCEINEFFECT).
Onestudy217comprising5participantsshowedasignificantdifferenceinsentenceproduction(using
treateditemsfromtheCaplan&Hannatest)infavouroftheparticipantsthatreceivedspeechand
languagetherapy(SLT)comparedtothosethatreceivedsocialsupportandstimulation(VERYLOW
CONFIDENCEINEFFECT).
Onestudy217comprising5participantsfoundnosignificantdifferenceinsentenceproduction(using
untreateditemsfromtheCaplan&Hannatest)betweentheparticipantsthatreceivedspeechand
languagetherapy(SLT)andthosethatreceivedsocialsupportandstimulation(VERYLOW
CONFIDENCEINEFFECT).
Expressivelanguage:picturedescription
Twostudies157iii217comprising23participantsfoundnosignificantdifferenceinpicturedescription
tasksbetweentheparticipantsthatreceivedspeechandlanguagetherapy(SLT)andthosethat
receivedsocialsupportandstimulation(VERYLOWCONFIDENCEINEFFECT).
Onestudy217comprising5participantsfoundnosignificantdifferenceinpicturedescriptiontasks
withstructuremodelling(treatedanduntreateditems)betweentheparticipantsthatreceived
speechandlanguagetherapy(SLT)andthosethatreceivedsocialsupportandstimulation(VERY
LOWCONFIDENCEINEFFECT).
Expressivelanguage:overallspoken
Onestudy157iiicomprising18participantsshowedasignificantdifferenceinoverallspokentest(using
thePICAverbalsubtest)infavouroftheparticipantsthatreceivedsocialsupportandstimulation
comparedtothosethatreceivedSLT(MODERATECONFIDENCEINEFFECT).
Expressivelanguage:written
Onestudy157iiicomprising18participantsshowedasignificantdifferenceinwrittenskills(usingthe
PICAgraphicsubtest)infavouroftheparticipantsthatreceivedsocialsupportandstimulation
comparedtothosethatreceivedSLT(LOWCONFIDENCEINEFFECT).
Severityofimpairment
Onestudy157iiicomprising18participantsshowedthatparticipantsthatreceivedsocialsupportand
stimulationweresignificantlylessimpairedasaresultofaphasia(usingtheshortenedPICA)
comparedtothosethatreceivedSLT(LOWCONFIDENCEINEFFECT).
Psychosocial:CommunicationOutcomesafterStrokescale(COAST)
Onestudy29comprising117participantsfoundnosignificantdifferenceintheCommunication
OutcomesAfterStrokescale(COAST))betweentheparticipantsthatreceivedspeechandlanguage
therapy(SLT)andthosethatreceivedsocialsupportandstimulation(HIGHCONFIDENCEINEFFECT)
at6monthfollowup.
Psychosocial:CarerCommunicationOutcomesafterStrokescale(COAST)
Onestudy29comprising129participantsfoundnosignificantdifferenceintheCarerCommunication
OutcomesAfterStrokescale(COAST)betweentheparticipantsthatreceivedspeechandlanguage
therapy(SLT)andthosethatreceivedsocialsupportandstimulation(HIGHCONFIDENCEINEFFECT)
at6monthfollowup.
Numberofdropouts(anyreason)
NationalClinicalGuidelineCentre,2013.
278
StrokeRehabilitation
Communication
Fourstudies29,5676236ii236iiicomprising427participantsshowedthatparticipantswhoreceivedSLT
weresignificantlylesslikelytodropoutcomparedtothosewhoreceivedsocialsupportand
stimulation(MODERATECONFIDENCEINEFFECT).
Noncompliancewithallocatedintervention(anyreason)
Fourstudies29,5676236ii236iiicomprising409participantsshowedthatparticipantswhoreceivedSLT
weresignificantlymorecompliantwiththeallocatedinterventioncomparedtothosewhoreceived
socialsupportandstimulation(MODERATECONFIDENCEINEFFECT).Economicevidencestatements
Acosteffectivenessstudydirectlyapplicableandwithpotentiallyseriouslimitationsshowsthatthe
incrementalcosteffectivenessratioofspeechandlanguagetherapycomparedtoother
interventions(attentioncontrol)isaround22,000perQALYgained.Howeverthereisahigh
uncertaintyaroundthisestimate.
12.2 Dysarthria
Dysarthriaismotorspeechdisorder,characterisedbyslowslurred,imprecisespeechandquietvocal
volume.Thecommoneffectofthesesymptomsisanimpactonintelligibility,makingcommunication
difficult.Thisinturncanaffectsocialinteraction,employmentandfeelingsofsocialstigmatisation65.
12.2.1
EvidenceReview:Inpeopleafterstrokeisspeechandlanguagetherapycomparedto
socialsupportandstimulationeffectiveinimprovingdysarthria?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention
Speechandlanguagetherapy:
Anyformoftargetedpracticetasksormethodologieswiththe
aimofimprovinglanguageorcommunicationabilities
Comparison
Socialsupportandcommunicativestimulation:Emotional,
psychologicalorcreativeinterventions(suchasart,danceor
music),conversationorotherinformal,unstructured
communicativeinteractions.Thiscomparisondoesnotinclude
targetedtherapeuticinterventionsthataimtoresolve
participantsexpressiveorreceptivespeechandlanguage
impairments
Outcomes
Measuresoffunctionalcommunication
Formalmeasuresofreceptivelanguageskills(language
understanding)
Formalmeasuresofexpressivelanguageskills(language
production)
Psychologicalorsocialwellbeingincludingdepression,anxietyand
distress
FrenchayDysarthriaAssessment.
Measuresofarticulation(range,speed,strength,andco
ordination)
Perceptualmeasuresofvoiceandprosody(forexample,Vocal
ProfileAnalysis)
Acousticmeasures(forexample,fundamentalfrequency,pitch
perturbation(jitter),amplitudeperturbation(shimmer),etc.as
measuredby,forexample,computerisedsoundspectrography)
NationalClinicalGuidelineCentre,2013.
279
StrokeRehabilitation
Communication
12.2.1.1
ClinicalEvidenceReview
AsearchwasconductedforsystematicreviewscomparingtheclinicaleffectivenessofSpeechand
LanguageTherapy(SLT)withsocialsupportandstimulationtoimprovedysarthriainadultsand
youngpeople16orolderwhohavehadastroke.
OneCochranesystematicreview(Sellars2005235)thatassessedtheeffectivenessofspeechand
languagetherapyfordysarthriawasidentified(Table71).Trialswereconsiderediftheyreporteda
comparisonbetweenagroupthatreceivedSLTinterventionandagroupthatreceived(1)noSLT
interventionand(2)aninterventionundertakenbynonSLTpersonnel,forexample,deliveredby
volunteers(i.e.,SLTversusnonSLT).Sixteentrialswereconsideredforinclusionbutrejectedfrom
thereview.
AfurthersystematicupdatesearchwasconductedforanytrialpublishedsinceSeptember2004
whichwasthesearchcutoffdateoftheincludedCochranereviewandonestudy(Bowen201229
ACTNoWstudy)wasidentified.
Table71: OverviewofadditionalRCT(Bowen201229)sincethesearchcutoffdateofthe
Cochranesystematicreview
STUDY
POPULATION
INTERVENTION
COMPARISON
Bowen
201229
ACTNoW
(Assessing
the
effectiveness
of
Communicat
ionTherapy
intheNorth
West)
Onehundredand
seventyadultswith
aphasia,dysarthria
orbothadmittedto
hospitalwithstroke.
Sixtysixparticipants
haddysarthria.
Participantsranged
inagefrom32to97
years(mean70
years)
Therapystarted2
weeksafterstokeand
involved22contacts,
for18hours(on
average),delivered
over13weeksinboth
hospitaland
communitysettingsby
qualifiedNHSSL
therapists.
Interventionwas
tailoredtoindividual
needsandabilities
(Outcomeinformation
availableonN=33
participantswith
dysarthria)
19contacts,for15
Functional
hours(onaverage),
communicative
deliveredover13
abilityonthe
weeksbyemployed
Therapy
visitorswithno
Outcome
professional
Measure
experienceofstroke
activitysubscale
orSLtherapy.
(TOM)
Visitorsweretrained
todeliversocial
attentionwhich
involvedgeneral
conversation,
involvingthepatient
invariousactivities
(reading,watching
televisionorvideos,
playingaselectionof
games)
(Outcome
information
availableonN=27
participantswith
dysarthria)
NationalClinicalGuidelineCentre,2013.
280
OUTCOMES
StrokeRehabilitation
Communication
Comparison:SpeechandLanguageTherapy(SLT)versussocialsupportandstimulation
Table72: SpeechandLanguageTherapy(SLT)versussocialsupportandstimulationStudyreferencesandsummaryoffindings
Qualityassessment
Summaryoffindings
Noof
studies
Social
supportand
SLT
stimulation
Mean(SD) Mean(SD)
Effect
Design
Limitations
Inconsistency
Indirectness
Imprecision
Mean
difference
(95%CI)
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
FunctionalcommunicationTherapyOutcomeMeasureactivitysubscale(TOM)(6monthfollowup)(Betterindicatedbyhighervalues)
1
Bowen
201229
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Serious
indirectness(
a)
Veryserious
imprecision(
b)
3.1(1.4)
3.1(1.7)
0.00(0.80, MD0.00
0.80)
(0.80lower
to0.80
higher)
(a) Amixedpopulationofconsistingofpeoplewithdysarthriaaloneandalsopeoplewithdysarthriaaswellasaphasia
(b) ConfidenceIntervalcrossesbothendsofdefaultMID(0.5)
NationalClinicalGuidelineCentre,2013.
281
Verylow
StrokeRehabilitation
Communication
12.2.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationswereincludedinthereview.TheeconomicanalysisintheACT
NoWstudy29wassetuptocoverbothpeoplewithaphasiaanddysarthria;however,sincethe
majorityofthepatientsintheeconomicanalysishadaphasia(90%),thisstudywasincludedinthe
reviewforaphasia(see12.1.1)butconsiderednotapplicabletopeoplewithdysarthria(39%ofthe
populationinthepaperhadeitherbothdysarthriaandaphasiaordysarthriaalone)asthetreatment
wouldbedifferentfromtheoneusedinthestudy.
Economicconsiderations
Theestimatedcostofaband6speechandlanguagetherapistis47perhourofclientcontactq.
12.2.1.3
Evidencestatements
Clinicalevidencestatements
SpeechandLanguageTherapy(SLT)versussocialsupportandstimulation
Functionalcommunication:TherapyOutcomeMeasureactivitysubscale(TOM)
Onestudy29comprising66participantswithdysarthriafoundnosignificantdifferenceinfunctional
communication(usingtheTherapyOutcomeMeasureactivitysubscale(TOM))betweenthe
participantsthatreceivedspeechandlanguagetherapy(SLT)andthosethatreceivedsocialsupport
andstimulationat6monthfollowup(VERYLOWCONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
12.2.2
Recommendationsandinktoevidence
AphasiaandDysarthria
62.Screenpeopleafterstrokeforcommunicationdifficultieswithin
72hoursofonsetofstrokesymptoms.
63.Eachstrokerehabilitationserviceshoulddeviseastandardisedprotocol
forscreeningforcommunicationdifficultiesinpeopleafterstroke.
64.Provideappropriateinformation,educationandtrainingtothe
multidisciplinarystroketeamtoenablethemtosupportand
communicateeffectivelywiththepersonwithcommunication
difficultiesandtheirfamilyorcarer.
65.Speechandlanguagetherapyforpeoplewithstrokeshouldbeledand
supervisedbyaspecialistspeechandlanguagetherapistworking
collaborativelywithotherappropriatelytrainedpeopleforexample,
q EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
50
reportandAgendaforChangesalaryband6 (typicalsalarybandidentifiedbyclinicalGDGmembers).
NationalClinicalGuidelineCentre,2013.
282
StrokeRehabilitation
Communication
speechandlanguagetherapyassistants,carersandfriends,and
membersofthevoluntarysector.
66.Provideopportunitiesforpeoplewithcommunicationdifficultiesafter
stroketohaveconversationandsocialenrichmentwithpeoplewho
havethetraining,knowledge,skillsandbehaviourstosupport
communication.Thisshouldbeinadditiontotheopportunitiesprovided
byfamilies,carersandfriends.
67.Speechandlanguagetherapistsshouldassesspeoplewithlimited
functionalcommunicationafterstrokefortheirpotentialtobenefitfrom
usingacommunicationaidorothertechnologies(forexample,home
basedcomputertherapiesorsmartphoneapplications).
68.Providecommunicationaidsforthosepeopleafterstrokewhohavethe
potentialtobenefit,andoffertraininginhowtousethem.
69.Tellthepersonwithcommunicationdifficultiesafterstrokeabout
communitybasedcommunicationandsupportgroups(suchasthose
providedbythevoluntarysector)andencouragethemtoparticipate.
70.Whenpersistingcommunicationdifficultiesareidentifiedatthepersons
6monthorannualstrokereviews,referthembacktoaspeechand
languagetherapistfordetailedassessment,andoffertreatmentifthere
ispotentialforfunctionalimprovement.
71.Makesurethatallwritteninformation(includingthatrelatingto
medicalconditionsandtreatment)isadaptedforpeoplewithaphasia
afterstroke.Thisshouldinclude,forexample,appointmentletters,
rehabilitationtimetablesandmenus.
Relativevaluesof
differentoutcomes
TheGDGrecognisedthatoutcomesinspeechandlanguagetherapycouldlookat
bothimpairmentandfunction.Whileimprovementinfunctionistheultimateaim,
smalltargetedstudiesmaydetectimpairmentchangesmoreeasily.
SomemembersoftheGDGquestionedtheresponsivenessoftheFunctional
communicativeabilityontheTherapyOutcomeMeasureactivitysubscaleTOM.(ACT
NOW)toreflectclinicallymeaningfulchange,butwerenotawareofanypublications
reportingresponsivenesswiththisinstrument.Itwasqueriedwhetherstepscouldbe
consideredequal(i.e.achangefrom0to1isfunctionallythesamechangeasone
from3to4).Itwasfeltthatitwasnotclearhowyoumovedfromonepointto
anotherwithinthescaleanditmaybeopentointerpretation.
TheCommunicationOutcomesAfterStrokescale(COAST)whichwasdevelopedfor
useintheACTNoWstudyhasreportedvalidityandreliabilitydataforuseinthe
study.TheGDGagreedthewaythishadbeendevelopedwasreasonable.
Tradeoffbetween
clinicalbenefitsand
harms
TheGDGwerenotawareofanypotentialharmsfromlanguagebasedtherapies.
Howevertheimportanceofregularreviewtoreassesspeoplewithcommunication
difficultieswasnoted,andtheGDGagreedtheguidanceof6monthandthenannual
reviewasgivenintheNationalStrokeStrategyshouldberecommended.61
Economic
considerations
TheGDGnotedthatSLTiscurrentlyroutinelyprovidedintheNHStopeoplewith
aphasia.Thecostofaband6or7speechandlanguagetherapistis47or57per
NationalClinicalGuidelineCentre,2013.
283
StrokeRehabilitation
Communication
hourofclientcontact. TheACTNoWstudyshowedthatthereisahighuncertainty
overthecosteffectivenessofspeechandlanguagetherapycomparedtoattention
control.
Inthisstudypatientsreceivedspeechandlanguagetherapyonaveragefor18hrsand
22contactsover13weeksperpatient.Thiswasconsideredtobealowlevelof
intensitybytheGDG.Atthislevelofintensitythereisuncertaintythatitiscost
effectiveoverandabovepaidvisitorsandahigherlevelofintensitycouldmakethe
interventionmorecosteffective.
Basedontheseconsiderationsandonthelevelofimprovementofpatientswho
receivedtheintervention,theGDGfeltthatthebenefitsofSLTarelikelytooutweigh
thecosts.
Therearehighcostsassociatedwithprovidingcommunicationaidsassomesoftware
areexpensive(upto2,000)andtrainingpeopletousethemwouldrequirearound
threemonths.Howeverotherapplicationsareavailablefromsmartphonesandthey
arenotveryexpensive.TheGDGhasdecidedthatgiventhepotentialhighcosts
involved,communicationaidsshouldbeprovidedonlytothosepeoplewhoarelikely
tobenefit.
Qualityofevidence
TheCochranesystematicreviewincludedstudiesiftheyreportedacomparison
betweenagroupthatreceivedSLTinterventionandagroupthatreceivednoSLTor
socialsupportorstimulation.Eightstudiesthatmatchedtheprotocolwereincluded
forSLTversusnoSLTandfiveforSLTversussocialsupportorstimulationplusan
additionalstudynotincludedintheCochranereview.
Themajorityofstudieswerefoundtobesmall,poorlydescribed,oldandestimated
effectsformostoftheoutcomesspecifiedwerenonsignificant.Theconfidencein
theeffectofspecifiedoutcomesrangedfromverylowtohighwiththemajority
beinglow.Outcomesweredowngradedforstudylimitations,inconsistencyand
imprecision.
Nosignificantdifferencewasfoundforanyofthefunctionalcommunication
outcomesinbotharms.SpeechandLanguageTherapy(SLT)showedsignificant
improvementinreceptivelanguagecomparedtonoSLTwhilesocialsupportand
communicativestimulationshowedsignificantimprovementinreceptivelanguage,
expressivelanguageandinreducingtheseverityofimpairment.Theimprovements
seenwithsocialsupportandcommunicativestimulationwereallfromonestudy
(Lincoln1982iii157)comprisingof18participants.TheGDGnotedthattheresult
shouldbereadwithcautionasitlacksgeneralisabilityasitisnotatrue
representationoftheaphasicpopulation.
TheGDGnotedthatcurrentpracticehaschangedsincemanyofthestudieswere
published,andthatthedetailsoftheinterventionwasnotprovidedorpoorly
describedinmanyofthestudies.ThismadeitdifficultfortheGDGtodraw
meaningfulconclusions.Thegroupdidagreethattheresultshighlightthebenefitsof
supportedcommunicationanditwasagreedthatSLThasbecomebroaderthanis
reflectedinthestudiesandwouldnowincludesupportandstimulation.
AmorerecentwellconductedstudywastheACTNOWstudydeliveredonaverage
18hrsoftherapyover13weeksinbothhospitalandcommunitysettings.The
interventionswerelargelydeliveredbyband6speechandlanguagetherapists,
althoughpatientsimproved,asimilarlevelofgainwasachievedbypaidvisitors.
Therewere6elementstothespeechandlanguagetherapyinterventions:
assessment,informationprovision,provisionofcommunicationmaterials,carer
contact,indirectcontact(includingdiscwithclinicalteamsandgoalsetting),and
directcontact(includingimpairment,activityandparticipationskills).However,the
outcomemeasurefocussedonfunctionalcommunicationwhichwasonlyoneaspect
oftheSpeechandLanguageTherapistsintervention.TheGDGnotedthatthetrial
wasconductedonparticipantssoonafteronsetofstroke(2weeks),andthatthis
doesnotreflectallstagesofstrokerecovery.Peoplemaymakechangeatalater
stageofrecoveryandthisstudydoesnotreflecttheentirescopeofSLTinput.
TheACTNoWstudywasawellconductedstudyofspeechandlanguagetherapy
NationalClinicalGuidelineCentre,2013.
284
StrokeRehabilitation
Communication
comparedwithpaidvisitors,earlyafterstroke.TheGDGnotedanumberofareas
thatmightinfluenceclinicalinterpretationincludingtherelativelylowdoseof
therapyandthepowerofthestudy.
Dysarthria
IntheACTNoWstudy39%oftheparticipantshaddysarthria,and29%hadboth
aphasiaanddysarthria,andasubanalysisforthetherapyoutcomemeasureactivity
subscaleoutcome(TOM)wasconducted.Nosignificantdifferencewasfound
betweentheSLTgroupandthesocialsupportandstimulationgroup.TheGDG
discussedtheuseofafunctionalinterventioninthispopulationandagreedthatfor
thissubgrouptheinterventionwouldinitiallybeimpairmentbasedandfunctional
communicationtherapywouldbegivenlaterinthepersonsrehabilitation.Itwas
feltthatthisanalysisdidnotprovideanyusefulinformationtoprovidespecific
guidanceforthisgroup.
Other
considerations
TheGDGwasawarethatpracticehaschangedinthepast30years.Thefindings
relatingtothebenefitsofsocialsupportinterventionsshouldnotbeoverlookedin
deliveringservices.Theincreasinguseofcommunicationtoolssuchascomputerand
smartphonetechnologieswasalsonoted.
TheGDGwereawareoftheCollegeofSpeechandlanguagetherapistsaphasia
commissioningdocument,whichprovidesanoverviewofthevarioustherapies.223.
ThegroupexpressedconcernthattheACTNoWstudyshouldnotbeover
interpretedandstressedthecentralroleofSpeechandLanguagetherapyinthe
organisation,assessmentandtreatmentofcommunicationdifficultiesoverthe
wholestrokepathway,shouldberecognised.Itwasagreedthatitwasimportantto
starttherapyassoonaspossibleafterstrokeandthatanassessmentof
communicationshouldbeundertakenwithin72hoursofadmissionandthiswas
currentlyusualpractice.Itwasacknowledgedthatscreeningwouldbeundertaken
onadmissiontoanacuteunit;howevernoparticularscreeningtoolcouldbe
recommended.Thestudieshaveemphasisedtheimportanceofsocialsupportand
providinganenrichedenvironmentindeliveryofSLTandtheGDGacknowledged
trainedvolunteersplayanimportantroleinprovidingthis.TheGDGagreedthat
therapyshouldbemanagedandledbyaspecialistspeechandlanguagetherapist.
Thepatientrepresentativehighlightedthatforpeoplewithdysarthriaguidanceon
helpwithmovementofthetongueisimportantandthiscouldonlybeprovideda
SLTtherapist.TheGDGalsonotedthatthenumberofdropoutsandnon
compliancetoallocatedinterventionwaslowerintheSLTgroupcomparedtosocial
supportandintervention.
TheGDGalsoconsideredthequalitativecomponentoftheACTNoWstudy.Inthis
sectionparticipantsandcarersviewsandexperiencesofSLTorvisitorsupport
whereevaluated.TheGDGfeltthatitwasimportantthatparticipants/carersvalued
outsidecontactregardlessofwhetheraSLTorvisitor.Personalqualitiesof
SLT/visitorswerehighlightedbyparticipants(forexampleputtingpeopleatease,
abilitytomakeparticipantsfeelimportantetc.).MembersoftheGDGfeltthatitwas
alsoimportantthatpatientsexperienceddifferentaspectsoftheirmeetingswith
SLTs/visitorasmeaningful,suchasthosewhohadtheSLThighlightedexplicit
strategiesthatwerehelpfultobuildconfidence;whereasthosewhohadvisitors
valuedthesocialengagementprocessesandeverydaypracticeaspectofthis
contact.
Itwasthereforenotedthatopportunitiestoengagewithcommunicationpartners
shouldplayapartintherehabilitationofpeoplewhohavehadastrokeandhave
languageimpairmentandthatprofessionalsshouldaimtoprovidesuchcontacts.
NationalClinicalGuidelineCentre,2013.
285
StrokeRehabilitation
Communication
12.3 Speechandlanguagetherapiesfordysarthriaandapraxiaofspeech
Therewasalackofdirectevidencefordysphasia,dysarthriaandapraxiaofspeech(sections12.2and
12.3).ThereforerecommendationsinthesesectionswerebasedonmodifiedDelphiconsensus
statements(basedonrecommendationsinpublishednationalandinternationalguidelines).Below
weprovidetablesofstatementsthatreachedconsensusandstatementsthatdidnotreach
consensusandgiveasummaryofhowtheywereusedtodrawuptherecommendations.Fordetails
ontheprocessandmethodologyusedforthemodifiedDelphisurveyseeAppendixF.
12.3.1
12.3.2
Whatinterventionsimprovecommunicationinpeopledysphasia,dysarthriaandapraxia
ofspeech?
Population
Adultsandyoungpeople16orolderwhohavehadastrokeandwhohavespeechand
languageimpairments
Components
8. Assessment
9. Speechandlanguagetherapies
10. Communicationaids
Outcomes
11. Qualityoflife
12. Communicationskills
13. Socialparticipation
Delphistatementswhereconsensuswasachieved
Table73: Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
1.
Results
%
Statement
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Forallpeoplewithspeechand
languageimpairmentstheSpeechand
LanguageTherapistneedstoexplain
anddiscusstheimpairmentwiththe
personwhohashada
stroke/family/carers/treatmentteam
andteachthemhowtomanagethe
condition.
78.6
3/28(11%)panelmembers
commented
Onepersoncommentedthatthis
doesnotneedtobecarriedoutbya
speechandlanguagetherapistas
longasitisundertheguidanceof
one.
Carerinvolvementwasalso
highlighted.
Onepersonexpressedsurprisethat
CommunicationSupportServices
werenotincludedinthewhole
speechandlanguagesection.
Earlyafterstrokethepersonwitha
speechandlanguageimpairment
shouldbefacilitatedtocommunicate
everydayneedsandwishes,and
supportedtounderstandand
93.1
4/29(14%)panelmembers
commented
Itwascommentedthatthereare
NationalClinicalGuidelineCentre,2013.
286
StrokeRehabilitation
Communication
Number
12.3.3
Statement
participateindecisionsaround,for
example,medicalcare,transfertothe
community,andhousing.Thismay
needalternativeandaugmentative
formsofcommunication.
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
interactionswithcognitionand
emotionandthereforeinputfrom
otherMDTmembersmaybe
needed.
ItwasstatedthatAACmaybelow
techandsimplepaperandpenor
highertechIpadappscouldbe
used.
Onecommentwasthatthisdepends
onthepersonsindividual
assessment,readinesstoparticipate
andhis/herstatedgoals.
Trainingforsomemembersofthe
MDTmayalsobenecessary.
2.
Peoplewhohavehadastrokeand
whohavepersistingspeechand
languagedeficitsshouldbeassessed
foralternativemeansof
communication(gesture,drawing,
writing,useofcommunicationaids).
73.1
2/26(8%)panelmembers
commented
Onepersonstatedthatmixing
peoplewithlanguageandthosewith
speechimpairmentstogetherisnot
appropriateinthisstatement.
Theotherpersonthoughtthatthis
statementwastooobvioustobe
useful.
3.
Theimpactofspeechandlanguage
impairmentsonliferolesforexample
family,leisure,work,etc.shouldbe
assessedandpossibleenvironmental
barriers(forexamplesigns,attitudes),
shouldbeaddressed,jointlywiththe
MDT.
81.5
3/27(11%)panelmembers
commented
Onepersonpointedoutthatthis
wouldnothappenintheacutestage
ofrehabilitation.
Anotherpersonthoughtthatit
shouldalsoinvolvefamilyand
friends,employersandrelevant
otheragencies
Athirdpersonindicatedthat
addressedwasnotclear.
Delphistatementwhereconsensuswasnotreached
Table74: Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
1.
Statement
Thekeyaimofspeechandlanguage
therapyearlyafterstrokeshouldbeto
Results
%
Amountandcontentofpanel
commentsorthemes
55.0
Inround217/27(63%)panel
memberscommented;11/20(55%)
NationalClinicalGuidelineCentre,2013.
287
StrokeRehabilitation
Communication
Number
2.
Statement
minimisethecommunication
impairment.
Thelistofapproachesthatmaybe
usedwithapatientwhoisdysphasic:
Picturecards
Drawings
SoundBoards
Writing
Phonologicalsoundcueing
Modellingwords
Sentencecompletion
Melodicintonationtherapy
Neurolinguisticapproach
Computerisedapproach
Results
%
Amountandcontentofpanel
commentsorthemes
inround3
Panelmembersthoughtthatthere
aremanyfacetstotheaimsof
speechandlanguagetherapythat
werenotcapturedbythis
statement.Suchas:
Todealwiththeimpactofthe
communicationimpairment
Toassessandeducateregardingthe
extentofthedifficulty
Toaddressthepersonsconfidence,
Toenhanceskillsofcommunication
partners
Toremovebarriersto
communication
Extract:
Thiscanbeverybroadlydefined.
Minimisingthecommunication
impairmentisnotnecessarilyjust
reducingtheactualimpairment.It
maybeprovidingadviceand
informationwhichenhances
understandingandindirectly
minimisestheproblem,itmaybe
usingstrategiestofacilitate
communication,itmaybeproviding
facilitatedemotionalsupportto
reducetraumawhichcanenhance
communication.
36.8
42.1
12.5
33.3
27.7
22.2
33.3
5.8
27.7
38.8
Inround220/27(74%)panel
memberscommented;11/19(58%)
inround3
Somefurtherapproacheswere
suggested:
Talkingmats
Semanticcueing
Gesture
Cognitiveneuropsychological
approaches
Constraintinducedtherapy(which
usespicturecards)
Augmentativeandalternative
communication
Onepersonhighlightedthatany
approachneedstobeevidence
based.
Itwasalsohighlightedthatthe
statementimpliesafocuson
NationalClinicalGuidelineCentre,2013.
288
StrokeRehabilitation
Communication
Number
3.
Statement
Thelistofapproachesthatmightbe
usedwithapatientwhoisdysarthric:
Oralmuscularexercises
Monitoringrateofspeechproduction
Pausing
Alphabetsupplementation
Results
%
Amountandcontentofpanel
commentsorthemes
languageimpairmentratherthan
focusontheskillsandcompetence
ofthepersonwhohasdysphasiaand
thoseintheircommunicative
environment.Thispanelmember
suggestedthefollowingapproaches
todothis:
Informationandsupportforthe
personandtheir
family/friends/serviceproviders
(andalsoaboutlanguage
strengths)
Trainingofconversationpartners
Accesstopeersupport
Othersspecificallyfavoured
impairmentbasedapproaches.
Itwashighlightedthatthiswould
varyfrompersontoperson(The
SpeechandLanguageTherapist
wouldmakeacommunicationbook
tailoredtotheindividualratherthan
alphabetchartand/ortalkingmats
toaiddiscussion).
21.7
34.7
26.0
27.2
Inround212/24(47%)panel
memberscommented;14
commentedinround3(freetext
prompt)
Thefollowingapproacheswere
suggested:
Initiationofvocalisation(exercises
forarticulation)
Coordinationbetweenbreathing
andspeech(breathingsupport
exercises)
Sustainingvoiceduringspeech
production
Pacing
Gesture
Adviceaboutcondition,andtraining
ofconversationpartners
Computertherapy
Writing/drawing
Augmentativeandalternative
communication
Compensatoryslowingofspeech
ratewithexaggerated
articulation
NationalClinicalGuidelineCentre,2013.
289
StrokeRehabilitation
Communication
Number
Statement
Results
%
Amountandcontentofpanel
commentsorthemes
Onepersonhighlightedthatthis
shouldbeafocusedapproachbased
onassessment(thesystemthatis
mostcompromisedwouldbe
targetedforexamplerespiration,
palatalmovement,voice,
articulation,rateofspeech,phrasing,
intonation).
4.
Listofapproachesthatmightbeused
withapatientwhohas
dysarthrophonia:
Biofeedback
Voiceamplifier
Intensetherapytoincreaseloudness
12.5
11.7
0.0
Inround29/23(39%)panel
memberscommented;7/17(41%)in
round3
Noclearapproachesweresuggested.
Itwasstatedthatthisdependson
thepatientspresentationand
severity.
Itwasalsohighlightedthatthisisa
rareproblemandthatthereisno
evidencetosupportaparticular
approach.
5.
Thelistofapproachesthatmightbe
usedwithapatientwhohas
articulatorydyspraxia:
Cognitivelinguistictherapy
Repetitivedrills
Auditoryinput/analysis
Automaticspeech
Singing
Phonemiccueing
Wordimitation
Computerprogrammes
Varleyapproach
AAC(AugmentativeandAlternative
Communication)readingaloud
Distractionpracticewithfeedback
Phonememanipulationtasks
SegmentbySegmentapproach
SWORD(computersoftware)
Prosodictherapy
20.0
41.6
33.3
38.4
8.3
16.6
8.3
38.4
27.2
18.1
11.1
9.0
18.1
27.2
25.0
Inround216panelmembers
commented(freetextprompt);
8/14(57%)inround3
Nofurtherapproacheswere
suggestedanditwashighlightedthat
anyapproachneedstobeevidence
based.
6.
Anypatientwithseverearticulation
difficulties(<50%intelligibility)
reasonablecognitionandlanguage
functionshouldbeassessedforand
providedwithalternativeor
augmentativecommunicationaids.
61.1
Inround23/25(12%)panel
memberscommented;3/18(17%)in
round3
Itwoulddependonstageofrehab,
successofrehabilitationand
prognosis.
Onepersonobjectedtoalevel(i.e.
below50%intelligibility)being
NationalClinicalGuidelineCentre,2013.
290
StrokeRehabilitation
Communication
Number
12.3.4
Results
%
Statement
Amountandcontentofpanel
commentsorthemes
stated(asitmaybedifferentfor
eachpatientandintelligibilitymay
dependonfamiliaritywiththe
patient.
RecommendationsandlinkstoDelphiconsensussurvey
Statements
28.ForallpeoplewithspeechandlanguageimpairmentstheSpeechand
LanguageTherapistneedstoexplainanddiscusstheimpairmentwith
thepersonwhohashadastroke/family/carers/treatmentteamand
teachthemhowtomanagethecondition.
29.Earlyafterstrokethepersonwithaspeechandlanguageimpairment
shouldbefacilitatedtocommunicateeverydayneedsandwishes,and
supportedtounderstandandparticipateindecisionsaround,for
example,medicalcare,transfertothecommunity,andhousing.This
mayneedalternativeandaugmentativeformsofcommunication.
30.Peoplewhohavehadastrokeandwhohavepersistingspeechand
languagedeficitsshouldbeassessedforalternativemeansof
communication(gesture,drawing,writing,useofcommunication
aids).
31.Theimpactofspeechandlanguageimpairmentsonliferolesfor
examplefamily,leisure,work,etc.shouldbeassessedandpossible
environmentalbarriers(forexamplesigns,attitudes),shouldbe
addressed,jointlywiththeMDT.
72.Helpandenablepeoplewithcommunicationdifficultiesafterstroke
tocommunicatetheireverydayneedsandwishes,andsupportthem
tounderstandandparticipateinbotheverydayandmajorlife
decisions.
73.Ensurethatenvironmentalbarrierstocommunicationareminimised
forpeopleafterstroke.Forexample,makesuresignageisclearand
backgroundnoiseisminimised.
Considerations
TheGDGagreedthatsupportandinformationforthepatientandtheir
carerwasextremelyimportant,butalsothatgoodcommunicationskills
arerequiredbyallofthemultidisciplinaryrehabilitationteam.The
speechandlanguagetherapisthasaroletoeducateandprovidetraining
toalloftheteaminordertoensurethatappropriatemethodsof
communicatingwiththepersonareused,andthatthereisgreater
awarenessofenvironmentalfactors,suchasnoise,signsandnotices
withintherehabilitationunitthatcouldimpactonthepersonsabilityto
communicate.
Noconsensuswasachievedonspecificformsoftherapytoofferto
peoplewithotherformsofspeechandlanguageimpairment,however
theGDGagreedthattherecommendationsmadefromthereviewof
peoplewithaphasiacouldapplytoallpeoplewithcommunication
NationalClinicalGuidelineCentre,2013.
291
StrokeRehabilitation
Communication
difficulties,andthepreferredapproachtoadoptwouldbedetermined
throughadetailedassessmentbythetherapistandtheneedsandwishes
ofthepatient.
Theuseofalternativemethodsofcommunicationaidssuchas
technologiesviacomputersandsmartphoneswasacknowledged,andit
wasagreedtheusageoftheseislikelytoincrease.Howeverthesewould
notbesuitableforallpeopleanduseofdrawingandwritingdown
informationsuchasappointmentletters,rehabilitationtimetablesetc.
shouldbeprovidedtothosethatneedthem.
12.4 Intensityofspeechandlanguagetherapy
12.4.1
Evidencereview:Inpeopleafterstrokewithcommunicationdifficultieswhatistheclinical
andcosteffectivenessofintensivespeechtherapyversusstandardspeechtherapy?
ClinicalMethodologicalIntroduction
12.4.1.1
Population
Adultsandyoungpeople16orolderwhohavehadastrokeand
havecommunicationdifficulties
Intervention
Intensivespeechtherapy:
aphasiatherapy,
constraintinducedaphasiatherapy
(Anystudyincludingmoreintensiveversuslessintensivespeech
therapy)
Comparison
Lessspeechtherapy
Notherapy
Outcomes
Anyoutcomereportedinthepapers.
Examplesinclude:
FunctionalAssessmentofCommunicationSkillsforAdults
(ASHAFACS)
BostonNamingTest
WesternAphasiaBattery
StrokeDyphasiaIndex
McKennaGradedNamingTest
Clinicalevidencereview
SearcheswereconductedforsystematicreviewsandRCTsthatcomparedtheeffectivenessof
intensivespeechtherapytolessspeechtherapyornotherapytoimprovespeechandlanguage
functioninadultsandyoungpeople16orolderafterstroke.Eight(8)RCTswereidentified.Table75
summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
Table75: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
AUTHOR
POPULATION
INTERVENTION
COMPARISON
OUTCOME
Bakheit,
200713
Firsttimestroke
patients(meandays
Five1hourlyspeech
therapysessionsa
Lessspeechtherapy
(i.e.thesameas
Western
AphasiaBattery
NationalClinicalGuidelineCentre,2013.
292
StrokeRehabilitation
Communication
AUTHOR
POPULATION
poststrokeonset:31)
withascoreof<93.8
onWesternAphasia
Battery.
INTERVENTION
weekfor12weeks,
targetedat
improving
understandingand
expressionofspoken
andwritten
language,including
picture/object
selection,naming
objects,describing
andrecognising
associationsbetween
items,facilitating
expressionoffeelings
andopinions,
improving
communicationskills,
usinggesturesand
nonverbal
communication,
usingcommunication
aidsandequipment
(partof
multidisciplinary
rehabilitation)(Mean
amount4.3hours(SD
1.0)perweek
achieved).(N=51)
COMPARISON
providedto
interventiongroupbut
onlyfortwo1hour
sessionsperweek;
actualmeanamount
achieved1.6hours(SD
0.5)perweek).(N=46)
Bowen
201229
ACTNoW
(Assessing
the
effectivene
ssof
Communic
ation
Therapyin
theNorth
West)
Onehundredand
seventyadultswith
aphasiaordysarthria
admittedtohospital
withstroke.
Participantsrangedin
agefrom32to97
years(mean70years)
and56%weremen.
Almostallhadaphasia
(90%).
Therapyonaverage
started2weeksafter
stokeandinvolved
22speechand
languagetherapy
contacts,for18
hours(mean),
deliveredover13
weeksinboth
hospitaland
communitysettings.
(N=85)
AttentionControl:An
Therapy
averageof19visitor
Outcome
contacts,forameanof
Measure
15hours.Visitorsdid
activity
notprovidetherapyor
subscale
anycommunication
(TOM)
strategies.Visitorshad
Communicatio
excellentsocialskills
nOutcomes
andgeneral
AfterStroke
competencyandwere
scale(COAST)
trainedtodeliversocial
attentionabsentofany Carers
Communicatio
intuitiveformof
nOutcomes
communication
AfterStroke
therapyorstrategy.
scale(COAST)
(N=85)
Denes,
199659
Patientswithstroke
(meanmonthspost
strokeonset;3)with
globalaphasiawith
lesionrestrictedtoleft
hemisphere.
Individualspeech
therapysessions
(totalmeannumber:
130(range94160))
of4560minutes
eachoverameanof
6months(range5.2
7months)including
conversational
setting,using
speaking,gesturing,
facialexpression.
Lessintensivespeech
Aachener
therapy:mean60
AphasiaTest
(range5670)
(AAT)5
individualspeech
SubtestsToken
therapysessionsof45
Test,
60minuteseachovera
Repetition,
meanof6months
Written
(range5.27months)
Language,
including
naming,
conversationalsetting,
Comprehension
usingspeaking,
andProfile
gesturing,facial
Level
NationalClinicalGuidelineCentre,2013.
293
OUTCOME
StrokeRehabilitation
Communication
AUTHOR
INTERVENTION
(N=8)
COMPARISON
expression.(N=9)
OUTCOME
Doesborgh, Chronicstroke
2004
patients(atleast11
67
monthspoststroke)
aged2086yearsold,
withsemantic(as
assessedbythe
SemanticAssociation
Testandthe
Psycholinguistic
Assessmentof
LanguageProcessing
inAphasiaPALPA)
andphonological
deficits(asmeasured
bytheAachenAphasia
TestAATRepetition
subtest).However
peoplewhowere
assessedashaving
globalaphasiaor
recoveredorno
aphasiaaccordingto
theAATwere
excluded.
Individualmulticue
treatment(computer
programmeforword
finding)total
duration1011
hoursinsessionsof
3045minuteseach
withafrequencyof
twotothreetimesa
weekinaperiodof
approximately2
months.(N=9)
Notherapyfor68
weeks.(N=10)
Hartman,
1987107
Firststroke(1month
poststrokeonset)
patientswithlesion
affectedtoleft
hemisphereandwith
functionallynormal
hearingandvision.
Individual
conventionalspeech
therapyincluding
languagedrills,home
practice,auditory
stimulationatsingle
wordandphrase
level,followspoken
commands,reading,
repetition,sentence
completion,cueing
strategies,twice
weeklyfor6months.
(N=30)
Unstructured
conversationbased
counselling/support
focusedonproblems
ofeverydaylife;
encouraging
independentproblem
solvingby
patient/family,twice
weeklyfor6months.
(N=30)
Katz,
1997133
Chronicstroke
patients(atleast1
yearpoststroke)with
aphasiasubsequentto
asingle,left
hemisphere,
thromboembolic
infarctandno
languagetreatment
duringthe3months
beforeentryintothe
study.
Individualcomputer Notherapy.(N=15)
providedreading
treatmentforchronic
aphasicadults.
(N=21)
Acutestroke(1stor
latertime)patients.
Patientswithverymild
aphasiaorsevere
Individualtwo1hour
speechtherapy
sessionsperweek
(nospecifictypeof
Lincoln,
1984156
POPULATION
PorchIndexof
Communicative
Ability(PICA)
NationalClinicalGuidelineCentre,2013.
294
Nospeechtherapy
(controlsoffered
treatmentatweek34).
(N=164)
Boston
NamingTest
(BNT)
Amsterdam
Nijmegan
Everyday
Language
Test,scaleA
(ANELTA).
ThePorch
Indexof
Communicati
veAbility
(PICA)
Western
aphasia
Battery
(WAB)
Aphasia
Quotient
(AQ)
PorchIndex
of
Communicati
veAbility
StrokeRehabilitation
Communication
AUTHOR
POPULATION
dysarthriawere
excluded.
INTERVENTION
therapywas
included;therapists
organisedtheirown
formoftreatment)
fromweek10post
stroketoweek34.
(N=163)
COMPARISON
OUTCOME
(PICA)
Functional
Communicati
onProfile
(FPA)
Wertz,
1986283
Acutefirsttimestroke
(224weekspost
stroke)maleveteran
patients75yearsor
underwithaleft
hemisphericlesion
Individualspeech
therapyadministered
byaspeechtherapist
for810hours/week
for12weeks,after
thennotreatment
wasgiven.(N=38)
Speechtherapy
administeredat
home:trainedfamily
memberorfriend
administered810
hours/weekfor12
weeks,afterthenno
treatmentwasgiven.
(N=43)
Notherapyfor12
weeks(afterthat
speechtherapist
administered810
hours/weekfor12
week.)(N=40)
PorchIndexof
Communicative
Ability(PICA)
NationalClinicalGuidelineCentre,2013.
295
StrokeRehabilitation
Communication
Comparison:Intensivespeechtherapyversuslessintensivespeechtherapyornothing
Table76:IntensivespeechtherapyversuslessintensivespeechtherapyClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Intensive
speech
therapy
Mean(SD)
Less
intensive
therapy
Mean(SD)
Relative
Mean
difference
(95%CI)
5.2(7.8)
6.2(3.32,
15.72)
MD6.2
higher
(3.32
lowerto
15.72
higher)
Moderate
6.1(6.1)
2.8(3.86,
9.46)
MD2.8
higher
(3.86
lowerto
9.46
higher)
Verylow
11(9.8)
2.1(3.1)
8.9(1.81,
15.99)
MD8.9
higher
(1.81to
15.99
higher)
Moderate
10.2(9.9)
4.5(4.2)
5.7(1.69,
MD5.7
Low
Absolute
Confidence
(ineffect)
AachenerAphasiaTest(AAT)tokentest(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Denes1996
59
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
11.4(11.6)
AachenerAphasiaTest(AAT)repetition(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Denes1996
59
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(B)
8.9(7.7)
AachenerAphasiaTest(AAT)writtenlanguage(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Denes1996
59
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
AachenerAphasiaTest(AAT)Naming(6monthsfollowup)(Betterindicatedbyhighervalues)
1
RCT
Serious
Noserious
Noserious
Serious
NationalClinicalGuidelineCentre,2013.
296
StrokeRehabilitation
Communication
Summaryoffindings
Qualityassessment
Noofstudies
Denes1996
59
Design
Effect
Limitations
limitations
(a)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
(c)
Intensive
speech
therapy
Mean(SD)
Less
intensive
therapy
Mean(SD)
Relative
Mean
difference
(95%CI)
13.09)
2.3(3.8)
10.3(0.52, MD10.3
21.12)
higher
(0.52
lowerto
21.12
higher)
Low
10(8.6)
4.3(3.8)
2.7(0.76,
12.16)
MD2.7
higher
(0.76
lowerto
12.16
higher)
Low
63.1(13.5)
63.2(13.5) 0.1(5.48,
5.28)
MD0.1
lower
(5.48
lowerto
5.28
higher)
High
Absolute
higher
(1.69
lowerto
13.09
higher)
Confidence
(ineffect)
AachenerAphasiaTest(AAT)Comprehension(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Denes1996
59
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(c)
12.6(15.2)
AachenerAphasiaTest(AAT)Profilelevel(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Denes1996
59
RCT
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(c)
WesternAphasiaBattery(2monthsfollowup)(Betterindicatedbyhighervalues)
1
Bakheit2007
13
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
WesternAphasiaBattery(3monthsfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
297
StrokeRehabilitation
Communication
Summaryoffindings
Qualityassessment
Effect
Noofstudies
Design
Limitations
Inconsistency
Indirectness
Imprecision
1
Bakheit2007
13
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(e)
Intensive
speech
therapy
Mean(SD)
Less
intensive
therapy
Mean(SD)
Relative
Mean
difference
(95%CI)
67.7(13.4)
69.2(13.4) 1.5(6.84,
3.84)
MD1.5
lower
(6.84
lowerto
3.84
higher)
Low
68.6(15.4)
71.4(15.4) 0.4(0.16,
0.96)
MD2.8
lower
(8.94
lowerto
3.34
higher)
Low
Absolute
Confidence
(ineffect)
WesternAphasiaBattery(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Bakheit2007
13
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(c)
(a)
Allocationconcealmentandblindingofoutcomeassessorsnotreported.
ConfidenceintervalcrossedbothendsofdefaultMID.
(c)
ConfidenceintervalcrossedoneendofdefaultMID.
(d)
Unclearrandomization.
(b)
NationalClinicalGuidelineCentre,2013.
298
StrokeRehabilitation
Communication
Table77: IntensivespeechtherapyversusnotherapyClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Intensive
speech
therapy
Mean(SD)
Effect
No
therapy
Mean(SD)
Mean
difference
(95%CI)
Mean
Difference Confidence
(95%CI)
(ineffect)
BostonNamingTest(BNT)(2monthfollowup)(Betterindicatedbyhighervalues)
1
Doesborgh
2004
67
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(b
)
75.6(38.7)
75.7(36.7) 0.1(
35.26,
35.06)
MD0.1
lower
(35.26
lowerto
35.06
higher)
Verylow
25.5(10.3) 8.8(0.16,
17.44)
MD8.8
higher
(0.16to
17.44
higher)
Low
MD5.1
higher(7
lowerto
17.2
higher)
Low
MD5
lower
(17.88
lowerto
7.88
higher)
Low
AmsterdamNijmegenEverydayLanguageTestANELTA(2monthfollowup)(Betterindicatedbyhighervalues)
1
Doesborgh
2004
67
RCT
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
34.3(8.4)
PorchIndexofCommunicativeability(PICA)moreintensiveversusnotreatment(26weekfollowup)(Betterindicatedbyhighervalues)
1
Katz1997
133
RCT
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(c)
66.4(19.4)
61.3(17.4) 5.1(7.0,
17.2)
PorchIndexofCommunicativeability(PICA):lessintensiveversusnotreatment(26weekfollowup)(Betterindicatedbyhighervalues)
1
Katz1997
133
RCT
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
26.3(20.9)
61.3(17.4) 5.0(
17.88,
7.88)
NationalClinicalGuidelineCentre,2013.
299
StrokeRehabilitation
Communication
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Intensive
speech
therapy
Mean(SD)
Effect
No
therapy
Mean(SD)
Mean
difference
(95%CI)
Mean
Difference Confidence
(95%CI)
(ineffect)
WesternAphasiaBattery:moreintensiveversusnotreatment(26weekfollowup)(Betterindicatedbyhighervalues)
1
Katz1997
133
RCT
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(b
)
73.6(22.6)
72.2(23.7) 1.40(14,
16.8)
MD1.4
higher(14
lowerto
16.8
higher)
Verylow
MD8.8
lower
(26.35
lowerto
8.75
higher)
Low
WesternAphasiaBattery:lessintensiveversusnotreatment(26weekfollowup)(Betterindicatedbyhighervalues)
1
Katz1997
133
RCT
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
63.4(28.5)
72.2(23.7) 8.8(
26.35,
8.75)
PorchIndexofCommunicativeAbility(PICA)(1monthfollowupafterlesiononset)(Betterindicatedbyhighervalues)
1
Hartman
1987
107
RCT
Serious
limitations(e
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
9.34(3.28)
PorchIndexofCommunicativeAbility(PICA)(10monthfollowupafterlesiononset)(Betterindicatedbyhighervalues)
1
Hartman
1987
107
RCT
Serious
limitations(e
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
11.22(2.88)
10.86
(4.02)
NationalClinicalGuidelineCentre,2013.
300
0.36(1.57, MD0.36
2.29)
higher
(1.57
lowerto
2.29
higher)
Low
StrokeRehabilitation
Communication
(a)
Blindingofoutcomeassessorsnotreported.
ConfidenceintervalcrossedtwoendsofdefaultMID.
(c)
ConfidenceintervalcrossesoneendofdefaultMID.
(d)
Detailsofblinding,randomisationandallocationconcealmentnotreported.
(e)
Allocationconcealmentnotreported.
(b)
Table78: Earlyintensivespeechtherapy(speechtherapydeliveredinthehospitalandcontinuedafterdischargeinthecommunity)versusnospeech
therapy(attentioncontrol)
Qualityassessment
Summaryoffindings
Noof
studies
Early
intensive
speech
Nospeech
therapy
therapy
Mean(SD) Mean(SD)
Design
Limitations
Inconsistency
Indirectness
Imprecision
Effect
Mean
difference
(95%CI)
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
TherapyOutcomeMeasureactivitysubscale(6monthfollowup)(Betterindicatedbyhighervalues)
1
Bowen
201229
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
a)
3.3(1.4)
3.0(1.6)
0.30(0.18, MD0.3
0.78)
higher
(0.18lower
to0.78
higher)
Moderate
73(18)
2.00(
8.59,4.59)
MD2.00
lower(8.59
lowerto
4.59
higher)
High
0.00(6.73, MD0.00
6.73)
(6.73lower
to6.73
higher)
High
CommunicationOutcomesAfterStrokescale(COAST)(6monthfollowup)(Betterindicatedbyhighervalues)
1
Bowen
201229
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
71(18)
CarerCommunicationOutcomesAfterStrokescale(CarerCOAST)(6monthfollowup)(Betterindicatedbyhighervalues)
1
Bowen
201229
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
62(21)
62(18)
NationalClinicalGuidelineCentre,2013.
301
StrokeRehabilitation
Communication
(c) ConfidenceIntervalcrossesoneendofdefaultMID(0.5)
Narrativesummaries
ThefollowingstudiesaresummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Onestudyofmoderateriskofbias(Lincolnetal,1984)156usingthePICAandtheFunctionalCommunicationProgramme(FCP)showedimprovementin
bothstandardspeechtherapygroupandmorespeechtherapygroupwithnosignificantdifferencesinlanguagerecoverybetweenthetwogroups
duringtreatmentandat34weekposttreatment.
Onestudyofhighriskofbias(Wertzetal,1986)283usingthePICApercentilesuggestedthatclinictreatmentforaphasiawasefficaciousand
delaying/deferringtreatmentfor12weeksdidnotcompromiseultimateimprovement.Resultsofhometreatmentdidnotdifferfromthoseofclinic
treatmentordeferredtreatment.
NationalClinicalGuidelineCentre,2013.
302
StrokeRehabilitation
Communication
12.4.1.2
Economicevidence
Onestudywasincludedthatincludedtherelevantcomparison169.Thisissummarisedintheeconomic
evidenceprofilebelow(Table79andTable80).SeealsothefullstudyevidencetablesinAppendixI.
Table79: EnhancedversusstandardspeechtherapyEconomicstudycharacteristics
Study
169
Marsh2010
(UKNHS)
Limitations
Applicability
Othercomments
Potentiallyserious
limitations(a)
Directlyapplicable
BasedontheRCTbyBakheitetal200713
includedinourclinicalreview.
UsualSLTdefinedas2hoursaweekfor
12weeks(inpracticewas0.57hoursper
week).
EnhancedSLTdefinedas2hoursaweek
for12weeks(inpracticewas1.6hours
perweek).
TheoriginalRCTcomparedenhanced
SLTalsowithintensiveSLT(definedas5
hoursaweekfor12weeks,inpractice
was3hoursperweek);however
intensiveSLThadnosignificanteffect
overandaboveenhancedSLT,therefore
intensiveSLTwasnotconsideredinthe
economicanalysis(itwouldbe
dominatedbyenhancedSLT).
(a) TheconversionofWABtestscoresintoQALYgainswasbasedonanumberofassumptions.Forexample,itassumed
thattheWABtestiscomparabletotheAphasiatestandthatbothscaleshaveasimilardistribution.TheWABtestis
scoredoutof1to100,whiletheaphasiatestisscoredbetween0and20.Issuesaroundtranslationaphasiascalestothe
Barthelindexwhichmeasuresfunction.TheeffectivenessdatausedintheanalysisisbasedononeRCT;howeverthe
NCGCclinicalreviewhasidentifiedadditionalrelevantstudies.
Table80: EnhancedversusstandardspeechtherapyEconomicsummaryoffindings
Study
169
Marsh2010
(UKNHS)
Incremental
cost()
Incremental
effects
(QALYs)
ICER
(/QALY)
844(a)
0.057(b)
14,807
Uncertainty
ThepercentageimprovementinWAB
testfollowingenhancedSLTwasvaried
between70%and80%.EnhancedSLT
remainedcosteffectiveaslongasthe
improvementwasabove72%.
ThechangeinQALYgainwasvaried
between0.040and0.058.Enhanced
SLTremainedcosteffectiveaslongas
theincrementalQALYgainwasabove
0.042.
(a) 2009UKpounds.Costsincorporated:communitySLTcosts(band7).StandardSLTcostwascalculatedas6.9hoursper
patientover12weeks.EnhancedSLTcostwascalculatedas19.3hoursperpatientover12weeks.
(b) TheimprovementinWABtestscores(frombaselineto24weeks)wascalculatedandconvertedtoaphasiatestscores.
ThesescoreswerethentranslatedtotheBarthelindex.TheQALYgainwasobtainedbymappingfromtheBarthelindex
toEQ5DusingalinearregressionanalysisreportedinapaperbyExeletal(2004).271
NationalClinicalGuidelineCentre,2013.
303
StrokeRehabilitation
Communication
12.4.1.3
Evidencestatements
Clinicalevidencestatements
Onestudy59comprisingof17participantsfoundthemoreintensivespeechinterventionwas
associatedwithastatisticallysignificantimprovementinthefollowingsubsetsofAachenerAphasie
Testcomparedwiththelessintensiveinterventionat6monthsfollowup:
o tokentest(MODERATECONFIDENCEINEFFECT)
o writtenlanguage(MODERATECONFIDENCEINEFFECT)
Onestudy59comprisingof17participantsfoundnosignificantdifferencebetweenthemoreintensive
speechgroupandthelessintensivegroupat6monthsfollowupatthefollowingsubsetsof
AachenerAphasieTest:
repetition(VERYLOWCONFIDENCEINEFFECT)
naming(LOWCONFIDENCEINEFFECT)
comprehension(LOWCONFIDENCEINEFFECT)
profilelevel(LOWCONFIDENCEINEFFECT)
Onestudy13comprisingof97participantsfoundnosignificantdifferenceinWesternAphasiaBattery
scorebetweentheintensivespeechgroupandlessintensivegroupattheendof2,3and6months
followup(MODERATE,LOWandMODERATECONFIDENCEINEFFECT,respectively).
Onestudy67comprisingof19participantsfoundnosignificantdifferenceintheBostonNamingTest
betweentheintensivespeechgroupandthenotherapygroupat2monthsfollowup(VERYLOW
CONFIDENCEINEFFECT).
Onestudy67comprisingof19participantsfoundthattheparticipantsreceivedmoreintensivespeech
therapyshowedastatisticallysignificantimprovementintheAmsterdamNijmeganEveryday
LanguageTestcomparedwiththenotherapygroupat2monthsfollowup(LOWCONFIDENCEIN
EFFECT).
Onestudy133comprisingof36participantsfoundnosignificantdifferencebetweenthemore
intensivespeechgroupandthenotherapygroupat6monthsfollowuponthefollowingoutcomes:
PorchIndexofCommunicativeAbility(PICA)(LOWCONFIDENCEINEFFECT)
WesternAphasiaBattery(LOWCONFIDENCEINEFFECT)
Onestudy133comprisingof36participantsfoundnosignificantdifferencebetweenthelessintensive
groupandthenotherapygroupat6monthsfollowuponthefollowingoutcomes:
PorchIndexofCommunicativeAbility(PICA)(VERYLOWCONFIDENCEINEFFECT)
WesternAphasiaBattery(LOWCONFIDENCEINEFFECT)
Onestudy107comprisingof60participantsfoundnosignificantdifferenceinthePorchIndexof
CommunicativeAbility(PICA)betweenthemoreintensivespeechgroupandthenotherapygroupat
1and10monthsfollowupafterlesiononset(LOWCONFIDENCEINEFFECT).
Onestudy29comprising153participantsfoundnosignificantdifferenceintheTherapyOutcome
MeasureSubscale(TOM)betweentheearlyintensivespeechtherapygroupandthenotherapy
groupat6monthfollowup(MODERATECONFIDENCEINEFFECT)
Onestudy29comprising117participantsfoundnosignificantdifferenceintheCommunication
OutcomesAfterStrokescale(COAST)betweentheearlyintensivespeechtherapygroupandtheno
therapygroupat6monthfollowup(HIGHCONFIDENCEINEFFECT)
NationalClinicalGuidelineCentre,2013.
304
StrokeRehabilitation
Communication
Onestudy29comprising129participantsfoundnosignificantdifferenceintheCarerCommunication
OutcomesAfterStrokescale(carerCOAST)betweentheearlyintensivespeechtherapygroupand
thenotherapygroupat6monthfollowup(HIGHCONFIDENCEINEFFECT)
Economicevidencestatements
Onedirectlyapplicablestudy169withpotentiallyseriouslimitationsshowedthatenhancedspeech
therapyiscosteffectivecomparedtostandardspeechtherapy.Enhancedspeechtherapyismore
costlybutalsomoreeffectivethanstandardspeechtherapyandtheICERisbelowthe20,000/QALY
threshold(14,807perQALYgained).TheseresultsweresensitivetotheimprovementinWABtest
andtotheQALYgainachievedwithenhancedtherapy.
12.4.2
Recommendationsandlinktoevidence
74.Referpeoplewithsuspectedcommunicationdifficultiesafter
stroketoaspeechandlanguagetherapistfordetailed
analysisofspeechandlanguageimpairmentsandassessment
oftheirimpact.
75.Speechandlanguagetherapistsshould:
providedirectimpairmentbasedtherapyfor
communicationimpairments(forexample,aphasiaor
dysarthria)
helpthepersonwithstroketouseandenhancetheir
remaininglanguageandcommunicationabilities
teachothermethodsofcommunicating,suchasgestures,
writingandusingcommunicationprops
coachpeoplearoundthepersonwithstroke(including
familymembers,carersandhealthandsocialcarestaff)to
developsupportivecommunicationskillstomaximisethe
personscommunicationpotential
helpthepersonwithaphasiaordysarthriaandtheirfamily
orcarertoadjusttoacommunicationimpairment
supportthepersonwithcommunicationdifficultiesto
rebuildtheiridentity
supportthepersontoaccessinformationthatenables
decisionmaking.
Tradeoffbetweenclinical
benefitsandharms
TheGDGagreedthatitwasunlikelythattherewereanysignificant
harmsassociatedwiththisformoftherapyprovidedpatient
expectationswerekeptrealistic,afocusremainedonparticipation,and
thepatient,familyandfriendsweresupportedtomanageanypersisting
disability.Itwasfeltthatanimprovementintheabilitytocommunicate
wouldhaveasignificantimpactintermsofqualityoflifeforthepatient
withstroke.
Economicconsiderations
TheGDGnotedthatspeechtherapiesarecurrentlyroutinelyofferedin
theUKNHStostrokepatientswithaphasia.Moreintensivetherapy
wouldbeassociatedwithincreasedpersonnelcosts.Oneeconomic
evaluation169basedonanRCTincludedinourreviewshowedthat
enhancedspeechtherapy(whichwas1.6hoursaweekfor12weeks)is
NationalClinicalGuidelineCentre,2013.
305
StrokeRehabilitation
Communication
morecostlybutalsomoreeffectivethanstandardspeechtherapy(which
was0.57hoursaweekfor12weeks)andtheICERisbelowthe20,000
perQALYthreshold,thereforethiswouldbeconsideredcosteffective.
However,intensivespeechtherapy(whichwas3hoursperweekfor12
weeks)wasnotincludedintheformalanalysissinceintheRCTthiswas
morecostlyandnotmoreeffectivethanenhancedspeechtherapy.
Intensivespeechtherapyisthereforenotcosteffective.
Qualityofevidence
Thestudiesshouldbeconsideredasfeasibilitystudiesduetothesmall
samplesize.Itwasnotedthattheywereunderpowered.
Onestudyshowedthatmoreintensivespeechtherapywasassociated
withanimprovementintokentestandwrittenlanguageasassessedby
AachenerAphasieTestat6monthsfollowupcomparedtolessintensive
speechtherapy(Denes59).Confidenceintheeffectshownforthese
outcomeswasgradedasmoderate.
TheGDGconsideredthatthepopulationsincludedinthetrialsdidnot
reflectthosewhowouldbeseeninclinicalpractice,manyofwhom
wouldhavehadspeechproblemsforasignificantlylongertime.
InstudiesofintensityintheACTNoWstudytheGDGagreedthe
evidencesuggeststhatgreaterintensitywhencomparedtosocial
supportwithoutdirectspeechandlanguagetherapyresultedingreater
benefit,althoughthiswasnotstatisticallysignificant.TheGDGagreed
thatthequestionstilltobeansweredwaswhethermoreintensespeech
andlanguagetherapydeliveredbymoreskilledstaffwoulddeliver
greaterbenefit.WhattheACTNoWstudyaddsisthevalueof
appropriatelytrainedpaidvisitorstoaidtherecoveryoffunctional
communicationafterstroke.
Otherconsiderations
Theincludedstudieslookedatmoreintensivespeechtherapycompared
tolessintensiveanditwastheviewoftheGDGthatnoneofthestudies
presentedwereparticularlyfocussedonintensity.Threestudies
includedinterventionarmsthatwereonlyshortsessions23timesper
week(Doesborgh2004,Hartman1987,Lincoln198467,107,156)theGDG
agreedthat18hoursand22contactsover13weeksdeliveredintheACT
NoWstudywasnotparticularlyintensive.Thegroupalsonotedthatthe
interventionwasdeliveredtwoweeksafterstroke,andthoughtthat
providingSLTthisearlyafterstrokewouldoftennotbeanappropriate
time.
TherewasadiscussionintheGDGaboutthenatureofintensityin
speechandlanguagetherapy,anditwasfeltthatoperationaldefinitions
ofintensityneedtobeagreed.TheGDGnotedthatalltherapieswere
providedinonetoonesessionsandconsideredthiswasappropriatefor
thistypeofrehabilitation.TheGDGalsorecognisedthatmanypeople
withaphasiaalsobenefitfromspeechtherapyingroupstoenhance
functionalcommunicationandconfidence.
Furtherresearchisrequiredthatconsidersover8hoursoflanguage
therapyaweekreflectingtheamountoftherapydeliveredbytheWertz
study283.
TheGDGacknowledgedthestandardspeechtherapyof5x45minsper
week.Thegroupfeltthatthistobetheminimumbutwereawarethat
thisisnotconsideredstandardcare.
Theassessmentoflanguagedeficitsandtherapytailoredtorestore
thosedeficitsshouldbeconductedbyspecialistspeechandlanguage
therapists.Theapproachtakenforeachindividualpatientwilldependon
theprofileofdeficitsandthereforeagenerictreatmentcannotbe
recommended.HoweverforpeoplewithaphasiaordysarthriatheGDG
NationalClinicalGuidelineCentre,2013.
306
StrokeRehabilitation
Communication
recognisedtheuseofimpairmentbasedtherapytohelprestorespeech
orlanguage.
12.5 Listeneradvice
Aphasiacanimpactonapersonsabilitytounderstandandtoexpresslanguagethrougha
rangeofmodalitiesincludingspeech,writing,drawingandgesturemakingcommunication
difficultfortheindividualwhohashadastrokeandthecommunicationpartner.Social
isolationisarguablythemostdevastatingconsequenceofaphasia.Itisthereforeimportant
thatfamilies/carersofthestrokepatientaskeycommunicationpartnersareactively
involvedinaphasiarehabilitationprogrammes.SpeechandLanguageTherapyaimsto
addressaphasiarehabilitationusingasupportedcommunicationapproachwhichincludes
listeneradviceandmultifunctionalskillstofacilitatecommunication.
Keycomponentsoflisteneradvicetraininginvolvesomeofthefollowing:
Useofandencouragementofmultimodalcommunicationi.e.writing,drawing,gesture,
speech,augmentativeandalternativecommunicationaids
Simplificationoflanguagebyusingshort,uncomplicatedsentences.
Facilitationofcomprehensionskillsbycheckingthatthemessageisunderstood.This
mayrequirerepetition
Writingdownkeywordstoclarifymeaningasneeded.
Minimisingdistractionswheneverpossible.
Encouragingengagementofthepersonwithaphasiainconversations.
Maintaininganaturalconversationalmannerappropriateforanadult.
Avoidingcorrectingthepersonsspeechorinterrupting
Allowingthepersonplentyoftimetotalk.
Encouragingconversationalturntaking
Askingforandvaluingtheopinionofthepersonwithaphasia,especiallyregardingfamily
matters.
12.5.1
Whatlisteneradviceskills/trainingorinformationwouldhelpfamilymembers/carers
improvecommunicationinpeoplewithaphasiaafterstroke?
ClinicalMethodologicalIntroduction
12.5.1.1
Population
Familiesandcarersofadultsandyoungpeople16orolderwith
aphasiaafterstroke
Intervention
Listeneradviceskills/trainingorinformation
Comparison
Usualcareornothing,shamoralternativeinterventions
Outcomes
Anyoutcomereportedinthepaper
Qualityoflife
Clinicalevidencereview
SearcheswereconductedforsystematicreviewsandRCTscomparingtheeffectivenessoflistener
adviceskills/trainingorinformationwithusualcare,shamoralternativeinterventionsthatwouldaid
NationalClinicalGuidelineCentre,2013.
307
StrokeRehabilitation
Communication
familymembers/carersimprovecommunicationinpeoplewithaphasiaafterstroke.TwoRCTswere
identified.
Table81summarisesthepopulation,interventionandoutcomesofeachofthestudiesincludedin
theevidencereview.
Table81:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOME
Kagan,
2001127
Atleastoneyear
poststroke
patientswith
moderateto
severeaphasia.
Conversationwith
trainedvolunteer;
Supported
Conversationfor
adultswith
Aphasia":for
examplekeeping
talkasnaturalas
possible,avoiding
beingpatronising,
ensuring
understanding,
allowingpersonto
expressknowledge,
thoughtsand
feelings,verifying,
usinggesture,
writinganddrawing.
(N=20)
Conversationwith
untrainedvolunteer
(N=20)
MeasureofSkill
inProviding
Supported
Conversation
forAdultswith
Aphasia(MSCA)
Measureof
Participationin
Conversation
forAdultswith
Aphasia(MPCA)
Worrall,
2000288
Patientswith
chronicaphasia
duetostroke(at
least12months
postonset)in
language
dominant
hemisphere.
Functional
communication
therapyprogramme
(SpeakingOut)
deliveredbytrained
volunteersin
patient'shome;
focusingon
strategiesto
improve
communication
activities(,for
examplepayingbills
directlyfrombank
account,using
multitripbus
tickets)for10
weeks.(N=6)
Nonverbalrecreational
programmefor10weeks.
(N=8)
Western
AphasiaBattery
(WAB),
American
Speech
Language
Hearing
Association
Functional
Assessmentof
Communication
Skills(ASHA
FACS)
Communication
Effectiveness
Index(CETI)
Functional
Communication
Therapy
Planner(FCTP)
ShortForm36
(SF36).
NationalClinicalGuidelineCentre,2013.
308
StrokeRehabilitation
Communication
Comparison:Listeneradviceskills/trainingorinformationversususualcare,nothing,shamoralternativeinterventionsforstrokerehabilitation
Table82:Listeneradviceskills/trainingorinformationversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Listener
advice
Mean(SD)
Usualcare Mean
difference
Mean(SD) (95%CI)
Mean
Difference
(MD)(95% Confidence
CI)
(ineffect)
MeasureofSkill(oflistener)inprovidingSupportedConversationforAdultswithAphasia(acknowledgecompetence)(Posttest)(Betterindicatedbylowervalues)
1
Kaganetal,
2001127
RCTquasi
randomised
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
2.6(0.7)
1.5(0.8)
1.1
(0.63,1.57)
MD1.1
higher
(0.63to
1.57
higher)
Moderate
MeasureofSkill(oflistener)inprovidingSupportedConversationforAdultswithAphasia(revealcompetence)(Posttest)(Betterindicatedbylowervalues)
1
Kaganetal,
2001127
RCTquasi
randomised
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
2.7(0.6)
0.7(0.4)
2.0(1.68,
2.32)
MD2
higher
(1.68to
2.32
higher)
Moderate
MeasureofParticipation(ofpersonwithaphasia)inConversationforAdultswithAphasia(interaction:socialconnection)(Posttest)(Betterindicatedbyhighervalues)
1
Kaganetal,
2001127
RCTquasi
randomised
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
2.6(9)
2.2(9)
0.4(0.16,
0.96)
MD0.4
higher
(0.16
lowerto
0.96
higher)
Low
MeasureofParticipation(ofpersonwithaphasia)inConversationforAdultswithAphasia(transaction:exchangeofcontent)(Posttest)(Betterindicatedbylower
values)
1
RCTquasi
Serious
Noserious
Noserious
Noserious
2.7(8)
2.0(8)
NationalClinicalGuidelineCentre,2013.
309
0.7(0.20,
MD0.7
Moderate
StrokeRehabilitation
Communication
Summaryoffindings
Qualityassessment
Noofstudies
Kaganetal,
2001127
(a)
(b)
Design
randomised
single
blinded
Effect
Limitations
limitations
(a)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
Listener
advice
Mean(SD)
Usualcare Mean
difference
Mean(SD) (95%CI)
1.20)
Mean
Difference
(MD)(95% Confidence
CI)
(ineffect)
higher(0.2
to1.2
higher)
Narrativesummary
ThefollowingstudyissummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Onematchedpairsingleblindedrandomisedstudy(Worralletal)288reportedthattherewasastatisticallysignificantimprovementontheWestern
AphasiaBatteryinthosewhoreceiveda10weekfunctionalcommunicationtherapy(SpeakingOut)programmedeliveredbytrainedvolunteers,
comparedtothosewhoreceivedusualcarewithrecreationalactivities/notreatment.Therewerenosignificantdifferencesinanyofthefunctional
communicationmeasures.Apositivemeanchangeof29.3(SD=19.3)ontheGeneralHealthscaleoftheaphasicpersonsSF36inonegroupwasobserved
andtherewasastatisticallysignificantnegativemeanchangeof18.1(SD=18.06)forthebodilypainscaleofthespousesSF36intheothergroup,
followingtheSpeakingOutprogramme.
NationalClinicalGuidelineCentre,2013.
310
StrokeRehabilitation
Communication
12.5.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparinglisteneradviceskills/informationtofamilymembers
/carerstoimprovecommunicationinpeoplewithaphasiaafterstrokewithusualcareorno
interventionwereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
CostsoftheinterventionwereestimatedbasedonastudybyVanderGaagetal,2008268.Thisstudy
wasacostanalysisofvariousspeechandlanguageprogrammes.Costswerecalculatedonthebasis
of7sessionswith17patientspersession.BasedontheresourceusebreakdownshowninTable83,
thetotalcostperpatientis517.Itisassumedthisisontopofusualcare.
Table83: Interventioncostslisteneradviceskills/informationtofamilymembers/carers
Resource
Quantity/hrs.(a)
Cost(b)
Seniorspeechandlanguagetherapist(band7)
57/hr.ofclientcontact
Speechandlanguagetherapist(band6)
47/hr.ofclientcontact
Assistant
25/hr.spentwithapatient
Transport
40/visit(a)
517
(c)
Totalcostperpatient
268
(a) Source:VanderGaagetal,2008
(b) EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandrelevantAgendaforChangesalarybands51(typicalsalarybandidentifiedbyclinicalGDGmembers)
(c) Assumedtobecostedsimilartoaband2clinicalsupportworkernursing(community)
12.5.1.3
Evidencestatements
Clinicalevidencestatements
Onequasirandomisedstudy127of40participantsfoundthatthetrainedvolunteersscoreda
statisticallysignificanthigherratingofskillinacknowledgingandrevealingcompetenceoftheir
partnerswithaphasiausingtheMSCA(measureofsupportedconversationforadultswithaphasia),
comparedwiththeuntrainedvolunteers(MODERATECONFIDENCEINEFFECT).
Onequasirandomisedstudy127of40participantsfoundthattherewasnosignificantdifferencein
theratingofMPCAinteraction:socialconnection(ratedbyaphasicindividuals)betweenthosewho
receivedtheSupportedConversationforAdultswithAphasia(SCA)interventionwithtrained
volunteerscomparedwithusualcareattheendoftheintervention(LOWCONFIDENCEINEFFECT).
Onequasirandomisedstudy127of40participantsfoundthatthosewhoreceivedtheSupported
ConversationwithAdultsinterventionwithtrainedvolunteersscoredastatisticallysignificanthigher
ratingofMPCAtransaction:messageexchange(ratedbyaphasicindividuals)comparedwithusual
careattheendoftheintervention(MODERATECONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
311
StrokeRehabilitation
Communication
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
12.5.2
Recommendationsandlinktoevidence
76.Offertrainingincommunicationskills(suchasslowingdown,
notinterrupting,usingcommunicationprops,gestures,
drawing)totheconversationpartnersofpeoplewithaphasia
afterstroke.
Relativevalueplacedonthe
outcomesconsidered
Anyoutcomereportedinthestudieswasincludedinthereview.
ItwasnotedthattheKaganstudy127measuredboththeacquiredskillsof
patientsandthevolunteerlisteners.TheGDGfeltthatinthecontextof
conversationpartners,itwasimportanttoconsideroutcomesforboth
groupsofparticipants.
Noqualityoflifeoutcomeswereaddressedintheevidencepresented.The
GDGconsideredthattheseoutcomemeasureswouldconsiderthesocial
andpsychologicalchallengesassociatedwithaphasiawhichareparticularly
relevanttothisintervention.Thelackofstandardisedoutcomesinthe
studieswasnoted.
Qualityofevidence
TheGDGagreedthatevidencepresentedwaslimitedtotwo,smallstudies
whichmightberegardedasfeasibilitystudiesandthereforenotrobust.
127 288
HowevertheGDGnotedthattheKaganstudydemonstratedtraining
volunteersresultedinapositivebenefitforthemeasureofskilloutcomes
andalsodemonstratedabenefitinparticipationinconversationbythe
personwithaphasia.Confidenceintheresultsshownfortheseoutcomes
wasgradedasmoderateduetounclearrandomisation.Anothersmall
study288reportedasignificantimprovementinthosewhoreceiveda10
weekfunctionalcommunicationtherapydeliveredbytrainedvolunteers.
Howevertheseresultswerenotpresentedinnumericaldatathatcouldbe
includedintheGRADEanalysis.
Thedifficultyofrecruitingasuitablylargepatientpopulationand
volunteersforthistypeofinterventionwasrecognised.Itwasagreedthat
thefeasibilityofofferingthisinterventionanditsapplicabilitytoUK
practicewasdemonstrated.
Tradeoffbetweenclinical
benefitsandharms
TheGDGdiscussedthedefinitionoflistenersadviceandtheelementsof
supportedconversationtraining.TheGDGconsideredtherewerelikelyto
bebenefitsfromprovidingsimplecommunicationskillstrainingtofamily
members,carersandvolunteersandthetypesofskillsemployedwithinthe
studiesreviewedshouldbegivenasexamplessuchasslowingspeech
down,notinterrupting,usinggesturesandwritingordrawing.
TheGDGacknowledgedthattheKaganstudy127mayhavedifferent
applicabilitywheninterventionsaredeliveredtofamilymemberswhere
otherfactorsmayimpact.
Economicconsiderations
Nocosteffectiveevidencewasfound.Thecostofprovidingtrainingin
communicationskillstotheconversationpartnersofpeoplewithaphasia
wasestimatedataround517perpatient.TheGDGconsideredthiscost
likelytobeoffsetbythebenefitsoftrainingtheconversationpartners.
NationalClinicalGuidelineCentre,2013.
312
StrokeRehabilitation
Movement
13 Movement
Weaknessiscommonafterstroke;thismayariseduetotheuppermotorneuronlesioncompounded
byinactivityasaconsequenceoflimitedphysicalmobility.Weaknesslimitspatientsabilitytomove
thebody,includingchangingbodyposition,transferringfromoneplacetoanother,andwalkingas
wellasupperlimbfunctionssuchascarrying,movingormanipulatingobjects.Italsolimits
performanceofactivitiesofdailylivingandmayleadtoamoregeneralisedlossoffitness.
Theabilitytowalkfollowingastrokeisoftenaffectedorlostduetomultipleandcomplexdeficitsof
motorandsensorysystemsgivingrisetolossofmovement,balanceandposturalcontrol.
Rehabilitationofgaitideallyattemptstorestoreanormalwalkingpatternorifthisisnotpossibleto
developacompensatoryapproachusingvariousaidstopromoteasafeandfunctionallevelof
mobility.Theabilitytowalkfollowingastrokeisoftenseenbyboththepatientwithstrokeandtheir
teamasamajordesirableoutcomeatanystagethroughouttherehabilitationjourneyandsignificant
timeandeffortisoftenrequiredbythepatientandtheirteaminordertoreeducategaitand
promoteindependentmobility.
Maximisingupperlimbrecoveryafterstrokesimilarlyrequiressignificanttimeandeffortbythe
patientafterstrokeandtherehabilitationteam.Ithasbeenestimatedthatupperlimbparalysis
affectsonethirdofthestrokepopulation.Thesiteandsizeofthelesionisamajordeterminantof
outcome,withsomepeopleafterstrokehavingsuchseverelossthatnoamountoftherapywill
affectfunctionalrecovery.However,asignificantproportionofpeoplefollowingstrokewillregain
goodarmfunctionthroughspontaneousrecovery.Intheremainder,additionaltherapymayimprove
outcomes.
Manyofthetechniquesusedtosupportthepatientinrelearningmotorskillsdependonrepetitive
taskpractice.Repetitivetasktrainingencompassesanumberofideas;firstlythatrepetitivepractice
earlyafterstrokemayleadtobeneficialneuroplasticchangeswithinthebrain;secondlythat
repetitivepracticereducesweakness;thirdly,thatcomplexmovementscanbebrokendowninto
theircomponentsallowingpracticeofsimpleelementsbeforeincorporatingtheentiremovement;
fourthly,thatvaryingtaskcomplexityandtrainingschedules(distributedpractice,contextual
interference)promotesmotorlearningandgeneralisationtoreallifesituationsandretentionof
skills;andfifthlythatfeedbackiscriticaltolearningthemotorskills.
Inadditionstrengthtrainingcanbeusedtoaddressthesecondarymuscleweaknessthatarisesasa
resultofinactivity.Theunderlyingmechanismsofneuromuscularweaknessafterstrokepossibly
includeatrophyoftypeIIfibers,increasedproportionoftypeIfibers,lossofmotorunits,collateral
reinnervation,andalteredfiringofmotorunitgroups.Itisthoughtthatremodellingofmotorunits
occursinthemonthsafterstrokeanditmaybepossibletoenhancethisprocesswiththerapies
directedtowardincreasingmusclestrengthandthusfunctionalability.
Inpracticethedistinctionbetweenstrengthtrainingandrepetitivetaskpracticemaybelessclear,
forexample,treadmilltrainingwithbodyweightsupportmaybeusedtofacilitateabettergait
patternwhilebuildingstrengthandendurance.
13.1 Strengthtraining
Decreasedmusclepoweriscommonafterstroke;thismaybeduetocompromisedmusclefunction
poststroke,compoundedbyinactivityasaconsequenceoflimitedphysicalmobility.Decreased
musclepowerlimitspatientsabilitiesinactivitiesofdailylivingandmayleadtoamoregeneralised
lossoffitness.
NationalClinicalGuidelineCentre,2013.
313
StrokeRehabilitation
Movement
Strengthtrainingthroughrepetitivepracticemayrepresentoneapproachtoimprovingupperand
lowerfunctionafterstroke.
13.1.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
strengthtrainingversususualcareonimprovingfunctionandreducingdisability?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
Upperlimbstrengthtrainingand/or
Lowerlimbstrengthtraining
TrunkTypesofinterventionsinclude:weighttraining,resistance
training,IsometricandIsotonicexercises,circuittrainingfor
strength
13.1.1.1
Comparison
Usualcare
Outcomes
UpperLimb
MRCScale
NewtonMetres
FuglmeyerAssessment
ActionResearchArmTest(ARAT)
FunctionalIndependenceMeasurement(FIM)
BarthelIndex
Adverseeventspainorspasticity
LowerLimb/Trunk
TimedUpandGoTest
Anytimedwalk
Walkingdistance
FunctionalIndependenceMeasure(FIM)
BarthelIndex
Adverseeventsfalls,painorspasticity
NewtonMetres
ClinicalEvidenceReview
SearcheswereconductedforsystematicreviewsandRCTscomparingtheclinicaleffectivenessof
strengthtrainingwithusualcaretoimprovefunctionandreducedisabilityforadultsandyoung
people16orolderwhohavehadastroke.Onlystudieswithaminimumsamplesizeof20
participants(10ineacharm)wereselected.NineRCTswereidentified.
Table84summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
Table84: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
47
Cooke,2010
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Inpatientswithstroke
(113weekspost
strokeonset)who
wereindependently
mobile(withor
withoutaids)priorto
theindexstrokeand
wereabletofollowa1
stagecommand.
Functionalstrength
trainingplus
conventional
physiotherapy;
focusedon
repetitive,
progressiveresistive
exerciseduringgoal
orientedfunctional
Routine
conventional
physiotherapy
byaclinical
physiotherapist
includedsoft
tissue
mobilisation,
facilitationof
NationalClinicalGuidelineCentre,2013.
314
Walkingspeed
(m/sec)
Kneeflexion
peaktorque
Kneeextension
peaktorque
StrokeRehabilitation
Movement
STUDY
POPULATION
Donaldson,
200968
Strokepatients(mean
timeafterstroke:20.2
days(SD=14.0)).
Flansbjer,
200884
Kim,2001136
INTERVENTION
activity.All
additionaltherapy
wasdeliveredusing
standardized
treatmentschedules
forupto1hour,4
days/weekfor6
weeks.(N=36)
COMPARISON
muscleactivity,
facilitationof
coordinated
multijoint
movement,
tactileand
proprioceptive
input,resistive
exerciseand
functional
retraining.
(N=38)
OUTCOMES
Functionalstrength
training(FST)plus
conventional
physiotherapybya
physiotherapistupto
1hour,4days/week
for6weeks(total24
hours).FSTwas
basedonthe
positioningofthe
handandthenusing
ittomanipulate
objects.Treatment
wasprogressedusing
repetition,altering
thesizeandweight
ofitems,andusing
heavierweights.
(N=10)
Strokepatients(mean Progressive
timesincestroke:18.9 resistancetraining
months(SD7.9)forthe providedfor10
traininggroupand20
weekstwiceweekly
months(SD11.6)for
usingLeg
thecontrolgroup.
Extension/Curl
Rehabilitation
exercisemachine.
Each(N=15)
Conventional
physiotherapy
instructedbya
physiotherapist.
(N=10)
Usualcare:
patientswere
encouragedto
continueusual
dailyactivities
andtrainingbut
nottoengagein
anyprogressive
resistance
training(PRT).
(N=9)
Gait
performance
TimedUp
andGotest
Fastwalking
speed
6minute
walktest
(m).
Chronicstroke
survivors,withresidual
unilateralweakness;
aged50yearsorolder;
historyofasingle
strokeatleast6
monthsbefore
participatinginthe
study.
Thesameas
intervention
exceptthe
resisted
contractions
replacedwith
passiverangeof
motion
movements.
(N=10)
Selfselectedgait
speed/Habitual
gaitspeed(m/s)
Maximalgait
speed(m/s)
Maximalisokinetic
strengtheningwith
anisokinetic
dynamometer
consistingofthree
45minutesessions
perweekfor6
consecutiveweeks
foratotalof18
sessions.
Participantswere
askedtowalkattheir
mostcomfortable
speed(i.e.,self
NationalClinicalGuidelineCentre,2013.
315
ActionResearch
ArmTest(ARAT)
HandGripforce
(N)
Pinchgrip
force(N)
Isometricelbow
flexionforce(N)
Isometricelbow
extensionforce
(N).
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
selectedspeed)for5
trialsandthen
safelyasfastas
possible(i.e.,
maximalspeed)for
another5trialsalong
an8mwalkway.
(N=10)
COMPARISON
Langhammer,
2007142
Strokepatientwith
hemispherelesion:
Right(n):19intensive
exercise;19regular
exercise
Left(n):16intensive
exercise;21regular
exercise.
Functionalexercise
programme;high
intensityof
endurance,strength,
andbalance.
Individualised
trainingprogrammes
aimedatfunctional
improvementsbut
withvariations,for
examplegettingup
fromachair,walking
indoors,Nordic
walkingoutdoors,
stationarybicycling
andstairwalking,
wherethe
physiotherapist
monitoredthelevels
ofintensity.Total
treatmentperiodat
least80hours
(minimum20hours
everythirdmonthfor
thefirstyearafter
discharge;2
3timesperweekifat
homeorattending
private
physiotherapy
practiceordailyifin
rehabilitationward)
(N=35)
Usualcare:
physical
exercisesin
accordancewith
theroutinesin
thecommunity,
andonlyif
neededthey
wouldhavea
followup
(N=40)
Moreland,
2003178
Strokepatients:
Nonlacunar36(53%)
progressiveresistance;
36(55%)control
group.
Strengthtraining:
conventional
physiotherapyplus
progressive
resistanceexercises
withweightsatwaist
oronlower
extremities.
Useofankle
exercisertowhich
variableweights
appliedfor30minute
exercisesessions(2
setsof10
Conventional
physiotherapy
butnoexternal
resistancewas
appliedwith
weights.(N=65)
NationalClinicalGuidelineCentre,2013.
316
OUTCOMES
BarthelIndex
(BI)
Gripstrength
(paretic,non
paretichand)
2minutewalk
test(m)
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
repetitions)3timesa
week.(N=68)
COMPARISON
Ouellette,
2004195
Patientsaged50
years;6monthsto6
yearsfollowingasingle
unilateralmildto
moderatestrokewith
residuallower
extremityhemiparesis,
communitydwelling;
independent
ambulationwithor
withoutanassistive
device
Progressive
ResistanceTraining:
Subjectsperformed
seatedbilateralleg
press(LP);unilateral
pareticandnon
pareticlimbknee
extension(KE),
unilateralankle
dorsiflexion(DF),and
planterflexion(PF)3
timesperweekfor
12weeks.
Habitualand
maximalGait
VelocitiesSubjects
wereinstructedto
walk10metresat
theirnormaland
maximalvelocity
respectively.
(N=21)
Bilateralrange
ofmotion
(ROM)and
upperbody
flexibility
exercises
performed3
timesperweek.
(N=21)
Sixminutewalk
(min)
HabitualGait
Velocity/self
gaitspeed
(m/sec)
MaximalGait
Velocity(m/sec)
Winstein,
2004285
Strokepatients(mean
timesincestroke16
days(SD17.7).85%of
patientshadischaemic
typeofstroke.
Strengthandmotor
controltraining(ST):
resistanceto
availablearm
motion.(N=21)
Standardcare
byan
occupational
therapist:
muscle
facilitation
exercises,
neuromuscular
electric
stimulation
applied
primarilyfor
shoulder
subluxation,
stretching
exercises,
activitiesof
dailyliving(self
carewherethe
upperlimbwas
usedasan
assist)and
caregiver
training.(N=21)
FuglMeyer
assessment
(FMA):
rangeof
movement
pain
sensory
motorfunction
FIM:
mobility
selfcare
40participantswere
assessedat2weeks
afterstrokeandwere
thosewithaconfirmed
diagnosisofafirst
unilateralstroke(MRI
orCT),noimpairment
Usualcare
Usualcare
physiotherapyplus
physiotherapy
familymediated
(N=20)
exerciseintervention.
IndividualisedFAME
programswere
conductedfor35
Galvin201188
NationalClinicalGuidelineCentre,2013.
317
OUTCOMES
Lowerlimbsection
oftheFuglMeyer
Assessment(LL
FMA).
Motorassessment
Scale(MAS),
BergBalanceScale
StrokeRehabilitation
Movement
STUDY
POPULATION
ofcognition(24of30
ontheMiniMental
StateExamination),
participatingina
physiotherapyprogram
andafamilymember
willingtoparticipatein
theprogram.To
controlfor
heterogeneity
individualswhoscored
from3.2to5.2onthe
OrpingtonPrognostic
Scalewererecruited.
Thefamilymember
hadtobemedically
stableandphysically
abletoassistinthe
deliveryofexercises.
INTERVENTION
minutesdailyatthe
bedsidewiththe
assistanceoftheir
nominatedfamily
member.The
emphasisofthe
lowerlimbexercise
interventionwason
achievingstability
andimprovinggait
velocityandlower
limbstrengthbased
onpatternsand
wereprogress
accordingtothe
individualsability
(N=20)
NationalClinicalGuidelineCentre,2013.
318
COMPARISON
OUTCOMES
(BBS),
6MWT
BarthelIndex(BI)
StrokeRehabilitation
Movement
Comparison:Functionalstrengthtraining(upper,lowerlimb)versususualcare
Table85:UpperlimbfunctionalstrengthtrainingusualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Indirectnes
Inconsistency s
Imprecision
Upper
limb
Functiona
lstrength
training
Mean
(SD)
Effect
Usual
care
Mean
(SD)
84.5
(23.9)
Mean
differenc
e
(95%CI)
Mean
Differen
ce(MD)
(95%CI)
91.2
(19.9)
6.7(
17.55to
4.15)
MD6.7
lower
(17.55
lowerto
4.15
higher)
Low
0.55
(0.42)
0.55
(0.41)
0(0.2to
0.2)
MD0
higher
(0.2
lowerto
0.2
higher)
Moderate
0.77
(0.35)
0.81
(0.31)
0.04(
0.2to
0.12)
MD0.04
lower
(0.2
lowerto
0.12
higher)
Low
Confidence
(ineffect)
BarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Langhammer
142
2007
RCTdouble
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
Serious
indirectness imprecision(
b)
Gripstrengthparetichand(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Langhammer
2007142
RCTdouble
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
Noserious
indirectness imprecision
Gripstrengthnonparetichand(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Langhammer
2007142
RCTdouble
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
Serious
indirectness imprecision(
c)
NationalClinicalGuidelineCentre,2013.
319
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Indirectnes
Inconsistency s
Imprecision
Upper
limb
Functiona
lstrength
training
Mean
(SD)
Effect
Usual
care
Mean
(SD)
80.8
(29.5)
Mean
differenc
e
(95%CI)
Mean
Differen
ce(MD)
(95%CI)
87.7
(27.8)
6.9(
21.05to
7.25)
MD6.9
lower
(21.05
lowerto
7.25
higher)
Verylow
0.63
(0.46)
0.67
(0.43)
0.04(
0.26to
0.18)
MD0.04
lower
(0.26
lowerto
0.18
higher)
Low
0.87
(0.4)
0.99
(0.32)
0.12(
0.3to
0.06)
MD0.12
lower
(0.3
lowerto
0.06
higher)
Low
43.6
(18.9)
45
(13.93
)
1.4(
16.58to
13.78)
MD1.4
lower
(16.58
Low
Confidence
(ineffect)
BarthelIndex(1yearfollowup)(Betterindicatedbyhighervalues)
1
Langhammer
2007142
RCTdouble
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
Veryserious
indirectness imprecision(
e)
Gripstrengthparetichand(1yearfollowup)(Betterindicatedbyhighervalues)
1
Langhammer
2007142
RCTdouble
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
Serious
indirectness imprecision(
c)
Gripstrengthnonparetichand(1yearfollowup)(Betterindicatedbyhighervalues)
1Langhammer
2007142
RCTdouble
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
Serious
indirectness imprecision(
c)
ActionResearchArmTest(ARAT)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Donaldson200968
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
Veryserious
indirectness imprecision(
d)
NationalClinicalGuidelineCentre,2013.
320
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Indirectnes
Inconsistency s
Imprecision
Upper
limb
Functiona
lstrength
training
Mean
(SD)
Effect
Usual
care
Mean
(SD)
Mean
differenc
e
(95%CI)
Mean
Differen
ce(MD)
(95%CI)
lowerto
13.78
higher)
58.5
(60.18)
64.75
(39.25
)
6.25(
52.41to
39.91)
MD6.25
lower
(52.41
lowerto
39.91
higher)
Low
25.8
(21.26)
24.5
(19.7)
1.3(
17.67to
20.27)
MD1.3
higher
(17.67
lowerto
20.27
higher)
Low
59.5
(44.69)
75
(38.67
)
15.5(
54.04to
23.04)
MD15.5
lower
(54.04
lowerto
23.04
higher)
Low
Confidence
(ineffect)
Gripforce(N)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Donaldson200968
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
Veryserious
indirectness imprecision(
c)
Pinchforce(N)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Donaldson200968
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
Veryserious
indirectness imprecision(
c)
Elbowflexionforce(N)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Donaldson200968
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
Veryserious
indirectness imprecision(
c)
Elbowextensionfore(N)(3monthsfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
321
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Indirectnes
Inconsistency s
1
Donaldson200968
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Imprecision
Noserious
Veryserious
indirectness imprecision(
c)
Upper
limb
Functiona
lstrength
training
Mean
(SD)
49.2
(34.19)
Effect
Usual
care
Mean
(SD)
Mean
differenc
e
(95%CI)
Mean
Differen
ce(MD)
(95%CI)
68.63
(39.61
)
19.43(
54.11to
15.25)
MD
19.43
lower
(54.11
lowerto
15.25
higher)
(a)
Allocationconcealmentnotreported.16%losttofollowup.
MeandifferencedidnotreachtheagreedMIDof9.25points.
(c)
ConfidenceintervalcrossedoneendofdefaultMID.
(d)
MeandifferencedidnotreachtheagreedMIDof12and17pointsfortheaffecteddominantandnondominantsidesrespectively
(e)
ConfidenceintervalcrossedbothendsofdefaultMID.
(b)
NationalClinicalGuidelineCentre,2013.
322
Confidence
(ineffect)
Low
StrokeRehabilitation
Movement
Comparison:lowerlimbfunctionalstrengthtrainingversususualcare
Table86: LowerlimbfunctionalstrengthtrainingversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitations
Inconsistency
Indirectness
Usual
care
Mea
n
(SD)
Effect
Imprecision
Lower
limb
Functional
strength
training
Mean(SD)
Noserious
imprecision
0.46
(0.37)
0.44
(0.39
)
0.02(
0.19to
0.23)
MD0.02
higher
(0.19
lowerto
0.23
higher)
Moderate
Serious
imprecision(b
)
29.4
(21.2)
25.2
(22.9
)
4.2(9.36
to17.76)
MD4.2
higher
(9.36
lowerto
17.76
higher)
Low
42.1
(27.5)
37.9
(27.8
)
4.2(
12.71to
21.11)
MD4.2
higher
(12.71
lowerto
21.11
higher)
Low
Relative
Mean
differenc
e(MD)
(95%CI)
Mean
Differen
ce(MD)
(95%CI)
Confidence
(ineffect)
Walkingspeed(m/sec)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Cooke201047
RCTsingle
blinded
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Kneeflexionpeaktorque(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Cooke201047
RCTsingle
blinded
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Kneeextensionpeaktorque(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Cooke201047
RCTsingle
blinded
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
(a)26%losttofollowupat12weeks.
(b)ConfidenceintervalcrossedoneendofdefaultMID
NationalClinicalGuidelineCentre,2013.
323
StrokeRehabilitation
Movement
Comparison:Resistancetrainingversususualcare
Table87:ResistancetrainingversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectnes
s
Imprecision
Usual
Resistance care
training
Mean
Mean(SD) (SD)
Mean
differenc
e(MD)
(95%CI)
15(7.14)
14.1
(7.58)
0.90(
3.66,
5.46)
MD0.90
higher
(3.66
lowerto
5.46
higher)
Low
2.44
(1.82)
5.67
(5.47)
3.23(
6.14to
0.32)
MD3.23
lower
(6.14to
0.32
lower)
Low
16.15
(5.81)
17(5.17)
0.85(
4.26,
2.56)
MD0.85
lower
(4.26
lowerto
2.56
higher)
Low
Confidenc
Mean
Differenc e
e(95%CI) (ineffect)
FIMmobilitychanges(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(b)
FIMmobilitychanges(9monthsfollowup)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(b)
FIMselfcarechanges(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(b)
NationalClinicalGuidelineCentre,2013.
324
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectnes
s
Imprecision
Usual
Resistance care
training
Mean
Mean(SD) (SD)
Mean
differenc
e(MD)
(95%CI)
2.75
(4.34)
3.32(
6.48to
0.16)
Confidenc
Mean
Differenc e
e(95%CI) (ineffect)
FIMselfcarechanges(9monthsfollowup)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(b)
6.07
(4.62)
MD3.32
lower
(6.48to
0.16
lower)
Low
0.15(
1.37,
1.07)
MD0.15
lower
(1.37
lowerto
1.07
higher)
Low
0.33
(1.45)
1.8(
3.43to
0.17)
MD1.8
lower
(3.43to
0.17
lower)
Low
0.6
(1.79)
0.10(
1.38,
1.18)
MD0.10
lower
(1.38
lowerto
1.18
higher)
Low
UpperextremityFuglMeyerAssessmentRangeofMovementchanges(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
0.75(2)
0.6
(1.93)
UpperextremityFuglMeyerAssessmentRangeofMovementchanges(9monthsfollowup)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
2.13
(2.96)
UpperextremityFuglMeyerAssessmentPainchanges(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
0.7(2.3)
NationalClinicalGuidelineCentre,2013.
325
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectnes
s
Imprecision
Usual
Resistance care
training
Mean
Mean(SD) (SD)
Mean
differenc
e(MD)
(95%CI)
Confidenc
Mean
Differenc e
e(95%CI) (ineffect)
UpperextremityFuglMeyerAssessmentPainchanges(9monthsfollowup)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
1.19(4)
1
(2.88)
0.19(
2.63to
2.25)
MD0.19
lower
(2.63
lowerto
2.25
higher)
Low
0.75
(1.33)
0.55(
0.59,
1.69)
0.55
higher
(0.59
lowerto
1.69
higher)
Low
0.07
(1.03)
0.18(
0.44to
0.8)
MD0.18
higher
(0.44
lowerto
0.8
higher)
Low
9.05(7.6) 9.15
(2.35,
15.95)
MD9.15
higher
(2.35to
15.95
higher)
Low
UpperextremityFuglMeyerAssessmentsensorychanges(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
1.3(2.23)
UpperextremityFuglMeyerAssessmentsensorychanges(9monthsfollowup)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
0.25
(0.68)
UpperextremityFuglMeyerAssessmentmotorfunctionchanges(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTunblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
18.2
(13.54)
NationalClinicalGuidelineCentre,2013.
326
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectnes
s
Imprecision
Usual
Resistance care
training
Mean
Mean(SD) (SD)
Mean
differenc
e(MD)
(95%CI)
Confidenc
Mean
Differenc e
e(95%CI) (ineffect)
UpperextremityFuglMeyerAssessmentmotorfunctionchanges(9monthsfollowup)(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCT
unblinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(c)
5.38
(9.11)
8.33
(11.26)
2.95(
10.19to
4.29)
MD2.95
lower
(10.19
lowerto
4.29
higher)
Low
23.6
(11.1)
26.7
(18.9)
3.1(
16.67to
10.47)
MD3.1
lower
(16.67
lowerto
10.47
higher)
Verylow
16.1
(9.9)
19.4
(17.8)
3.3(
15.96to
9.36)
MD3.3
lower
(15.96
lowerto
9.36
higher)
Low
58.6
(52.7)
63.2
(49.1)
4.6(
23.98to
14.78)
MD4.6
lower
(23.98
lowerto
14.78
Moderate
Timedupandgotest(sec)(5monthsfollowup)(betterindicatedbylowervalues)
1Flansbjer
200884
RCTsingle
blinded
Serious
limitations
(e)
Noserious
inconsistency
Noserious Veryserious
indirectness imprecision(d)
Fastwalkingspeed(m/sec)(5monthsfollowup)(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blinded
Serious
limitations
(e)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(h)
2minutewalktest(m)(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Moreland
2003178
RCTsingle
blinded
Serious
limitations(g
)
Noserious
inconsistency
Noserious Noserious
indirectness imprecision
NationalClinicalGuidelineCentre,2013.
327
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Usual
Resistance care
training
Mean
Mean(SD) (SD)
Mean
differenc
e(MD)
(95%CI)
239.1
(30.3)
234.8
(36.9)
4.30(
16.12,
24.72)
Veryserious
imprecision(f)
251(144)
Noserious Serious
indirectness imprecision(j)
Kim:0.04
(0.13)
Ouellette:
0.64(0.08)
Indirectnes
s
Imprecision
Confidenc
Mean
Differenc e
e(95%CI) (ineffect)
higher)
6minutewalktest(m)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Ouellette,
2004195
RCTSingle
Serious
Noserious
blinded limitation(e) inconsistency
Noserious Noserious
indirectness imprecision
MD4.3
higher
(16.12
lowerto
24.72
higher)
Moderate
240(140) 11.00(
105.95,
127.95)
MD11
higher
(105.95
lowerto
127.95
higher)
Verylow
Kim:0.09
(0.07)
Ouellette
:0.64
(0.09)
MD0.01
lower
(0.06
lowerto
0.03
higher
Low
6minutewalktest(m)(5monthsfollowup)(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blinded
Serious
limitations(
e)
Noserious
inconsistency
Noserious
indirectness
Selfselected/Habitualgaitspeed(Betterindicatedbyhighervalues)
2
Kim,2001136,
Ouellette,
2004195
RCTdouble
blinded(Kim)
RCTsingle
blinded
(Ouellette)
Serious
limitations(i
)
Noserious
inconsistency
NationalClinicalGuidelineCentre,2013.
328
0.01(
0.06,
0.03)
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectnes
s
Imprecision
Usual
Resistance care
training
Mean
Mean(SD) (SD)
Mean
differenc
e(MD)
(95%CI)
Kim:0.05
(0.09)
Ouellette:
0.86(0.11)
0.01(
0.07,
0.04)
Confidenc
Mean
Differenc e
e(95%CI) (ineffect)
Maximalgaitspeed(Betterindicatedbyhighervalues)
2
Kim,2001136,
Ouellette,
2004195
RCTdouble
blinded(Kim)
RCTsingle
blinded
(Ouellette)
Serious
limitations(i
)
Noserious
inconsistency
Noserious Serious
indirectness imprecision(j)
Kim:0.07
(0.08)
Ouellette
:0.87
(0.12)
(a)
Unblindedstudy.Unclearrandomizationandinadequateallocationconcealment.27%losttofollowupat9months.
MeandifferencedidnotreachtheagreedMIDof17pointsforthemotorscaleandthe3pointsforthecognitivescale.
(c)
MeandifferencedidnotreachtheagreedMIDof10%betweentheinterventionandcontrolgroups.
(d)
MeandifferencedidnotreachtheagreedMIDof10secbetweentheinterventionandcontrolgroups.
(e)
Allocationconcealmentnotreported;Nodetailsofrandomisation
(f)
ConfidenceintervalcrossedbothendsofdefaultMID.
(g)
Unclearblinding;20%losttofollowupat6months.
(h)
ConfidenceintervalcrossedoneendofthedefaultMID.
(i)Unclearrandomisation;unclearallocationconcealment
(J)MeandifferencedidnotreachtheagreedMIDof0.2m/secbetweentheinterventionandcontrolgroups.
(b)
NationalClinicalGuidelineCentre,2013.
329
MD0.01
lower
(0.07
lowerto
0.04
higher
Low
StrokeRehabilitation
Movement
Comparison:Familymediatedexerciseinterventionversususualcare(physiotherapy)
Table88: GRADEcharacteristicsandclinicalsummaryoffindings
Qualityassessment
Noof
studies
Design
SummaryofFindings
Riskof
bias
Inconsistency
Indirectness
Imprecision
Family
mediated
strength
trainingMean
(SD)
Usualcare
(physiotherapy)
Mean(SD)
Effect
Mean
Differenc
e
(95%CI)
Mean
Difference
(95%CI)
Confidence
(ineffect)
LowerlimbFuglMeyerAssessmentPostintervention(followup8weeks;measuredwith:Meanchangefrombaseline;rangeofscores:036;Betterindicatedby
highervalues)
1
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(b)
9.5(9.9)
1.75(6.3)
7.75
(2.61,
12.89)
MD7.75higher LOW
(2.61to12.89
higher)
LowerlimbFuglMeyerAssessmentFollowupafter3months(measuredwith:Meanchangefrompostintervention;rangeofscores:036;Betterindicatedbyhigher
values)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
1.6(2.4)
1.3(5.2)
0.3(
2.21,
2.81)
MD0.3higher
(2.21lowerto
2.81higher)
MODERAT
E
MotorassessmentscalePostintervention(followup8weeks;measuredwith:Meanchangefrombaseline;rangeofscores:048;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(c)
11.9(7.8)
4.75(6.2)
7.15
(2.78,
11.52)
MD7.15higher LOW
(2.78to11.52
higher)
MotorassessmentscaleFollowupafter3months(measuredwith:Meanchangefrompostintervention;rangeofscores:048;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(c)
1.8(3.8)
0.7(2.6
1.1(
0.92,
NationalClinicalGuidelineCentre,2013.
330
MD1.1higher
(0.92lowerto
LOW
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
SummaryofFindings
Riskof
bias
Inconsistency
Indirectness
Imprecision
Family
mediated
strength
trainingMean
(SD)
Usualcare
(physiotherapy)
Mean(SD)
single
(assessor)
blinded
Effect
Mean
Mean
Differenc Difference
e
(95%CI)
(95%CI)
3.12)
3.12higher)
Confidence
(ineffect)
BergBalanceScalePostintervention(followup8weeks;measuredwith:Meanchangefrombaseline;rangeofscores:056;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(c)
22.8(18.1)
9(9)
13.8
(4.94,
22.66)
MD13.8higher LOW
(4.94to22.66
higher)
BergBalanceScaleFollowupafter3months(measuredwith:Meanchangefrompostintervention;rangeofscores:056;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(d)
0.9(2.5)
1.8(8.5)
0.9(
4.78,
2.99)
MD0.9lower
(4.78lowerto
2.98higher)
LOW
6MinuteWalkTestPostintervention(followup8weeks;measuredwith:Meanchangefrombaselinemetres;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
1
noserious
inconsistency
noserious
indirectness
noserious
imprecision
164.1(128.7)
47.2(50.6)
116.9
(56.29,
177.51
)
MD116.9
higher(56.29
to177.51
higher)
MODERAT
E
6MinuteWalkTestFollowupafter3months(measuredwith:Meanchangefrompostinterventionmetres;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
serious(e)
39.8(55.4)
3.5(32.7)
43.3
MD43.3higher LOW
(15.11, (15.11to71.49
71.49) higher)
NationalClinicalGuidelineCentre,2013.
331
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
SummaryofFindings
Riskof
bias
Inconsistency
Indirectness
Imprecision
Family
mediated
strength
trainingMean
(SD)
Usualcare
(physiotherapy)
Mean(SD)
Effect
Mean
Differenc
e
(95%CI)
Mean
Difference
(95%CI)
Confidence
(ineffect)
BarthelIndexPostintervention(followup8weeks;measuredwith:Meanchangefrombaseline;rangeofscores:0100;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
32.3(24)
16.3(14.2)
16
(3.78,
28.22)
MD16higher
(3.78to28.22
higher)
MODERAT
E
BarthelIndexFollowupafter3months(measuredwith:Meanchangefrompostintervention;rangeofscores:0100;Betterindicatedbyhighervalues)
Galvin
201188
randomise
dtrials
single
(assessor)
blinded
serious
(a)
noserious
inconsistency
noserious
indirectness
very
serious(f)
3.8(8.3)
1.5(11.6)
2.3(
3.95,
8.55)
(a)
MD2.3higher
(3.95lowerto
8.55higher)
VERYLOW
NationalClinicalGuidelineCentre,2013.
332
StrokeRehabilitation
Movement
13.1.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingstrengthtrainingwithusualcarewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Thecostofprovidingstrengthtraining(Table89)wasestimatedbasedontheresourcesusedintwo
studies(Flansbjer,200884andCooke,201047)includedintheclinicalreview.Theremainingstudies
includedintheclinicalreview68,142,178,285werenotusedastheydidnotprovidesufficientinformation
aboutthetypeoramountofresourcesused.
Table89: Interventioncostsmusclepowertraining
Source
Unitcosts(a)
Resourceuse
Incrementalcost
(interventionoverusual
care)
Resistancetraining
Flansbjer
200884
90minutesessionsbyphysiotherapist
includingprogressiveresistance
training(b)(<6minutes)for10weeks,
twiceweekly(1.5hrsx20=30hrs)
45perhourofclient
contact(band6)
1350
45perhourofclient
contact(band6)
1080
Lowerlimbstrengthtraining
Cooke
201047
1hoursessionbyphysiotherapist,for
4daysperweekfor6weeks(1hrx24
=24hrs)
(a) EstimatedbasedondataandmethodsfromthePersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcarereportand
51
AgendaforChangesalaryband6 (typicalsalarybandidentifiedbyclinicalGDGmembers).
(b) Trainingwasdoneusingalegextension/curlrehabilitationexercisemachine.
TheseestimatesrepresentthecostofmusclepowertrainingprovidedbyNHSorPSSstaffinthe
earlyphaseafterstroke.However,inlaterstages,strengthtrainingmaybehandedovertoan
appropriatelyqualifiedgyminstructorandthiswouldhavelowercosts.
13.1.1.3
Evidencestatements
Clinicalevidencestatements
Onestudy142of75participantsfoundnosignificantdifferenceinBarthelIndexbetweenthose
participantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceivedusualcareata
followupof6months(LOWCONFIDENCEINEFFECT).
Onestudy142of75participantsfoundnosignificantdifferenceingripstrength(paretichand)
betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceived
usualcareatafollowupof6months(MODERATECONFIDENCEINEFFECT
NationalClinicalGuidelineCentre,2013.
333
StrokeRehabilitation
Movement
Onestudy142of75participantsfoundthattherewasnosignificantdifferenceingripstrength(non
paretichand)betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewho
receivedusualcareatafollowupof6months(LOWCONFIDENCEINEFFECT).
Onestudy142of75participantsfoundnosignificantdifferenceinBarthelIndexbetweenthose
participantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceivedusualcareata
followupof1year(VERYLOWCONFIDENCEINEFFECT).
Onestudy142of75participantsfoundnosignificantdifferenceingripstrength(paretichand)
betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceived
usualcareatafollowupof1year(LOWCONFIDENCEINEFFECT).
Onestudy142of75participantsfoundnosignificantdifferenceingripstrength(nonparetichand)
betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceived
usualcareatafollowupof1year(LOWCONFIDENCEINEFFECT).
Onestudy68of20participantsfoundnosignificantdifferenceinActionResearchArmTest(ARAT)
betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceived
usualcareat3monthsfollowup(LOWCONFIDENCEINEFFECT)
Onestudy68of20participantsfoundthattherewasnosignificantdifferenceinGripforce(N)
betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceived
usualcareat3monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy68of20participantsfoundthattherewasnosignificantdifferenceinpinchforce(N)
betweenthoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceived
usualcareat3monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy68of20participantsfoundnosignificantdifferenceinElbowflexionforce(N)between
thoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceivedusualcareat
3monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy68of20participantsfoundnosignificantdifferenceinElbowextensionforce(N)between
thoseparticipantswhoreceivedupperlimbfunctionaltrainingandthosewhoreceivedusualcareat
3monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy47of74participantsfoundnosignificantdifferenceinwalkingspeed(m/sec)between
thoseparticipantswhoreceivedlowerlimbfunctionaltrainingandthosewhoreceivedusualcareat
afollowupof3months(MODERATECONFIDENCEINEFFECT).
Onestudy47of74participantsfoundnosignificantdifferenceinkneeflexionpeaktorquebetween
thoseparticipantswhoreceivedlowerlimbfunctionaltrainingandthosewhoreceivedusualcareat
afollowupof3months(LOWCONFIDENCEINEFFECT).
Onestudy47of74participantsfoundnosignificantdifferenceinkneeextensionpeaktorquebetween
thoseparticipantswhoreceivedlowerlimbfunctionaltrainingandthosewhoreceivedusualcareat
afollowupof3months(LOWCONFIDENCEINEFFECT).
Onestudy285of43participantsfoundthattherewasnosignificantdifferenceinFIMmobilityscore
betweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareafter
treatment(LOWCONFIDENCEINEFFECT).
Onestudy285of43participantsfoundthatusualcarewasassociatedwithstatisticallysignificant
improvementinFIMmobilityscorecomparedtoresistancetrainingatafollowupof9months,
althoughthisdifferencewasnotofclinicalsignificance(LOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
334
StrokeRehabilitation
Movement
Onestudy285of43participantsfoundthattherewasnosignificantdifferenceinFIMselfcarescore
betweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareafter
treatment(LOWCONFIDENCEINEFFECT).
Onestudy285of43participantsfoundthatusualcarewasassociatedwithstatisticallysignificant
improvementinFIMselfcarescorecomparedtoresistancetrainingat9monthsfollowup,
althoughthisdifferencewasnotofclinicalsignificance(LOWCONFIDENCEINEFFECT).
Onestudy285of43participantsfoundthattherewasnosignificantdifferenceinFuglMeyerROM
scorebetweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcare
aftertreatment(LOWCONFIDENCEINEFFECT).
Onestudy285of64participantsfoundthatusualcarewasassociatedwithstatisticallysignificant
improvementinFuglMeyerrangeofmotionscorecomparedtoresistancetrainingatafollowup
of9months(LOWCONFIDENCEINEFFECT).
Onestudy285of43participantsfoundthattherewasnosignificantdifferenceinFuglMeyerpain
scorebetweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcare
aftertreatmentandatafollowupof9months(LOWCONFIDENCEINEFFECT).
Onestudy285of64participantsfoundnosignificantdifferenceinFuglMeyersensoryscore
betweenthoseparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcare
aftertreatmentandatafollowupof9months(LOWCONFIDENCEINEFFECT).
Onestudy285of43participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinFIMmotorfunctionscorecomparedtousualcareaftertreatment,
althoughthisdifferencewasnotofclinicalsignificance(LOWCONFIDENCEINEFFECT).
Onestudy285of64participantsfoundnosignificantdifferenceinFuglMeyermotorfunctionscore
betweenthoseparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareata
followupof9months(LOWCONFIDENCEINEFFECT).
Onestudy84of24participantsfoundnosignificantdifferenceingaitperformanceassessedbytimed
upandgotest(sec)betweenthoseparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareatafollowupof5months(VERYLOWCONFIDENCEINEFFECT).
Onestudy84of24participantsfoundnosignificantdifferenceingaitperformanceassessedbyfast
gaitspeed(10m/sec)betweenthoseparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareatafollowupof5months(LOWCONFIDENCEINEFFECT)
Onestudy178of133participantsfoundnosignificantdifferencein2minutewalktestbetweenthose
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareatafollowupof6
months(MODERATECONFIDENCEINEFFECT).
Onestudy195of42participantsfoundnosignificantdifferenceingaitperformanceassessedby6
minutewalktestbetweenthoseparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareatafollowupof3months(MODERATECONFIDENCEINEFFECT).
Onestudy84,of24participantsfoundnosignificantdifferenceingaitperformanceassessedby6
minutewalktestbetweenthoseparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareatafollowupof5months(VERYLOWCONFIDENCEINEFFECT).
Twostudies136,195of62participantsfoundnosignificantdifferenceinselfselected/habitualgait
speedbetweenthoseparticipantswhoreceivedresistancetrainingandthosewhoreceivedusual
care(LOWCONFIDENCEINEFFECT)
NationalClinicalGuidelineCentre,2013.
335
StrokeRehabilitation
Movement
Twostudies136,195of62participantsfoundnosignificantdifferenceinmaximalgaitspeedbetween
thoseparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcare(LOW
CONFIDENCEINEFFECT)
Familymediatedexercise(FAME)interventioncomparedtousualcare(physiotherapy)
Onestudy88of40participantsfoundasignificantimprovementinLowerLimbFuglMeyermotor
functionassociatedwiththeFAMEinterventioncomparedtousualcareattheendofthe8week
intervention(LOWCONFIDENCEINEFFECT).Thisimprovementwasnotmaintainedattheendofthe
3monthsfollowupperiod(MODERATECONFIDENCEINEFFECT).
Onestudy88of40participantsfoundasignificantimprovementineverydaymotorfunction(as
assessedbytheMotorAssessmentScale)associatedwiththeFAMEinterventioncomparedtousual
careattheendofthe8weekintervention(LOWCONFIDENCEINEFFECT).Thisimprovementwasnot
maintainedattheendofthe3monthsfollowupperiod(LOWCONFIDENCEINEFFECT).
Onestudy88of40participantsfoundasignificantimprovementinperson'sstaticanddynamic
balanceabilities(asassessedbytheBergBalanceScale)associatedwiththeFAMEintervention
comparedtousualcareattheendofthe8weekintervention(LOWCONFIDENCEINEFFECT).This
improvementwasnotmaintainedattheendofthe3monthsfollowupperiod(LOWCONFIDENCEIN
EFFECT).
Onestudy88of40participantsfoundasignificantimprovementinfunctionalexercisecapacity(as
assessedbythe6minutewalktest)associatedwiththeFAMEinterventioncomparedtousualcareat
theendofthe8weekintervention(MODERATECONFIDENCEINEFFECT).Thisimprovementwasstill
significantattheendofthe3monthsfollowupperiodbuttheeffectwasnotaslargeaspost
intervention(LOWCONFIDENCEINEFFECT).
Onestudy88of40participantsfoundasignificantimprovementintheperformanceinactivitiesof
dailyliving(asassessedbytheBarthelIndex)associatedwiththeFAMEinterventioncomparedto
usualcareattheendofthe8weekintervention(MODERATECONFIDENCEINEFFECT).This
improvementwasstillsignificantattheendofthe3monthsfollowupperiodbuttheeffectwasnot
aslargeaspostintervention(LOWCONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.1.2
Recommendationsandlinktoevidence
77.Providephysiotherapyforpeoplewhohaveweaknessintheir
trunkorupperorlowerlimb,sensorydisturbanceorbalance
difficultiesafterstrokethathaveaneffectonfunction.
78.Peoplewithmovementdifficultiesafterstrokeshouldbe
treatedbyphysiotherapistswhohavetherelevantskillsand
traininginthediagnosis,assessmentandmanagementof
movementinpeoplewithstroke.
79.Treatmentforpeoplewithmovementdifficultiesafterstroke
shouldcontinueuntilthepersonisabletomaintainor
progressfunctioneitherindependentlyorwithassistance
fromothers(forexample,rehabilitationassistants,family
members,carersorfitnessinstructors).
NationalClinicalGuidelineCentre,2013.
336
StrokeRehabilitation
Movement
80.Considerstrengthtrainingforpeoplewithmuscleweakness
afterstroke.Thiscouldincludeprogressivestrengthbuilding
throughincreasingrepetitionsofbodyweightactivities(for
example,sittostandrepetitions),weights(forexample,
progressiveresistanceexercise),orresistanceexerciseon
machinessuchasstationarycycles.
Relativevaluesofdifferent
outcomes
Therangeofoutcomesreflectedimpairment(force),activity(walking
speedanddistance,ActionResearchArmtest),anddependence(Barthel
IndexandFunctionalIndependenceMeasure).Improvementsin
strengthwouldbepostulatedtoleadtoimprovementsoffunctionand
thusmeasuresofmobility,activity,anddependenceareofpotentially
moreinterest.However,insmallstudiesmeasuresofimpairmentmay
beresponsivetotheintervention.
Adverseeventswerealsoregardedasanimportantoutcome,
particularlythedevelopmentofincreasedtone.TheGDGnotedthat
somehealthprofessionalshaveexpressedaconcernthatstrength
trainingmaybeassociatedwithanincreaseintonethatintime,may
leadtodeteriorationinfunction.Inthiscontext,theGDGconsideredit
importanttorecognisetheincidenceofdisablingspasticityinstroke
whichhasbeenreportedas4%byLumstrometal.161
Tradeoffbetweenclinical
benefitsandharms
Onestudy84statedtherewerenoinjuriesassociatedwiththeresistance
trainingexercisemachineused.Noneoftheotherstudiesreportedany
adverseeventsfromthestrengthtraininginterventions.TheGDG
agreedthatthattherewouldnotnormallybeanydetrimentaleffect
fromthesetypesofinterventions.
Weaknessoftheface,upperlimb,trunkandlowerlimbarecommon
deficitsafterstroke,Aswellasstrength,sensorydisturbanceand
balancedifficultiesimpactonmovement.Itwasfeltthattrained
physiotherapistswiththerelevantskillsandtraininginthediagnosis,
assessmentandmanagementofmovementinpeoplewithstrokeshould
regularlymonitorandtreatpeoplewithmovementdifficultiesuntilthey
areabletomaintainorprogressfunctioneitherindependentlyorwith
assistancefromothers(rehabilitationassistants,carers,fitness
instructorsetc.).Onestudyreportedsignificantimprovementsinmotor
andbalancefunctionassociatedwithstrengthtrainingusingfamily
membersascotrainers.Theevidenceforoutcomesfromthisstudywere
oflowtomoderatequality.However,improvementswerenot
maintainedoverathreemonthfollowupperiod.
Economicconsiderations
Nocosteffectivenessstudieswereidentifiedforthisquestion.Themain
differenceincostsbetweentheprovidingmusclepowertrainingand
usualcarewasduetotheamountofadditionalpersonneltimerequired.
Inaddition,theremayalsobesomedevicecosts,forexamplean
exercisemachinewasusedinoneofthestudiesincludedintheclinical
review.However,whenthecostofthemachineisspreadoverthe
lifetimeoftheequipmentandtheamountofusage,thecostperpatient
persessionisexpectedtobelow.BasedonresourceusefromCooke
(2010lowerlimbstrengthtrainingintervention)47theadditionalcostof
strengthtrainingoverusualcarewasestimatedtobe1080andbased
onFlansbjer(2008resistancetrainingintervention)84itwasestimated
tobe1350(personnelcostsonly).
TheGDGconsidereditlikelythatstrengthtrainingwouldbecost
effectiveasthepotentialimprovementforpatientsintermsofqualityof
lifefromimprovedfunctionwouldjustifyanyadditionalcostsofthe
NationalClinicalGuidelineCentre,2013.
337
StrokeRehabilitation
Movement
intervention.
Qualityofevidence
Manyofthestudieswerelimitedbysmallnumbers(maximumsample
size=133),somewerefeasibilitystudies,inadequatelypoweredand
durationoffollowupandtimesinceonsetofstrokevariedbetween
studies.Becauseofthewiderangeofdifferenttypesofstrengthtraining
andoutcomemeasuresincludedwithinthestudiesitwasnotpossibleto
carryoutanymetaanalysisandthereforeinterpretationoftheresults
waslimited.
Confidenceintheresultsshownforthemajorityoftheoutcomeswas
lowtoverylowbecauseofstudylimitations(unclearblinding,unclear
randomisationandlackofallocationconcealment)andimprecision
aroundtheeffectestimate.
Thegroupagreeditwasnotclearwhethertherewouldbeapersistent
differencebetweenthegroupsat6monthsor1year.TheGDGagreed
thatitwasdifficulttodeterminethetreatmenteffectfromthesmall
studysizespresented.
Thegroupagreedthattherewasnoclearevidencetoshowthatstrength
trainingisbetterthanthecontrolinterventions(usualpractice)butboth
strengthtrainingandusualpracticeledtoimprovementssothe
consensusofthegroupwasthatthisstrengthtrainingisusefulforthose
withweaknessinupperorlowerlimbs,andthereforecouldbe
consideredaspartofapersonsrehabilitation.
Otherconsiderations
Definitionsofstrengthtrainingvaryfromtraditionalresistancetraining
tofunctionalstrengthtraining.Conventionalresistancetrainingwould
includeexercisessuchasliftingweightsinagym,whereasfunctional
strengthtrainingfocusesonbuildingstaminathrougharangeoftasks
suchaswalkingandgradedactivitiesdeliveredbyarehabilitation
professional.Thenatureofstrengthtrainingvariedaccordingtowhether
itwasupperorlowerlimbandtimesinceonsetofthestroke.
Therewasnoindicationfromthestudiespresentedofwhatdoseof
strengthtrainingisappropriate.Itwasalsohighlightedthatevidencefor
strengthtrainingwhichinvolvedafamilymembershowedshortterm
improvements.Yetthesewereshortlivedandthereforeseemnotto
makeacontributiontolongtermfunctionalgains.
13.2 FitnessTraining
13.2.1
Inpeopleafterstroke,doescardiorespiratoryorresistancefitnesstrainingimprove
outcome(fitness,function,qualityoflife,mood)andreducedisability?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention
Anycardiorespiratoryorresistancefitnesstrainingsuchas:
Aquaticphysicalexercise
Cycle,rowingortreadmillergometry
Weightbearingresistancetraining
Dynamicandisokineticmusclestrengthtraining
Comparison
Usualcare(otherphysiotherapy)
Outcomes
Mortalityrate
Dependenceorlevelofdisability
Physicalfitness
NationalClinicalGuidelineCentre,2013.
338
StrokeRehabilitation
Movement
ClinicalMethodologicalIntroduction
Mobility
Physicalfunction
Qualityoflife
Mood
Indicesandscalesmayinclude:
Bloodpressure
Bodymass
Maximaloxygenupdate(peakVO2(ml/kg/min)
Endurance
Barthel
Rivermeadmobilityindex
SF36
EuroQol
HADS
BeckDepressionIndex
Geriatricdepressionscale
Epidemiologicstudiesfordepressionscale(CESD)
13.2.1.1
ClinicalEvidenceReview
Searcheswereconductedforsystematicreviewscomparingtheclinicaleffectivenessoffitness
training(cardiorespiratoryorresistance)withusualcaretoimprovefunctionandreducedisabilityfor
adultsandyoungpeople16orolderwhohavehadastroke.
OneCochranesystematicreview(Brazzelli201131)wasidentified.ThisCochranereviewwasadapted
toaddressthecurrentprotocol(thecomparisonofmixedcardiorespiratoryvs.usualcarewas
removedandoutcomesthathadalreadybeenincludedinthereviewin12.1wereremovedfromthe
resistancevs.usualcarecomparison).TheCochranereviewincluded32trials.Fromthesetrials(32),
21trialsmatchingourprotocolwereincludedforthisreview.
AfurthersystematicsearchwasconductedforanytrialpublishedsincetheCochranesearchcutoff
(March2010)andfourtrials(Globas201294,Holmgren2010113,Jin2012125andVanDePort2012267)
wasidentified.
Inthesystematicreviewthefollowingstrategyofanalysiswasadopted:
Theeffectsoftheinterventionswereseparatelyanalysedattheendoftheinterventionandat
theendoffollowup.Endofinterventionreferstothetimepointwhenatrainingprogramme
finishes(rangedfrom214weeks)andendoffollowupreferstoanytimepointoccurringafter
theendofintervention(rangedfrom1236weeks).(SeeindividualGRADETable92/Table93
forcardiorespiratoryandTable94/Table95forresistancetraining).Retainedtrainingeffectswere
measuredattheendoffollowup.
Studieswereincludedinwhichcontrolswereexposedtoeitherphysicalactivityoccurringduring
usualcareornotrainingafterusualcare.Notrainingreferstonointerventionoranonexercise
intervention.TheseweresubgroupanalyseswithineachGRADEtable
Cardiorespiratorytrainingwasalsocomparedwithresistancetrainingusingonemobilityoutcome
(seeGRADETable96)
Whenthereisanoutcomewithsubgroup,overalleffectsaswellassubgroupanalyses(initalics)
arepresented(seeGRADEtables)
Theevidencestatementsalsoreflectthetotaleffectsaswellasthesubgroupanalysis.
NationalClinicalGuidelineCentre,2013.
339
StrokeRehabilitation
Movement
PleaseseeAppendixMforexcludedtrialsfromtheCochranereview.
Table90summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
Table90: OverviewofstudiesincludedintheCochranereview
COMPARISON
Cardiorespirat
oryvs.usual
care
STUDIES
3
Aidar2007 ;Bateman
200119;Cuviello
Palmer198852;da
Cunha200252;Eich
200474;Glasser1986
93
;KatzLeurer2003
134
;Lennon2008152;
Moore2010177;
Mudge2009180;Pohl
2002208;Potempa
1995211;Salbach2004
229
;Smith2008244
TOTALNUMBEROF
PARTICIPANTS
RANGEOF
INTERVENTIONS
718participants
Twoofthese
trialsassessed
circuittraining
(Mudge2009;
Salbach2004),
onetrialassessed
aquatictraining
(Aidar2007),
whilethe
remainingtrials
employed
differentformsof
ergometry(cycle,
treadmillor
Kinetron)
NationalClinicalGuidelineCentre,2013.
340
OUTCOMES
Disability
Functional
Independence
Measure(FIM)
Rivermead
MobilityIndex
PhysicalActivity
andDisability
Scale
Nottingham
ExtendedADL
Frenchay
ActivitiesIndex
Mobility
Functional
Ambulation
Categories
Maximalgait
speed(m/min)
Preferredgait
speed(m/min)
6MinuteWalk
Test(metres)
Gaitendurance
6metrewalking
time(sec)
StrokeImpact
Scale(mobility
domain)
Peakactivity
index(steps/min)
Maximumstep
ratein1min
Riskfactors
Bloodpressure
(systolicand
Diastolic)
HealthRelatedQoL
SF36
EuroQoL
Mood
StrokeRehabilitation
Movement
COMPARISON
STUDIES
TOTALNUMBEROF
PARTICIPANTS
RANGEOF
INTERVENTIONS
OUTCOMES
BeckDepression
Index
HospitalAnxiety
andDepression
Scale
Geriatric
DepressionScale
Centrefor
Epidemiologic
Studiesfor
DepressionScale
Casefatality
Resistancevs.
usualcare
192participants
Bale200814;Cooke
47
2010* ;Flansbjer
200884;Kim2001136;
Ouellette2004195;
Sims2009238;Winstein
2004285
Allemployed
muscle
contractions
resistedby
weights,exercise
machines,or
elasticdevices.
Fivetrialslimited
thestrength
trainingtothe
lowerlimbs,one
trialtotheupper
limbs(Winstein
2004),andone
trialtrainedboth
theupperand
lowerlimbs(Sims
2009).
Physicalfitness
PeakVO2
(ml/kg/min)
Gaiteconomy,
VO2
(ml/kg/metre)
Maximumcycling
workrate(Watts)
Bodymass(Kg)
Composite
measureof
musclestrength
Knee
flexion/knee
extension(Nm)
Physicalfunction
BergBalance
Scale
TimedUpandGo
(sec)
Weightbearing
(affectedside)
Stairclimbing,
maximal
(sec/step)
Cardiorespirat
oryvs.
resistance
CuvielloPalmer1988
52
;KatzLeurer2003
134
;Moore2010177;
Salbach2004229;Bale
200814;Kim2001136;
Ouellette2004195
Walkingspeed
Preferredgait
speed(m/min)
301participants
*Cooke2010wasincludedinthemixedcardiovascularresistancegroupwhichisnotinourprotocol.However,thisstudy
includedonlyresistancetrainingasaninterventionandthereforewasmovedtothiscomparisoninoureditedversionofthe
Cochranereview.
NationalClinicalGuidelineCentre,2013.
341
StrokeRehabilitation
Movement
Table91.DetailsoffouradditionalRCTsthatwerecompletedsincetheCochranereviewandwere
addedtothecurrentreview.SeeAppendixHforextraction
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Holmgren
2010113
34patientswith
stroke.
36monthsafter
strokeonset
15allocatedto
interventionand19
allocatedtocontrol.
Sevensessionsaweek
dividedover3days
withindividualized
grouptraining,
supervisedbya
physiotherapist(PT)
plusonesessiona
weekfor1hrwith
educationalgroup
discussionsaboutfall
riskandsecurity
aspectsledbyaPTand
anOT
(N=15)
Onesessionaweek
for1hreachduring
the5weekperiod.
Sessionwasan
educationalgroup
discussionsession
ledbyone
occupational
therapist(OT)
(N=19)
Globas
201294
38patientswith
stroke(>6months)
aged>60yearswith
residualhemiparetic
gait(atleast1
clinicalsignfor
paresis,spasticityor
circumductionof
affectedlegwhile
walking);abilityto
walkontreadmillat
0.3km/hr.for3
minuteswith
handrailsupport.
Highintensityaerobic
treadmillexercise
(TAEX)for3months
(39sessions)starting
with1020minutesat
4050%heartrate
reserve(HRR)building
upto3050minutesat
6080%HRR
ConventionalCare
Physiotherapy(13
sessionsof1hour
each/week)
includingpassive
muscletone
regulatingexercises
forupperandlower
extremity,balance
training
Bodymass
adjustedpeak
VO2;
Sustained
walkingability(6
minutewalk).
2ry:
10mtimedwalk
atcomfortable
andmaximal
speeds;
5chairrisetest;
Bergbalance
scale;
Rivermead
Index;
SF12
Jin2012125
133participantsage
50orolder;single
stroke>6months
ago;independentin
ambulationwithor
withoutwalkingaid
Cyclingexercisegroup:
8weekaerobiccycling
training+paretic
lowerlimbweights40
mins./day5timesa
week,targetaerobic
intensity5070%heart
ratereserve
Lowintensityover
groundwalking
training2030%
heartratereserve.
Bothgroupshad
balanceexercise30
minutesand
supervised
stretching20
minutes
6minutewalking
distance,
Rivermead
MobilityIndex.
Kneemuscle
strength
(dynamometer);
balance(Berg
scale);
Spasticity
(Modified
AshworthScale)
VanDePort
2012267
250participants
withverifiedstroke
whohadcompleted
inpatient
rehabilitation
(dischargedhome)
assoonasthey
wereabletostart
Taskorientedcircuit
training:90minute
gradedtaskoriented
circuittrainingtwicea
weekfor12weeks
aimedatimproving
walkingcompetency
(warmup5minutes;
Usualphysiotherapy
carefor12weeks
accordingto
guidelines;no
restrictionson
content,timeor
duration
StrokeImpact
ScaleMobility
domainsSIS;
RMI,
NEADL,
HADS,
fatigueseverity,
NationalClinicalGuidelineCentre,2013.
342
SF36
Geriatric
Depression
Scale15
StrokeRehabilitation
Movement
STUDY
POPULATION
outpatient
rehabilitation;able
towalkaminimum
of10mwithout
physicalassistance;
needtocontinue
physiotherapyto
improvewalking
competencyor
physicalconditionor
both;abletogive
informedconsent;
motivatedto
participatein12
weekintensive
physiotherapy
programme.
INTERVENTION
circuittraining60
minutes;
evaluation/break10
minutes;groupgame
15minutes)
NationalClinicalGuidelineCentre,2013.
343
COMPARISON
OUTCOMES
MotricityIndex,
functional
ambulation,
6minwalk,
5mcomfortable
walk,
timedbalance,
timedupandgo,
StrokeRehabilitation
Movement
Comparison:Cardiorespiratorytrainingversususualcare
Table92:CardiorespiratorytrainingendofinterventionversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings.
Subgroupsareinitalics.
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
Seesubgroupsfor
means
Seesub
groupsfor
means
0.21(
0.10,
0.52)
SMD0.21 Low
higher
(0.1
lowerto
0.52
higher)
100.38
(18.92)
0.23(
0.32,
0.78)
SMD0.23 Moderate
higher
(0.32
lowerto
0.78
higher)
DisabilityFunctionalIndependenceMeasure(Betterindicatedbyhighervalues)
3
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
DisabilityFunctionalIndependenceMeasureDuringusualcare(Betterindicatedbyhighervalues)
1
Bateman
200119
RCTsingle
blind
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
104.74(17.7)
DisabilityFunctionalIndependenceMeasureAfterusualcare(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
344
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
2
Cuviello
Palmer
198852;
Katz
Leurer
2003134
RCTsingle
blind
Summaryoffindings
Limitations
Serious
limitations(
a)
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Serious
imprecision(b
)
CuvielloPalmer:
44.79(8.77)
KatzLeurer:
105.8(12.5)
Cuviello
0.17(
Palmer:
0.29,
47.18(9.88) 0.63)
KatzLeurer:
101.4(16)
SMD0.17 Low
higher
(0.29
lowerto
0.63
higher)
Serious
imprecision(c
)
Seesubgroups
below
See
subgroups
below
MD0.57
higher
(0.03
lowerto
1.17
higher)
Bateman10.06
(3.6)
Jin10.5(1.7)
VandePort13.47
(1.44)
Bateman9.9 0.41
(3.65)
(0.01,
Jin10.4(1.6) 0.81)
VandePort
12.82(1.44)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
Confidence
(ineffect)
DisabilityRivermeadMobilityIndex(Betterindicatedbyhighervalues)
4
Seesub
groups
below
(next4
rows)
RCTssingle
blind
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
0.57(
0.03,
1.17)
Moderate
DisabilityRivermeadMobilityIndexDuringusualcare(Betterindicatedbyhighervalues)
3
Bateman
200119
Jin
2012125
VanDe
Port267
RCTssingle
blind
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
NationalClinicalGuidelineCentre,2013.
345
MD0.41 Moderate
higher
(0.01
higherto
0.81
higher)
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
13.3(1.7)
11.3(2.7)
2(0.53,
3.47)
MD2
Low
higher
(0.53
higherto
3.47
higher)
DisabilityRivermeadMobilityIndexAfterusualcare(Betterindicatedbyhighervalues)
1
Globas
201294
RCTssingle
blind
Serious
limitations(
d)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
DisabilityPhysicalActivityandDisabilityScaleAfterusualcare(Betterindicatedbylowervalues)
1
Mudge
2009180
RCTsingle
blind
Serious
limitations(
d)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
77.8(55.7)
60.9(67.2)
16.9(
15.15,
48.95)
MD16.9
higher
(15.15
lowerto
48.95
higher)
Low
Noserious
imprecision
Seesubgroupsfor
means
Seesub
groupsfor
means
0.4(
8.38,
9.18)
MD0.4
higher
(8.38
lowerto
9.18
higher)
Low
Riskfactorsbloodpressure,systolic(Betterindicatedbylowervalues)
4
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitations(
a)
Serious
inconsistency(
e)
Noserious
indirectness
Riskfactorsbloodpressure,systolicDuringusualcare(Betterindicatedbylowervalues)
NationalClinicalGuidelineCentre,2013.
346
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
1
daCunha
200253
RCTsingle
blind
Summaryoffindings
Limitations
Serious
limitations(
a)
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
daCunha:191.33
(9.93)
DaCunha:
165(28.81)
26.33
(1.95to
50.71)
MD26.33 Low
higher
(1.95to
50.71
higher)
Noserious
imprecision
KatzLeurer:130.3
(15.7)
Lennon:136(13.3)
Potempa:127.3
(18.31)
KatzLeurer: 2.69(
136.2(19.5) 8.03,
2.66)
Lennon:
133.5(16.7)
Potempa:
131.5
(22.54)
MD2.69
lower
(8.03
lowerto
2.66
higher)
Moderate
Noserious
imprecision
Seesubgroupsfor
means
Seesub
groupsfor
means
MD0.33
lower
(2.97
lowerto
2.31
higher)
Moderate
Serious
imprecision(c
)
Confidence
(ineffect)
Riskfactorsbloodpressure,systolicAfterusualcare(Betterindicatedbylowervalues)
3
Katz
Leurer
2003134;
Lennon
2008152;
Potempa
1995211
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Riskfactorsbloodpressure,diastolic(Betterindicatedbylowervalues)
4
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Riskfactorsbloodpressure,diastolicDuringusualcare(Betterindicatedbylowervalues)
NationalClinicalGuidelineCentre,2013.
347
0.33(
2.97,
2.31)
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
1
daCunha
200253
RCTsingle
blind
Summaryoffindings
Limitations
Serious
limitations(
a)
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Veryserious 95.33(9.69)
imprecision(j)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
94.33
(10.54)
1(
10.46,
12.46)
MD1
higher
(10.46
lowerto
12.46
higher)
Verylow
Riskfactorsbloodpressure,diastolicAfterusualcare(Betterindicatedbylowervalues)
3
Katz
Leurer
2003134;
Lennon
2008152;
Potempa
1995211
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
KatzLeurer:79
(9.7)
Lennon:81.4(8.4)
Potempa:78.4
(9.15)
KatzLeurer:
80.8(10.2)
Lennon:82
(9)
Potempa:
76.4(7.67)
0.41(
3.12,
2.31)
MD0.41
lower
(3.12
lowerto
2.31
higher)
Moderate
Serious
imprecision(c
)
Seesubgroupsfor
means
Seesub
groupsfor
means
2.73
(1.29to
4.17)
MD2.14
higher
(0.5to
3.78
higher)
Low
PhysicalfitnesspeakVO2(ml/kg/min)(Betterindicatedbyhighervalues)
6
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
348
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
DaCunha11.55
(2.76)
Jin16.8(1)
DaCunha
8.12(2.3)
Jin13.3(1)
3.50
(3.16,
3.84)
MD3.50
higher
(3.16to
3.84
higher)
Moderate
Lennon:12(2.2)
Moore:18(5.4)
Potempa:18.8
(4.79)
Globas24.4(6.6)
Lennon:
11.1(1.9)
Moore:16
(7.1)
Potempa:
15.2(4.32)
Globas20.9
(7.8)
1.85
(0.31,
3.39)
MD1.85
higher
(0.31to
3.39
higher)
Low
0.371
(0.234)
0.08(
0.28,
0.12)
MD0.08
lower
(0.28
lowerto
Verylow
PhysicalfitnesspeakVO2(ml/kg/min)Duringusualcare(Betterindicatedbyhighervalues)
2
daCunha
200253;
Jin
2012125
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
PhysicalfitnesspeakVO2(ml/kg/min)Afterusualcare(Betterindicatedbyhighervalues)
4
Lennon
2008152;
Moore
2010177;
Potempa
1995
211
;
Globas
201294
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
Physicalfitnessgaiteconomy,VO2(ml/kg/metre)Afterusualcare(Betterindicatedbyhighervalues)
1
Moore
2010177
RCTsingle
blind
Serious
limitations(f
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(g
)
0.291(0.228)
NationalClinicalGuidelineCentre,2013.
349
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
0.12
higher)
Seesubgroupsfor
means
Seesub
groupsfor
means
0.6
(0.18,
1.02
SMD0.6
higher
(0.18to
1.02
higher)
Bateman:
4.13(0.59)
DaCunha:
41.67
(12.91)
0.32(
0.34,
0.98)
SMD0.32 Moderate
higher
(0.34
lowerto
0.98
higher)
KatzLeurer:
12.9(12.6)
Potempa:
0.83
(0.47to
1.18)
SMD0.83 Low
higher
(0.47to
1.18
Physicalfitnessmaximumcyclingworkrate(Watts)(Betterindicatedbyhighervalues)
4
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
Low
Physicalfitnessmaximumcyclingworkrate(Watts)Duringusualcare(Betterindicatedbyhighervalues)
2
Bateman
200119;
daCunha
200253
RCTsingle
blind
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
Bateman:4.22
(0.72)
daCunha:62.5
(26.22)
Physicalfitnessmaximumcyclingworkrate(Watts)Afterusualcare(Betterindicatedbyhighervalues)
2
Katz
Leurer
RCTsingle
blind
Serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
KatzLeurer:25.2
(14.9)
Potempa:94.2
NationalClinicalGuidelineCentre,2013.
350
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
2003134;
Potempa
1995211
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
(46.64)
66.1(30.69)
80.79(15.78)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
higher)
75.41
(14.58)
5.38(
1.69,
12.45)
MD5.38
higher
(1.69
lowerto
12.45
higher)
Moderate
DaCunha:
1.86(1.77)
Pohl:4.3
(0.7)
Pohl*:4.3
(0.7)Vande
Port4.74
(0.55)
0.33[0.0
1,0.65]
MD0.33
higher
(0.01to
0.65
higher)
Low
Seesub
8.66
MD8.66
Moderate
PhysicalfitnessDuringusualcareBodyMass(Kg)(Betterindicatedbylowervalues)
1
Bateman
200119
RCTsingle
blind
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
MobilityfunctionalambulationcategoriesDuringusualcare(Betterindicatedbyhighervalues)
3
RCTsingle
daCunha blind
200253;
Pohl2002
208
;Pohl
2002
208
*Van
dePort267
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
daCunha:2.33
(1.37)
Pohl:5(0.01)
Pohl*:4.6(0.6)
VandePort4.87
(0.36)
Mobilitymaximalgaitspeed(m/minover5to10metres)(Betterindicatedbyhighervalues)
7
RCTsingle
Noserious
Noserious
Noserious
Serious
Seesubgroupsfor
NationalClinicalGuidelineCentre,2013.
351
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Seesub
groups
below
(next4
rows)
blind
Summaryoffindings
Limitations
limitations
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision(f)
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
means
groupsfor
means
(2.98,
14.34)
daCunha:
16.2(13.8)
Pohl*:58.2
(38.4)
Bateman:
16.22
(19.49)
Eich:36
(13.2)
Pohl:58.2
(38.4)
10(
0.05,
20.05
MD10
higher
(0.05
lowerto
20.05
higher)
Moderate
Salbach:48
(29.4)
9.93
(3.38,
MD9.93
higher
Low
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
higher
(2.98to
14.34
higher)
Confidence
(ineffect)
Mobilitymaximalgaitspeed(m/minover5to10metres)Duringusualcare(Betterindicatedbyhighervalues)
4
RCTsingle
daCunha blind
200253;
Pohl2002
208
*;
Bateman
200119;
Eich2004
74
;Pohl
2002208
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
daCunha:35.5
imprecision(f) (17.4)
Pohl*:73.2(44.4)
Bateman:16
(11.06)
Eich:42.6(18)
Pohl:97.8(48)
Mobilitymaximalgaitspeed(m/minover5to10metres)Afterusualcare(Betterindicatedbyhighervalues)
3
Salbach
RCTsingle
blind
Serious
limitation(a
Noserious
inconsistency
Noserious
indirectness
Serious
Salbach:59.4
imprecision(f) (33.6)
NationalClinicalGuidelineCentre,2013.
352
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
2004229;
Moore
2010177;
Mudge
2009180
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
(3.38to
16.48
higher)
Confidence
(ineffect)
Mobilitypreferredgaitspeed(m/min)(Betterindicatedbyhighervalues)
4
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Seesubgroupsfor
means
Seesub
groupsfor
means
4.68
(1.4,
7.96)
MD4.68
higher
(1.4to
7.96
higher)
Moderate
12.07(6.41)
6.04(
MD6.04
0.92,13) higher
(0.92
lowerto
13
higher)
Low
KatzLeurer:
27(9.6)
29(0.57, MD4.29
8.01)
higher
Moderate
Mobilitypreferredgaitspeed(m/min)Duringusualcare(Betterindicatedbyhighervalues)
1
Cuviello
Palmer
198852
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
18.11(9.22)
imprecision(f)
Mobilitypreferredgaitspeed(m/min)Afterusualcare(Betterindicatedbyhighervalues)
3
Katz
RCTsingle
blind
Serious
limitation(a
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
KatzLeurer:30.6
(10.8)
NationalClinicalGuidelineCentre,2013.
353
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Leurer
2003134;
Moore
2010177;
Salbach
2004229
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Moore:37.8(18)
Salbach:46.8(24)
Moore:34.8
(13.8)
Salbach:
38.4(22.2)
Seesubgroupsfor
means
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
(0.57to
8.01
higher)
Seesub
groupsfor
means
39.39
(13.53,
65.25)
MD39.39 VeryLow
higher
(13.53to
65.25
higher)
Eich164.4
(69.3)
Jin213.5
(50.6)
VandePort
354(145)
30.25(
4.70,
65.21)
MD30.25 Verylow
higher
(4.70
lowerto
65.21hig
her)
Mobilitygaitendurance(6MWTmetres)(Betterindicatedbyhighervalues)
7
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitation(a
)
Serious
inconsistency()
Noserious
indirectness
Serious
imprecision(n
)
Mobilitygaitendurance(6MWTmetres)Duringusualcare(Betterindicatedbyhighervalues)
3
RCTsingle
Eich2004 blind
74
;Jin
2012125
vande
Port
2012267
Serious
limitation
Veryserious
inconsistency(
h)
Noserious
indirectness
Serious
imprecision(n
)
Eich198.8(81.1)
Jin218.5(63.7)
VandePort412
(117)
NationalClinicalGuidelineCentre,2013.
354
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
Moore:201
(134)
Mudge:200
(99)
Salbach:209
(132)
Globas
265.9(189)
58.10
(23.02,
93.17)
MD58.10 Low
higher
(23.02to
93.17
higher)
0.87(0.62)
0.15(
0.19,
0.49)
MD0.15
lower
(0.19
lowerto
0.49
higher)
Low
See
subgroups
below
0.20
(0.12,
0.28)
MD0.20
higher
(0.12to
Low
Mobilitygaitendurance(6MWTmetres)Afterusualcare(Betterindicatedbyhighervalues)
4
Moore
2010177;
Mudge
2009180;
Salbach
2004229;
Globas
201294
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(n
)
Moore:226(130)
Mudge:282(117)
Salbach:249(136)
Globas332.1(138)
Mobilitymaximalgaitspeed(m/secover10metres)Afterusualcare(Betterindicatedbyhighervalues)
1
Globas
201294
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
1.02(0.38)
imprecision(f)
Mobilitycomfortablegaitspeed(m/secover5to10metres)(Betterindicatedbyhighervalues)
2
See
subgroup
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
Seesubgroups
imprecision(f) below
NationalClinicalGuidelineCentre,2013.
355
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
0.28
higher)
0.89(0.36)
0.21
(0.13,
0.29)
MD0.21
higher
(0.13to
0.29
higher)
Low
Serious
0.79(0.29)
imprecision(f)
0.70(0.46)
0.09(
0.16,
0.34)
MD0.09
higher
(0.16
lowerto
0.34
higher)
Low
Serious
imprecision(h
)
Seesub
groupsfor
means
8.87
(1.35,
16.4)
MD8.87
higher
(1.35to
16.4
higher)
Low
sbelow
Confidence
(ineffect)
Mobilitycomfortablegaitspeed(m/secover10metres)Duringusualcare(Betterindicatedbyhighervalues)
Vande
Port
2012267
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
1.1(0.3)
imprecision(f)
Mobilitycomfortablegaitspeed(m/secover10metres)Afterusualcare(Betterindicatedbyhighervalues)
1
Globas
201294
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Mobilitygaitendurance(m/min)(Betterindicatedbyhighervalues)
3
Seesub
groups
below
(next4
rows)
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Seesubgroupsfor
means
NationalClinicalGuidelineCentre,2013.
356
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
daCunha:34.17
(17.17)
Eich:33.13(13.52)
daCunha:
12.14
(10.87)
Eich:27.4
(11.55)
12.24(
3.41,
27.89)
MD12.24 Moderate
higher
(3.41
lowerto
27.89
higher)
41.4(22.8)
34.8(22.2)
6.6(
2.66,
15.86)
MD6.6
higher
(2.66
lowerto
15.86
higher)
Low
Serious
9.98(3.03)
imprecision(i)
13.3(7.82)
3.32(
8.52,
1.88)
MD3.32
lower
(8.52
lowerto
1.88
higher)
Low
Mobilitygaitendurance(m/min)Duringusualcare(Betterindicatedbyhighervalues)
2
RCTsingle
daCunha blind
200253;
Eich2004
74
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(h
)
Mobilitygaitendurance(m/min)Afterusualcare(Betterindicatedbyhighervalues)
1
Salbach
2004229
RCTsingle
blind
Serious
Noserious
limitation(c) inconsistency
Noserious
indirectness
Serious
imprecision(h
)
Mobility6metrewalkingtime(sec)Duringusualcare(Betterindicatedbylowervalues)
1
Glasser
198693
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
MobilityStrokeImpactScale(mobilitydomain)(Betterindicatedbylowervalues)
NationalClinicalGuidelineCentre,2013.
357
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
2See
subgroup
sbelow
Summaryoffindings
Limitations
Noserious
limitation
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Serious
imprecision(c
)
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
Seesubgroups
below
See
subgroups
below
3.20
(0.04,
6.35
MD3.20
(0.04to
6.35
higher)
Moderate
83.73
(13.25)
3.54
(0.30,
6.78)
MD3.54
higher
(0.30to
6.78
higher)
Moderate
68(15.4)
3.2(
17.14,
10.74)
MD3.2
lower
(17.14
lowerto
10.74
higher)
Verylow
49(17.5)
18.1
(7.71,
28.49)
MD18.1
higher
(7.71to
28.49
Low
MobilityStrokeImpactScale(mobilitydomain)Duringusualcare(Betterindicatedbylowervalues)
1
RCTsingle
blind
Vande
Port2012
267
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
87.27(12.38)
MobilityStrokeImpactScale(mobilitydomain)Afterusualcare(Betterindicatedbylowervalues)
1
Smith
2008244
RCTsingle
blind
Very
serious
limitations(
a)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(g
)
64.8(16.4)
Mobilitypeakactivityindex(steps/min)Afterusualcare(Betterindicatedbyhighervalues)
1
Mudge
2009180
RCTsingle
blind
Serious
limitation(d
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
67.1(22.8)
NationalClinicalGuidelineCentre,2013.
358
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
higher)
Serious
imprecision(c
)
90.7(21.9)
75.2(20.5)
15.5
(4.58,
26.42)
MD15.5
higher
(4.58to
26.42
higher)
Low
Serious
imprecision(c
)
Seesubgroupfor
means
Seesub
groupfor
means
1.28(
1.74,
3.30)
MD1.28
higher
(1.74
lowerto
3.30
higher)
Low
Bateman45(11.9)
Jin48.6(2.9)
Bateman
0.27(
45.3(11.3)
0.87,
Jin48.3(3.9) 1.41)
MD0.27
higher
(0.87
lowerto
1.41
higher)
Moderate
Mobilitymaxstepratein1minAfterusualcare(Betterindicatedbyhighervalues)
1
Mudge
2009180
RCTsingle
blind
Serious
limitation(d
)
Noserious
inconsistency
Noserious
indirectness
PhysicalfunctionBergBalancescale(Betterindicatedbyhighervalues)
5
Seesub
group
below
(next4
rows)
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
PhysicalfunctionBergBalancescaleDuringusualcare(Betterindicatedbyhighervalues)
2
Bateman
200119;
Jin
2012125
RCTsingle
blind
Serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
PhysicalfunctionBergBalancescaleAfterusualcare(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
359
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
3
Moore
2010177;
Salbach
2004229;
Globas
201294
RCTsingle
blind
Summaryoffindings
Limitations
Serious
limitation(a
)
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
Serious
imprecision(c
)
Moore:48(10)
Salbach:44(11)
Globas51.1(6.4)
Moore:46
(10)
Salbach:41
(13)Globas
44.3(11.9)
4.06
(0.52,
7.60)
MD4.06
higher
(o.52to
7.60
higher)
Low
Noserious
imprecision
Seesubgroups
below
See
subgroups
below
3.99(
6.91,
1.08)
MD3.99
lower
(6.91to
1.08
lower)
High
15(16)
4.00(
7.15,
0.85)
MD4.00
lower
(7.15to
0.85
lower)
High
Moore:24
(16)
3.94(
11.65,
MD3.94
lower
Low
PhysicalfunctionTimedUpandGo(sec)(Betterindicatedbylowervalues)
3
See
subgroup
sbelow
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
PhysicalfunctionTimedUpandGo(sec)Duringusualcare(Betterindicatedbylowervalues)
1
Vande
Port
2012267
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
11(7)
PhysicalfunctionTimedUpandGo(sec)Afterusualcare(Betterindicatedbylowervalues)
2
Moore
RCTsingle
blind
Serious
limitation(a
Noserious
inconsistency
Noserious
indirectness
Serious
Moore:20(12)
imprecision(j) Salbach:23.2
NationalClinicalGuidelineCentre,2013.
360
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
2010177;
Salbach
2004229
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
(20.6)
Salbach:
27.1(27.1)
3.77)
Aidar59.3
(6.9)
Globas43.7
(8.3)
Holmgren
45.8.(26.6)
0.82(
0.13,
1.77)
58.2(8.3)
11(6.15, MD11
15.85)
higher
(6.15to
15.85
higher)
Low
86(32)
14.6(
39.54,
10.34)
Low
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
(11.65
lowerto
3.77
higher)
Confidence
(ineffect)
HealthrelatedQoLSF36orSF12physicalfunctioningAfterusualcare(Betterindicatedbyhighervalues)
3
RCTsingle
blind
Aidar
3
2007 ;
Globas
201294;
Holmgren
2010113
Serious
limitation(a
)
Veryserious
inconsistency(k
)
Noserious
indirectness
Serious
imprecision(c
)
Aidar69.9(3.2)
Globas46.5(5)
Holmgren52.1
(22.2)
SMD0.82 Verylow
higher
(0.13
lowerto
1.77
higher)
HealthrelatedQoLSF36emotionalrolefunctioningAfterusualcare(Betterindicatedbyhighervalues)
1
Aidar
20073
RCTsingle
blind
Very
serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
69.2(3.5)
HealthrelatedQoLSF36emotionalrolefunctioningPostintervention(Betterindicatedbyhighervalues)
1
RCTsingle
Holmgren blind
2010113
Serious
limitation(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
71.4(38.9)
NationalClinicalGuidelineCentre,2013.
361
MD14.6
lower
(39.54
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
lowerto
10.34
higher)
79.4(20.8)
5.2(
6.19,
16.59)
MD5.2
higher
(6.19
lowerto
16.59
higher)
Low
33.2(12)
1(
8.74,
6.74)
MD1
lower
(8.74
lowerto
6.74
higher)
Verylow
54.8(10)
0.4(
7.42,
6.62)
MD0.4
lower
(7.42
lowerto
Verylow
HealthrelatedQoLSF36mentalhealthPostintervention(Betterindicatedbyhighervalues)
1
RCTsingle
Holmgren blind
2010113
Serious
limitation(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
86.4(12.4)
HealthrelatedQoLSF36PhysicalComponentScalePostintervention(Betterindicatedbyhighervalues)
1
RCTsingle
Holmgren blind
2010113
Serious
limitation(d
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(g
)
32.2(10.6)
HealthrelatedQoLSF36MentalComponentScalePostintervention(Betterindicatedbyhighervalues)
1
RCTsingle
Holmgren blind
2010113
Serious
limitation(d
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(g
)
54.4(10.3)
NationalClinicalGuidelineCentre,2013.
362
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
6.62
higher)
9.4(1.9)
8.8(3)
0.6(
1.6,2.8)
MD0.6
higher
(1.6
lowerto
2.8
higher)
Verylow
0.87(
2.52,
0.78)
MD0.78
lower
(2.58
lowerto
078
higher)
Moderate
MoodBeckDepressionIndexAfterusualcare(Betterindicatedbylowervalues)
1
Smith
2008244
RCTsingle
blind
Very
serious
limitation(a
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(g
)
MoodHospitalAnxietyandDepressionScale(HADS)anxietyscoreDuringusualcare(Betterindicatedbylowervalues)
2
Bateman
200119;
Vande
Port
2012267
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
Bateman4.42
(3.69)
VandePort3.8
(3.4)
6.36(3.47)
VandePort
4.01(3.6)
MoodHospitalAnxietyandDepressionScale(HADS)andGeriatricDepressionscoredepression(Betterindicatedbylowervalues)
1
Bateman
200119
Vande
Port
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
serious
imprecision
Bateman5.54
(3.26)
VandePort4.92
(3.62)
Bateman
6.94(3.82)
VandePort
4.42(3.69)
NationalClinicalGuidelineCentre,2013.
363
0.25(
0.77,
0.27)
SMD0.25 Moderate
lower
0.77
lowerto
0.27
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
intervention
Mean(SD)
Usualcare
Mean(SD)
2012267;
Holmgren
2010113
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
(l)
Effect
Mean
differen
ce(95%
CI)
Mean
Differenc
e(MD)
/Standar
dised
Mean
Differenc
e(SMD)
(95%CI)
higher)
Confidence
(ineffect)
Unclear allocation concealment and blinding (outcome assessor). Limitations were considered by study weights in the meta-analysis
Mean difference did not reach the agreed MID of 17 points for the motor scale between the intervention and control group
Confidence interval crosses default MID (0.5) for single studies or default 0.5*median control SD for 2 or more studies
Unclear allocation concealment
2
Heterogeneity: I = 59%
Unclear blinding (outcome assessor)
Confidence interval crosses both ends of default MID (0.5) for single studies or default 0.5*(median control SD) for 2 or more studies
Mean difference did not reach agreed MID of 0.16m/sec for the walking speed between the intervention and control group for acute stroke patients or 0.2 m/sec for chronic stroke
patients
Mean difference did not reach agreed MID of 28m for the 6 MWT between the intervention and control group
Mean difference did not reach the agreed MID of 10 secs. for the Time Up and Go between the intervention and control group
2
Heterogeneity: I = 74%
Pohl 2002*: Pohl 2002 data were subdivided into two relevant comparisons. Half of the controls (10 participants) were used for each comparison
NationalClinicalGuidelineCentre,2013.
364
StrokeRehabilitation
Movement
Comparison:Cardiorespiratorytrainingversususualcare
Table93:CardiorespiratorytrainingendofretentionfollowupversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
1/42
(2.40%)
2/39
(5.10%)
0.46(0.04,
4.92)
28fewer
per1000
(from49
fewerto
201
more)
Verylow
10.72(3.3)
10.97
(3.35)
0.25(
1.85,1.35)
MD0.25
lower
(1.85
lowerto
1.35
higher)
Moderate
31.59
(17.17)
2.64(
5.57,
10.85)
MD2.64
higher
(5.57
lowerto
10.85
higher)
Moderate
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
Confidence
(ineffect)
Casefatality
1
Katz
Leurer
2003
134
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Noserious
indirectness
Veryserious
imprecision(
b)
DisabilityRivermeadMobilityIndexDuringusualcare(Betterindicatedbyhighervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
DisabilityNottinghanExtendedADLsDuringusualcare(Betterindicatedbyhighervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
34.23(16.3)
NationalClinicalGuidelineCentre,2013.
365
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
Confidence
(ineffect)
DisabilityPhysicalActivityandDisabilityScaleAfterusualcare(Betterindicatedbylowervalues)
1
Mudge
2009180
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Noserious
indirectness
Serious
imprecision(c
)
82.1(72.8)
62.2(72.5)
19.9(
17.58,
57.38)
MD19.9
higher
(17.58
lowerto
57.38
higher)
Low
26(5)
1(1.55,
3.55)
MD1
higher
(1.55
lowerto
3.55
higher)
Low
84.06(75.52)
77.94
(44.76)
6.12(
24.06,
36.3)
MD6.12
higher
(24.06
lowerto
36.3
higher)
Moderate
80.39(15.83)
77.58
(14.43)
2.81(
4.63,
10.25)
MD2.81
higher
(4.63
Moderate
DisabilityFrenchayActivitiesIndex(FAI)Afterusualcare(Betterindicatedbyhighervalues)
1
Katz
Leurer
2003134
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Noserious
indirectness
Serious
imprecision(c
)
27(6.5)
Physicalfitnessmaximumcyclingworkrate(Watts)Duringusualcare(Betterindicatedbyhighervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
PhysicalfitnessBodyMass(Kg)Duringusualcare(Betterindicatedbylowervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
NationalClinicalGuidelineCentre,2013.
366
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
Seesubgroupsfor
means
Seesub
groupsfor
means
8.21(3.38,
13.05)
MD8.21
higher
(3.38to
13.05
higher)
Moderate
Bateman:21.04
(12.31)
Eich:46.2(21)
Bateman:
15(21.86)
Eich:34.8
(13.2)
8.1(1.98,
14.22)
MD8.1
higher
(1.98to
14.22
higher)
Moderate
46.2(15.6)
37.8(15)
8.4(0.52,
16.28)
MD8.4
higher
(0.52to
16.28
higher)
Low
Seesubgroupsfor
Seesub
69.3
MD69.3
High
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
lowerto
10.25
higher)
Confidence
(ineffect)
Mobilitymaximalgaitspeed(m/min)(Betterindicatedbyhighervalues)
3
Seesub
groups
below
(next4
rows)
RCTsingle Noserious
blind
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
d)
Mobilitymaximalgaitspeed(m/min)Duringusualcare(Betterindicatedbyhighervalues)
2
Bateman
200119;
Eich2004
74
RCTsingle Noserious
blind
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
d)
Mobilitymaximalgaitspeed(m/min)Afterusualcare(Betterindicatedbyhighervalues)
1
Mudge
2009180
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Noserious
indirectness
Serious
imprecision(
d)
Mobilitygaitendurance(6MWTmetres)(Betterindicatedbylowervalues)
2
RCTsingle Noserious
Noserious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
367
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Seesub
groups
below
(next4
rows)
blind
Summaryoffindings
Limitations
limitation
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
means
groupsfor
means
(33.38,
105.23)
163(70.2)
61.8
(16.48,
107.12)
MD61.8
higher
(16.48to
107.12
higher)
Moderate
195(104)
82(23.05,
140.95)
MD82
higher
(23.05to
140.95
higher)
Low
51.5(20.5)
12.2(1.38,
23.02)
MD12.2
higher
(1.38to
23.02
higher)
Low
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
higher
(33.38to
105.23
higher)
Confidence
(ineffect)
Mobilitygaitendurance(6MWTmetres)Duringusualcare(Betterindicatedbyhighervalues)
1
Eich2004
74
RCTsingle Noserious
blind
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(f
)
224.8(90)
Mobilitygaitendurance(6MWTmetres)Afterusualcare(Betterindicatedbyhighervalues)
1
Mudge
2009180
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Noserious
indirectness
Serious
imprecision(f
)
277(125)
Mobilitypeakactivityindex(steps/min)Afterusualcare(Betterindicatedbyhighervalues)
1
Mudge
2009180
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Noserious
indirectness
Serious
imprecision(c
)
63.7(21.5)
Mobilitymaxstepratein1minAfterusualcare(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
368
StrokeRehabilitation
Movement
Qualityassessment
Summaryoffindings
Noof
studies
Design
Limitations
Inconsistency
1
Mudge
2009180
RCTsingle Serious
Noserious
blind
limitation(a) inconsistency
Indirectness
Noserious
indirectness
Imprecision
Serious
imprecision(c
)
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
87.7(21)
75.6(22.2)
12.1(0.93,
23.27)
MD12.1
higher
(0.93to
23.27
higher)
Low
78.3(13.3)
72.4(18)
5.9(7.97,
19.77)
MD5.9
higher
(7.97
lowerto
19.77
higher)
Verylow
46.16(12.09)
49.09
(8.01)
2.93(
7.91,2.05)
MD2.93
lower
(7.91
lowerto
2.05
higher)
Moderate
7.3(2.5)
8.6(2.9)
1.3(3.67, MD1.3
1.07)
lower
(3.67
lowerto
1.07
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
Confidence
(ineffect)
MobilityStrokeImpactScale(mobilitydomain)Afterusualcare(Betterindicatedbylowervalues)
1
Smith
2008244
Noserious
indirectness
Veryserious
imprecision(
e)
PhysicalfunctionBergBalancescaleDuringusualcare(Betterindicatedbyhighervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
MoodBeckDepressionIndexAfterusualcare(Betterindicatedbylowervalues)
1
Smith
2008244
Noserious
indirectness
Serious
imprecision(c
)
NationalClinicalGuidelineCentre,2013.
369
Verylow
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
higher)
Confidence
(ineffect)
MoodHospitalAnxietyandDepressionScale(HADS)anxietyscoreDuringusualcare(Betterindicatedbylowervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
3.57(3.36)
5.17(3.99)
1.6(3.58, MD1.6
0.38)
lower
(3.58
lowerto
0.38
higher)
Moderate
2.7(4.4,
1)
Moderate
MoodHospitalAnxietyandDepressionScale(HADS)depressionscoreDuringusualcare(Betterindicatedbylowervalues)
1
Bateman
200119
RCTsingle Noserious
blind
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
)
3.3(2.36)
6(3.92)
MD2.7
lower
(4.4to1
lower)
HealthrelatedQoLSF36PhysicalComponentScale6monthspostintervention(Betterindicatedbyhighervalues)
1
RCTsingle Serious
Noserious
limitation(a) inconsistency
Holmgren blind
2010113
Noserious
indirectness
Veryserious
imprecision(
e)
35.3(13.3)
35.4(12.9)
0.1(9.47, MD0.1
9.27)
lower
(9.47
lowerto
9.27
higher)
Verylow
5(14.22,
4.22)
Low
HealthrelatedQoLSF36MentalComponentScale6monthspostintervention(Betterindicatedbyhighervalues)
1
RCTsingle Serious
Noserious
limitation(a) inconsistency
Holmgren blind
2010113
Noserious
indirectness
Serious
imprecision(c
)
50.4(15)
55.4(9.3)
NationalClinicalGuidelineCentre,2013.
370
MD5
lower
(14.22
lowerto
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
4.22
higher)
Confidence
(ineffect)
HealthrelatedQoLSF36physicalfunctioning6monthspostintervention(Betterindicatedbylowervalues)
1
RCTsingle Serious
Noserious
blind
limitation(a)
inconsistency
Holmgren
113
2010
Noserious
indirectness
Veryserious
imprecision(
e)
51.5(18.6)
46.7(26.9)
4.8(
11.22,
20.82)
MD4.8
higher
(11.22
lowerto
20.82
higher)
Verylow
90.7(27.6)
18.9(
44.34,
6.54)
MD18.9
lower
(44.34
lowerto
6.54
higher)
Low
77.3(21.2)
3.9(7.84,
15.64)
MD3.9
higher
(7.84
lowerto
15.64
higher)
Verylow
3.7(2.9)
0.7(2.27, MD0.7
HealthrelatedQoLSF36emotionalrolefunctioning6monthspostintervention(Betterindicatedbylowervalues)
1
RCTsingle Serious
Noserious
blind
limitation(a)
inconsistency
Holmgren
113
2010
Noserious
indirectness
Serious
imprecision(c
)
71.8(40.5)
HealthrelatedQoLSF36mentalhealth6monthspostintervention(Betterindicatedbylowervalues)
1
RCTsingle Serious
Noserious
limitation(a) inconsistency
Holmgren blind
2010
Noserious
indirectness
Veryserious
imprecision(c
)
81.2(11.9)
MoodGeriatricDepressionScale156monthspostintervention(Betterindicatedbylowervalues)
1
RCTsingle Serious
Noserious
Noserious
Serious
3(1.5)
NationalClinicalGuidelineCentre,2013.
371
Low
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Holmgren blind
2010
(a)
(b)
(c)
(d)
(e)
(f)
Summaryoffindings
Limitations
Inconsistency
limitation(a) inconsistency
Indirectness
indirectness
Imprecision
Cardiorespiratory
trainingendof
retentionfollow
up
Mean(SD)/
Frequency
Usualcare Effect
Mean(SD)/ Relative
Frequency risk/mean
difference
(95%CI)
imprecision(c
)
0.87)
Unclear allocation concealment/unclear assessor blinding. Limitations were considered by study weights in the meta-analysis
Confidence interval crosses one end of default MID (0.75, 1.25)
Confidence interval crosses one end of default MID (0.5)
Mean difference did not reach agreed MID of 0.16m/sec for the walking speed between the intervention and control group
Confidence interval crosses both ends of default MID (0.5)
Mean difference did not reach agreed MID of 28m for the 6 MWT between the intervention and control group
NationalClinicalGuidelineCentre,2013.
372
Absolute
effect/
Mean
differenc
e(MD)
(95%CI)
lower
(2.27
lowerto
0.87
higher)
Confidence
(ineffect)
StrokeRehabilitation
Movement
Comparison:Resistancetrainingversususualcare
Table94:ResistancetrainingendofinterventionversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
Seesubgroup
formeans
Seesubgroup
formeans
0.58
(0.06,
1.1)
SMD0.58
higher
(0.06to
1.1
higher)
Low
220.58
(260.26)
0.47(
0.16,
1.1)
SMD0.47
higher
(0.16
lowerto
1.1
higher)
Low
142(193)
0.84(
0.09,
1.76)
SMD0.84
higher
(0.09
lowerto
1.76
Low
Physicalfitnesscompositemeasureofmusclestrength(Betterindicatedbyhighervalues)
2
Seesub
group
below
(next4
rows)
RCTsingle
anddouble
blind
Serious
limitation(d)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
PhysicalfitnesscompositemeasureofmusclestrengthDuringandafterusualcare(Betterindicatedbyhighervalues)
1
Winstein
2004285
RCTsingle
blind
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
353.53
(296.25)
PhysicalfitnesscompositemeasureofmusclestrengthAfterusualcare(Betterindicatedbyhighervalues)
1
Kim2001
136
RCTdouble
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
507(559)
NationalClinicalGuidelineCentre,2013.
373
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
higher)
Seesubgroup
formeans
Seesubgroup
formeans
12.01(
4.46,
28.47)
MD12.01
higher
(4.46
lowerto
28.47
higher)
Verylow
12.9(13.5)
4.8(
5.98,
15.58)
MD4.8
higher
(5.98
lowerto
15.58
higher)
Verylow
41.3(20.9)
21.8
(4.92,
38.68)
MD21.8
higher
(4.92to
38.68
higher)
Low
Physicalfitnessmusclestrength,kneeextension(Nm)(Betterindicatedbyhighervalues)
2
Seesub
group
below
(next4
rows)
RCTsingle
blind
Veryserious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(e
)
Physicalfitnessmusclestrength,kneeextension(Nm)Duringusualcare(Betterindicatedbyhighervalues)
1
Bale
200814
RCTsingle
blind
Veryserious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(e
)
17.7(9.8)
Physicalfitnessmusclestrength,kneeextension(Nm)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(e
)
63.1(19.6)
Physicalfitnessmusclestrength,kneeflexion(Nm)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
374
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
2
Seesub
group
below
(next4
rows)
RCTsingle
blind
Summaryoffindings
Limitations
Veryserious
limitation(c)
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
Seesubgroup
formeans
9.61(
5.01,
24.24)
MD9.61
higher
(5.01
lowerto
24.24
higher)
Verylow
2.8(4.8)
4.5(
1.13,
10.13)
MD4.5
higher
(1.13
lowerto
10.13
higher)
Verylow
Serious
74(27.7)
imprecision(f)
53.5(21.1)
20.5
(0.84,
40.16)
MD20.5
higher
(0.84to
40.16
higher)
Low
Serious
imprecision(g
)
Seesubgroup
formeans
3.98(
4.88,
12.85)
MD3.98
higher
(4.88
Verylow
Serious
Seesubgroup
imprecision(f) formeans
Confidence
(ineffect)
Physicalfitnessmusclestrength,kneeflexion(Nm)Duringusualcare(Betterindicatedbyhighervalues)
1
Bale
200814
RCTsingle
blind
Veryserious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
7.3(6.9)
imprecision(f)
Physicalfitnessmusclestrength,kneeflexion(Nm)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Mobilitymaximalgaitspeed(m/min)(Betterindicatedbyhighervalues)
2
Seesub
group
RCTsingle
blind
Veryserious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Seesubgroup
formeans
NationalClinicalGuidelineCentre,2013.
375
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
lowerto
12.85
higher)
17.4(6)
9(6)
8.4
(2.82,
13.98)
MD8.4
higher
(2.82to
13.98
higher)
Verylow
3.98(7.89)
4.63(7.29)
0.65(
6.86,
5.56)
MD0.65
lower
(6.86
lowerto
5.56
higher)
Low
13.8(6)
4.8(6)
9(3.42,
14.58)
MD9
higher
(3.42to
14.58
higher)
Verylow
below
(next4
rows)
Confidence
(ineffect)
Mobilitymaximalgaitspeed(m/min)Duringusualcare(Betterindicatedbyhighervalues)
1
Bale
200814
RCTsingle
blind
Veryserious
limitation(c)
Veryserious
Noserious
inconsistency(j) indirectness
Serious
imprecision(g
)
Mobilitymaximalgaitspeed(m/min)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(g
)
Mobilitypreferredgaitspeed(m/min)Duringusualcare(Betterindicatedbyhighervalues)
1
Bale
200814
RCTsingle
blind
Veryserious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(g
)
MobilityRivermead(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
376
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
1
Cooke
201047
RCTsingle
blind
Summaryoffindings
Limitations
Noserious
limitation
Inconsistency
Noserious
inconsistency
Indirectness
Noserious
indirectness
Imprecision
Serious
imprecision(b
)
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
37.7(8.6)
34.6(10.8)
3.1(
1.58,
7.78)
MD3.1
higher
(1.58
lowerto
7.78
higher)
Moderate
11.8
(0.89,
22.71)
MD11.8
higher
(0.89to
22.71
higher)
Verylow
SMD0.04
lower
(0.86
lowerto
0.77
higher)
Verylow
MD1.2
lower
(11.84
Low
Physicalfunctionweightbearing(%bodyweightaffectedside)Duringusualcare(Betterindicatedbyhighervalues)
1
Bale
200814
RCTsingle
blind
Veryserious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
17.4(8.8)
5.6(14.6)
Physicalfunctionstairclimbing,maximal(sec/step)Afterusualcare(Betterindicatedbyhighervalues)
2
Kim2001
136
;
Ouellette
2004195
RCTsingle
anddouble
blind
Serious
limitation(c)
Serious
inconsistency(k
)
Noserious
indirectness
Veryserious
imprecision(h
)
Kim:0.03
Kim:0.08(0.1) 0.04(
(0.08)
Ouellette:0.53 0.86,
0.77)
Ouellette:0.65 (0.34)
(0.41)
PhysicalfunctionTimedUpandGo(sec)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
23.1(10.3)
imprecision(i)
24.3(14.2)
NationalClinicalGuidelineCentre,2013.
377
1.2(
11.84,
9.44)
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
lowerto
9.44
higher)
0.74(7.15)
0.73(5.81)
1.47(
4.24,
7.18)
MD1.47
higher
(4.24
lowerto
7.18
higher)
Verylow
Veryserious
imprecision(h
)
1.73(7.34)
1.07(10.13)
2.8(
4.95,
10.55)
MD2.8
higher
(4.95
lowerto
10.55
higher)
Verylow
Serious
imprecision(b
)
69.9(18.9)
60.8(19.6)
9.1(
0.14,
18.34)
MD9.1
higher
(0.14
lowerto
18.34
Moderate
HealthrelatedQoLSF36physicalfunctioningAfterusualcare(Betterindicatedbyhighervalues)
1
Kim2001
136
RCTdouble
blinded
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(h
)
HealthrelatedQoLSF36mentalhealthAfterusualcare(Betterindicatedbyhighervalues)
1
Kim2001
136
RCTdouble
blinded
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
EuroQolSelfperceivedhealth(Betterindicatedbyhighervalues)
1
Cooke
201047
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
378
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Resistance
trainingend
of
intervention
Mean(SD)
Usualcare
Mean(SD)
Effect
Confidence
(ineffect)
Meanor
Standard
ised
Mean
Differen
ce
(95%CI)
Mean
Differenc
e(MD)/
Standardi
sedMean
Differenc
e(SMD)
(95%CI)
higher)
5.49(
9.78,
1.2)
MD5.49
Verylow
lower
(9.78to
1.2lower)
MoodCentreforEpidemiologicStudiesforDepressionscale(CESD)Afterusualcare(Betterindicatedbylowervalues)
1
Sims
2009238
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
RCTsingle
blind
Veryserious
limitation(a);
(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
15.13(8.49)
20.62(11.79)
NationalClinicalGuidelineCentre,2013.
379
StrokeRehabilitation
Movement
Comparison:Resistancetrainingversususualcare
Table95:ResistancetrainingendofretentionfollowupversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Qualityassessment
Noof
studies
Design
Summaryoffindings
Limitations
Inconsistency
Indirectness
Imprecision
Resistance
training
endof
retention
followup
Mean(SD)
Control
Mean
(SD)
Effect
Mean
differenc
e
(95%CI)
Mean
Difference
(MD)(95%
CI)
Confidence
(ineffect)
Physicalfitnessmusclestrength,kneeextension(Nm)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
59.4(22.6)
imprecision(b)
42(20.1) 17.4(
0.01,
34.81)
MD17.4
higher
(0.01
lowerto
34.81
higher)
Low
53(22.1) 17.6(
2.17,
37.37)
MD17.6
higher
(2.17
lowerto
37.37
higher)
Low
Physicalfitnessmusclestrength,kneeflexion(Nm)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
70.6(26.7)
imprecision(b)
Mobilitymaximalgaitspeed(m/min)Afterusualcare(Betterindicatedbyhighervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
96.6(59.4)
116.4
(106.8)
19.8(
95.77,
56.17)
MD19.8
lower
(95.77
lowerto
56.17
higher)
Moderate
Noserious
Veryserious
39.9(7.2)
39.7
0.2(
MD0.2
Low
MobilityRivermead(Betterindicatedbylowervalues)
1
RCTsingle
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
380
StrokeRehabilitation
Movement
Qualityassessment
Noof
studies
Design
Cooke
201047
blind
Summaryoffindings
Limitations
limitation
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
Resistance
training
endof
retention
followup
Mean(SD)
imprecision(d)
Control
Mean
(SD)
Effect
Confidence
(ineffect)
Mean
differenc
e
(95%CI)
Mean
Difference
(MD)(95%
CI)
(5.7)
3.34,
3.74)
higher
(3.34
lowerto
3.74
higher)
26.7
(18.9)
3.1(
16.67,
10.47)
MD3.1
lower
(16.67
lowerto
10.47
higher)
Low
PhysicalfunctionTimedUpandGo(sec)Afterusualcare(Betterindicatedbylowervalues)
1
Flansbjer
200884
RCTsingle
blind
Serious
limitation(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
23.6(11.1)
MoodCentreforEpidemiologicStudiesforDepressionscale(CESD)Afterusualcare(Betterindicatedbylowervalues)
1
RCTsingle
Sims2009 blind
238
Veryserious
limitation(a);(
c)
Noserious
inconsistency
Noserious
indirectness
Serious
13.78(8.02)
imprecision(b)
22.7
(11.17)
8.92(
13.03,
4.81)
MD8.92
lower
(13.03to
4.81
lower)
Verylow
Noserious
indirectness
Serious
69.6(19.3)
imprecision(b)
66.2
(18.9)
3.4(
7.31,
14.11)
MD3.4
higher
(7.31
lowerto
14.11
higher)
Moderate
EuroQolSelfperceivedhealth(Betterindicatedbyhighervalues)
1
Cooke
201047
RCTsingle
blind
Noserious
limitation
Noserious
inconsistency
NationalClinicalGuidelineCentre,2013.
381
StrokeRehabilitation
Movement
(a)
(b)
(c)
(d)
Unclear blinding.
Confidence interval crosses one end of default MID (0.5)
Unclear allocation concealment
Confidence interval crosses both ends of default MID (0.5)
NationalClinicalGuidelineCentre,2013.
382
StrokeRehabilitation
Movement
Comparison:Cardiorespiratoryversusresistancetraining
Table96:CardiorespiratoryversusresistancetrainingClinicalstudycharacteristicsandclinicalsummaryoffindings
Qualityassessment
Noof
studies
Design
Noofpatients
Limitations
Inconsistency
Indirectness
Imprecision
Cardiorespiratory
versusresistance
training
Mean(SD)
Usual
care
Mean
(SD)
Effect
Mean
differenc
e(95%
CI)
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
Mobilitygaitpreferredspeed(m/min)Cardiorespiratorytraining(Betterindicatedbylowervalues)
4
Cuviello
Palmer
198852;
Katz
Leurer
2003134;
Moore
2010177;
Salbach
2004229
RCTsingle
blind
Serious
limitation(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(b
)
CuvielloPalmer:
18.11(9.22)
KatzLeurer:30.6
(10.8)
Moore:37.8(18)
Salbach:46.8(24)
Cuviello
4.68
Palmer:
(1.4,
12.07
7.96)
(6.41)
Katz
Leurer:27
(9.6)
Moore:
34.8
(13.8)
Salbach:
38.4
(22.2)
MD4.68
higher
(1.4to
7.96
higher)
Low
Bale:4.8 2.34(
(6)
6.77,
11.45)
Kim:5.4
(4.2)
Ouellette:
38.4
(24.75)
MD2.34
higher
(6.77
lowerto
11.45
higher)
Verylow
Mobilitygaitpreferredspeed(m/min)Resistancetraining(Betterindicatedbylowervalues)
3
RCTsingle
anddouble
Bale
blind
200814;
Kim2001
136
;
Ouellette
2004195
(a)
(b)
Serious
limitation(c)
Veryserious
inconsistency(d
)
Noserious
indirectness
Serious
imprecision(b
)
Bale:13.8(6)
Kim:2.4(7.8)
Ouellette:38.4(22)
Unclear blinding and unclear allocation concealment. Limitations were considered by study weights in the meta-analysis
Mean difference did not reach agreed MID of 0.16m/sec for the walking speed between the intervention and control group
NationalClinicalGuidelineCentre,2013.
383
StrokeRehabilitation
Movement
(c)
(d)
NationalClinicalGuidelineCentre,2013.
384
13.2.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationswereidentifiedoncardiorespiratoryorresistancefitnesstraining.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Cardiorespiratorytrainingisdeliveredaspartofusualrehabilitationprogrammesby
physiotherapists.Thecostperhourofacommunitybasedphysiotherapistis30.51TheGDG
acknowledgedthatadditionalcostswouldbeincurredifpeoplearereferredfortrainingprogrammes
postrehabilitation.
13.2.1.3
Evidencestatements
Clinicalevidencestatements
Endofintervention:
Functionalindependencemeasure
Threestudies1952134of162participantsfoundnosignificantdifferenceinFIMDisabilitybetween
theparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcareatthe
endofintervention(LOWCONFIDENCEINEFFECT)
Onestudy19of52participantsfoundnosignificantdifferenceinFIMDisabilitybetweenthe
participantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcareat
theendofintervention(MODERATECONFIDENCEINEFFECT)
Twostudies52134of110participantsfoundnosignificantdifferenceinFIMDisability
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceived
usualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
RivermeadMobilityIndex
Fourstudies19,125,26794of488participantsfoundthatcardiorespiratorytrainingwasassociatedwitha
significantimprovementinthelevelofdisabilityasmeasuredbyRivermeadMobilityIndex,
comparedtousualcareattheendofintervention(MODERATECONFIDENCEINEFFECT).Rivermead
MobilityIndexwasthensubdividedintogroupsofstudieswhereparticipantsinthecontrolgroup
stillreceivedusualcarerehabilitation(duringusualcare)andthosewhereparticipantswere
recruitedwhomaynotcurrentlyreceiveusualcarerehabilitation(afterusualcare).
Duringusualcare:Threestudies19,125,26794comprising452participantsfoundthat
cardiorespiratorytrainingwasassociatedwithasignificantimprovementinthelevelof
disability,asmeasuredbytheRivermeadMobilityIndex,comparedtousualcareattheend
ofintervention(MODERATECONFIDENCEINEFFECT).
Afterusualcare:Onestudy94,94of36participantsfoundthatcardiorespiratorytraining
significantlyimprovedlevelofdisabilityasassessedwiththeRivermeadMobilityIndex,
comparedtousualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
PhysicalActivityandDisabilityscale
NationalClinicalGuidelineCentre,2013.
385
Onestudy180of58participantsfoundnosignificantdifferenceinthePhysicalActivityandDisability
scalebetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusual
careattheendofintervention(LOWCONFIDENCEINEFFECT)
Systolicbloodpressure
Fourstudies53134152211of190participantsfoundnosignificantdifferenceinsystolicbloodpressure
Riskfactorsbetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Onestudy53of12participantsfoundthatusualcarewasassociatedwithstatistically
significantimprovementinsystolicbloodpressureRiskfactorscomparedtothe
cardiorespiratorytrainingattheendofintervention(LOWCONFIDENCEINEFFECT)
Threestudies134152211of178participantsfoundnosignificantdifferenceinsystolicblood
pressureRiskfactorsbetweentheparticipantswhoreceivedcardiorespiratorytrainingand
thosewhoreceivedusualcareattheendofintervention(MODERATECONFIDENCEIN
EFFECT)
Diastolicbloodpressure
Fourstudies53134152211of190participantsfoundnosignificantdifferenceindiastolicbloodpressure
Riskfactorsbetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(MODERATECONFIDENCEINEFFECT)
Onestudy53of12participantsfoundnosignificantdifferenceindiastolicbloodpressure
Riskfactorsbetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Threestudies134152211of178participantsfoundnosignificantdifferenceindiastolicblood
pressureRiskfactorsbetweentheparticipantswhoreceivedcardiorespiratorytrainingand
thosewhoreceivedusualcareattheendofintervention(MODERATECONFIDENCEIN
EFFECT)
PeakVO2
Sixstudies5315217794,125,211comprising289participantsfoundasignificantimprovementinpeak
oxygenuptake(VO2)Physicalfitnessinfavouroftheparticipantsthatreceivedcardiorespiratory
trainingcomparedtotheparticipantsthatreceivedusualcareattheendofintervention(LOW
CONFIDENCEINEFFECT)
Twostudies53,125of145participantsfoundasignificantdifferenceinpeakoxygenuptake
(VO2)Physicalfitnessinfavouroftheparticipantsthatreceivedcardiorespiratorytraining
comparedtotheparticipantsthatreceivedusualcareattheendofintervention
(MODERATECONFIDENCEINEFFECT)
Fourstudies15217794,211of144participantsfoundasignificantimprovementinpeakoxygen
uptake(VO2)Physicalfitnessinfavouroftheparticipantsthatreceivedcardiorespiratory
trainingcomparedtotheparticipantsthatreceivedusualcareattheendofintervention
(LOWCONFIDENCEINEFFECT)
Gaiteconomy,VO2
Onestudy177of20participantsfoundnosignificantdifferenceinGaiteconomy,VO2Physicalfitness
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Maximumcyclingworkrate
Fourstudies1953134211of221participantsfoundthatcardiorespiratorytrainingwasassociatedwitha
statisticallysignificantdifferenceinmaximumcyclingworkratePhysicalfitnesscomparedtousual
careattheendofintervention(LOWCONFIDENCEINEFFECT)
NationalClinicalGuidelineCentre,2013.
386
Twostudies1953of89participantsfoundnosignificantdifferenceinmaximumcyclingwork
ratePhysicalfitnessbetweentheparticipantswhoreceivedcardiorespiratorytrainingand
thosewhoreceivedusualcareattheendofintervention(MODERATECONFIDENCEIN
EFFECT)
Twostudies134211of132participantsfoundthatthecardiorespiratorytrainingwasassociated
withstatisticallysignificantimprovementinmaximumcyclingworkratePhysicalfitness
comparedtousualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Bodymass(KG)
Onestudy19of72participantsfoundnosignificantdifferenceinBodymass(KG)Physicalfitness
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofintervention(MODERATECONFIDENCEINEFFECT)
FunctionalAmbulationCategories
Twostudies53208of73participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementinFunctionalAmbulationCategoriesMobilitycomparedtothe
usualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Maximalgaitspeed
Sevenstudies5320819,74229177180of365participantsfoundthatcardiorespiratorytrainingwas
associatedwithstatisticallysignificantimprovementinmaximalgaitspeedMobilitycomparedto
theusualcareattheendofintervention,althoughthisdifferencewasnotofclinicalsignificance
(MODERATECONFIDENCEINEFFECT)
Fourstudies5320819,74of196participantsfoundnosignificantdifferenceinmaximalgaitspeed
Mobilitybetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(MODERATECONFIDENCEINEFFECT)
Threestudies229177180of169participantsfoundthatcardiorespiratorytrainingwas
associatedwithstatisticallysignificantimprovementinmaximalgaitspeedMobility
comparedtotheusualcareattheendofintervention.Thisdifferencewasnotclinically
significant(LOWCONFIDENCEINEFFECT)
Preferredgaitspeed
Fourstudies52134177229of221participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementinpreferredgaitspeedMobilitycomparedtotheusualcareat
theendofintervention.Thisdifferencewasofclinicalsignificance(MODERATECONFIDENCEIN
EFFECT)
Onestudy52of20participantsfoundnosignificantdifferenceinpreferredgaitspeed
Mobilitybetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Threestudies134177229of201participantsfoundthatcardiorespiratorytrainingwas
associatedwithstatisticallysignificantimprovementinpreferredgaitspeedMobility
comparedtotheusualcareattheendofintervention.Thisdifferencewasofclinical
significance(MODERATECONFIDENCEINEFFECT)
Gaitendurance(6MWTmetres)
Sevenstudies7417718094,125,229,267of630participantsfoundthatcardiorespiratorytrainingwas
associatedwithstatisticallysignificantimprovementingaitendurance(6MWTmetres)Mobility
comparedtotheusualcareattheendofintervention.(VERYLOWCONFIDENCEINEFFECT)
Threestudies74,125,267of425participantsfoundnosignificantdifferenceingaitendurance(6
MWTmetres)Mobilitybetweentheparticipantswhoreceivedcardiorespiratorytraining
NationalClinicalGuidelineCentre,2013.
387
andthosewhoreceivedusualcareattheendofintervention(VERYLOWCONFIDENCEIN
EFFECT)
Fourstudies17718022994of205participantsfoundthatcardiorespiratorytrainingwas
associatedwithstatisticallysignificantimprovementingaitendurance(6MWTmetres)
Mobilitycomparedtotheusualcareattheendofintervention.Thisdifferencewasnot
clinicallysignificant(LOWCONFIDENCEINEFFECT)
Maximalgaitspeed(m/secover10meters)
Onestudy94of36participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinmaximalgaitspeedMobility(m/secover10meters)comparedtothe
usualcareattheendofintervention,althoughthisdifferencewasnotofclinicalsignificance(Low
CONFIDENCEINEFFECT)
Comfortablegaitspeed(m/secover5to10meters)
Twostudies94,267of278participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementincomfortablegaitspeedMobilitycomparedtotheusualcare
attheendofintervention.(LOWCONFIDENCEINEFFECT)
Onestudy267of242participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementincomfortablegaitspeedMobilitycomparedtothe
usualcareattheendofintervention.(LOWCONFIDENCEINEFFECT)
Onestudy94of36participantsfoundcardiorespiratorytrainingwasnotassociatedwith
statisticallysignificantimprovementincomfortablegaitspeedMobilitycomparedtothe
usualcareattheendofintervention.(LOWCONFIDENCEINEFFECT)
Gaitendurance(m/min)
Threestudies5374229of154participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementingaitendurance(m/min)Mobilitycomparedtotheusualcare
attheendofintervention,althoughthisdifferencewasnotofclinicalsignificance(LOWCONFIDENCE
INEFFECT)
Twostudies5374of63participantsfoundnosignificantdifferenceingaitendurance(m/min)
Mobilitybetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(MODERATECONFIDENCEINEFFECT)
Onestudy229of91participantsfoundnosignificantdifferenceingaitendurance(m/min)
Mobilitybetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
6metrewalkingtime(sec)
Onestudy93of20participantsfoundnosignificantdifferencein6metrewalkingtime(sec)Mobility
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofintervention(LOWCONFIDENCEINEFFECT)
StrokeImpactScale
Twostudies244,267of262participantsfoundthatcardiorespiratorytrainingwasassociatedwitha
statisticallysignificantimprovementintheimpactofthestroke,asmeasuredbythestrokeimpact
scaleMobility,comparedtotheusualcareattheendofintervention.(MODERATECONFIDENCEIN
EFFECT)
Onestudy267of242participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementintheimpactofthestroke,asmeasuredbythestroke
impactscaleMobility,comparedtotheusualcareattheendofintervention.(MODERATE
CONFIDENCEINEFFECT)
NationalClinicalGuidelineCentre,2013.
388
Onestudy244of20participantsfoundnosignificantdifferenceinStrokeImpactScale(mobility
domain)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Peakactivityindex(steps/min)
Onestudy180of58participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinPeakactivityindex(steps/min)mobilitycomparedtotheusualcareat
theendofintervention(LOWCONFIDENCEINEFFECT)
Maximumsteprate
Onestudy180of58participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinMaximumstepratemobilitycomparedtotheusualcareattheendof
intervention(LOWCONFIDENCEINEFFECT)
BergBalancescale
Fivestudies1922994,177125of357participantsfoundnosignificantdifferenceinBergBalancescale
Physicalfunctionbetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Duringusualcare:Twostudies19,125of210participantsfoundnosignificantdifferencein
BergBalancescalePhysicalfunctionbetweentheparticipantswhoreceived
cardiorespiratorytrainingandthosewhoreceivedusualcareattheendofintervention
(MODERATECONFIDENCEINEFFECT)
Afterusualcare:Threestudies22994,177of147participantsfounddifferencesignificant
improvementinBergBalancescalePhysicalfunctionassociatedwithcardiorespiratory
trainingcomparedtousualcare.(LOWCONFIDENCEINEFFECT)
TimeUpandGomeasure
Threestudies229177,267of353participantsfoundthatcardiorespiratorytrainingsignificantlyimproved
TimedUpandGoresponsePhysicalfunctioncomparedtousualcareattheendofintervention.
Thisimprovementwasnotlargeenoughtoindicateclearclinicalbenefit(HIGHCONFIDENCEIN
EFFECT).
Duringusualcare:Onestudy267of142participantsthatcardiorespiratorytraining
significantlyimprovedTimedUpandGoresponsePhysicalfunctioncomparedtousual
careattheendofintervention.Thisimprovementwasnotlargeenoughtoindicateclear
clinicalbenefit(HIGHCONFIDENCEINEFFECT)
Twostudies177229of111participantsfoundnosignificantdifferenceintheTimeUpandGo
measurePhysicalfunctionbetweentheparticipantswhoreceivedcardiorespiratory
trainingandthosewhoreceivedusualcareattheendofintervention(LOWCONFIDENCEIN
EFFECT)
HealthrelatedQoL
Onestudy3of28participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinHealthrelatedQoL(SF36Emotionalrolefunctioningdomain)compared
totheusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Threestudies3,94,113of97participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36or
SF12Physicalfunctioningdomain)betweentheparticipantswhoreceivedcardiorespiratory
trainingandthosewhoreceivedusualcareattheendofintervention(VERYLOWCONFIDENCEIN
EFFECT)
NationalClinicalGuidelineCentre,2013.
389
Onestudy113of33participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36
Emotionalrolefunctioningdomain)betweentheparticipantswhoreceivedcardiorespiratorytraining
andthosewhoreceivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Onestudy113of33participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Mental
healthdomain)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Onestudy113of33participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Physical
Componentscale)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy113of33participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Mental
Componentscale)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Mood
Onestudy244of20participantsfoundnosignificantdifferenceinMood(BeckDepressionIndex)
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Twostudies19,267of302participantsfoundthatcardiorespiratorytrainingwasnotassociatedwith
statisticallysignificantimprovementinanxiety(HADSanxietyscore)comparedtotheusualcareat
theendofintervention(MODERATECONFIDENCEINEFFECT)
Threestudies19,113,267of60participantsfoundnosignificantdifferenceindepression(measuredby
HADSdepressionscoreortheGeriatricDepressionScale)betweentheparticipantswhoreceived
cardiorespiratorytrainingandthosewhoreceivedusualcareattheendofintervention(MODERATE
CONFIDENCEINEFFECT)
Endofretentionfollowup:
Casefatality
Onestudy134of81participantsfoundnosignificantdifferenceincasefatalitybetweenthe
participantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(VERYLOWCONFIDENCEINEFFECT)
RivermeadMobilityIndex
Onestudy19of66participantsfoundnosignificantdifferenceintheRivermeadMobilityIndex
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofretentionfollowup(MODERATECONFIDENCEINEFFECT)
NottinghamExtendedADL
Onestudy19of64participantsfoundnosignificantdifferenceintheNottinghamExtendedADL
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofretentionfollowup(MODERATECONFIDENCEINEFFECT)
PhysicalActivityandDisabilityScale
Onestudy180of58participantsfoundnosignificantdifferenceinthePhysicalActivityandDisability
Scalebetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusual
careattheendofretentionfollowup(LOWCONFIDENCEINEFFECT)
FrenchayActivitiesIndex
NationalClinicalGuidelineCentre,2013.
390
Onestudy134of79participantsfoundnosignificantdifferenceinFrenchayActivitiesIndexbetween
theparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcareatthe
endofretentionfollowup(LOWCONFIDENCEINEFFECT)
Maximumcyclingworkrate
Onestudy19of66participantsfoundnosignificantdifferenceinmaximumcyclingworkratebetween
theparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcareatthe
endofretentionfollowup(MODERATECONFIDENCEINEFFECT)
Bodymass(Kg)
Onestudy19of64participantsfoundnosignificantdifferenceinBodymass(Kg)betweenthe
participantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(MODERATECONFIDENCEINEFFECT)
Maximalgaitspeed
Threestudies19,74180of186participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementinmaximalgaitspeedMobilitycomparedtotheusualcareat
theendofretentionfollowup,althoughthisdifferencewasnotofclinicalsignificance(MODERATE
CONFIDENCEINEFFECT)
Twostudies19,74of128participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementinmaximalgaitspeedMobilitycomparedtotheusual
careattheendofretentionfollowup,althoughthisdifferencewasnotofclinical
significance(MODERATECONFIDENCEINEFFECT)
Onestudy180of58participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementinmaximalgaitspeedMobilitycomparedtotheusual
careattheendofretentionfollowup,althoughthisdifferencewasnotofclinical
significance(LOWCONFIDENCEINEFFECT)
6MinuteWalkTest
Twostudies74180of107participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementin6MinuteWalkTestMobilitycomparedtotheusualcareat
theendofretentionfollowup.Thisdifferencewasofclinicalsignificance(HIGHCONFIDENCEIN
EFFECT)
Onestudy74of49participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementin6MinuteWalkTestMobilitycomparedtotheusual
careattheendofretentionfollowup.Thisdifferencewasnotclinicallysignificant
(MODERATECONFIDENCEINEFFECT)
Onestudy180of58participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementin6MinuteWalkTestMobilitycomparedtotheusual
careattheendofretentionfollowup.Thisdifferencewasnotclinicallysignificant(LOW
CONFIDENCEINEFFECT)
Peakactivityindex(steps/min)
Onestudy180of58participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinpeakactivityindex(steps/min)Mobilitycomparedtotheusualcareat
theendofretentionfollowup(LOWCONFIDENCEINEFFECT)
Maximumsteprate
NationalClinicalGuidelineCentre,2013.
391
Onestudy180of58participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinmaximumsteprateMobilitycomparedtotheusualcareattheendof
retentionfollowup(LOWCONFIDENCEINEFFECT)
StrokeImpactScale
Onestudy244of20participantsfoundnosignificantdifferenceinStrokeImpactScale(mobility
domain)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceived
usualcareattheendofretentionfollowup(VERYLOWCONFIDENCEINEFFECT)
BergBalancescale
Onestudy19of66participantsfoundnosignificantdifferenceinBergBalancescalePhysical
functionbetweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceived
usualcareattheendofretentionfollowup(MODERATECONFIDENCEINEFFECT)
Mood
Onestudy244of20participantsfoundnosignificantdifferenceinMood(BeckDepressionIndex)
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofretentionfollowup(VERYLOWCONFIDENCEINEFFECT)
Onestudy19of53participantsfoundnosignificantdifferenceinMood(HADSanxietyscore)
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
attheendofretentionfollowup(MODERATECONFIDENCEINEFFECT)
Onestudy19of53participantsfoundthatcardiorespiratorytrainingwasassociatedwithstatistically
significantimprovementinMood(HADSdepressionscore)comparedtotheusualcareattheend
ofretentionfollowup(MODERATECONFIDENCEINEFFECT)
Onestudy113of31participantsfoundnosignificantdifferenceinMood(GeriatricDepressionScale)
betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewhoreceivedusualcare
at6monthspostintervention(LOWCONFIDENCEINEFFECT)
HealthrelatedQoL
Onestudy113of31participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Physical
Componentscale)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareat6monthspostintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy113of31participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Mental
Componentscale)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareat6monthspostintervention(LOWCONFIDENCEINEFFECT)
Onestudy113of31participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Physical
functioningdomain)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthose
whoreceivedusualcareat6monthspostintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy113of31participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36
Emotionalrolefunctioningdomain)betweentheparticipantswhoreceivedcardiorespiratorytraining
andthosewhoreceivedusualcareat6monthspostintervention(LOWCONFIDENCEINEFFECT)
Onestudy113of31participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Mental
healthdomain)betweentheparticipantswhoreceivedcardiorespiratorytrainingandthosewho
receivedusualcareat6monthspostintervention(VERYLOWCONFIDENCEINEFFECT)
Resistancetraining:Endofintervention
NationalClinicalGuidelineCentre,2013.
392
Musclestrength
Twostudies285136of60participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementincompositemeasureofmusclestrengthcomparedtousualcareattheend
ofintervention(LOWCONFIDENCEINEFFECT)
Onestudy285of40participantsfoundnosignificantdifferenceincompositemeasureof
musclestrengthbetweentheparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Onestudy136of20participantsfoundnosignificantdifferenceincompositemeasureof
musclestrengthbetweentheparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareattheendofintervention(LOWCONFIDENCEINEFFECT)
Kneeextension(Nm)
Twostudies1484of42participantsfoundnosignificantdifferenceinkneeextension(Nm)between
theparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
intervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy14of18participantsfoundnosignificantdifferenceinkneeextension(Nm)
betweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcare
attheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy84of24participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinkneeextension(Nm)comparedtousualcareattheendof
intervention(LOWCONFIDENCEINEFFECT)
Kneeflexion(Nm)
Twostudies1484of42participantsfoundnosignificantdifferenceinkneeflexion(Nm)betweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
intervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy14of18participantsfoundnosignificantdifferenceinkneeflexion(Nm)between
theparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareatthe
endofintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy84of24participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinkneeflexion(Nm)comparedtousualcareattheendof
intervention(LOWCONFIDENCEINEFFECT)
Maximalgaitspeed
Twostudies1484of42participantsfoundnosignificantdifferenceinmaximalgaitspeedbetweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
intervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy14of18participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinmaximalgaitspeedcomparedtousualcareattheendof
intervention,althoughthisdifferencewasnotofclinicalsignificance(VERYLOW
CONFIDENCEINEFFECT)
Onestudy84of24participantsfoundnosignificantdifferenceinmaximalgaitspeedbetween
theparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareatthe
endofintervention(LOWCONFIDENCEINEFFECT)
Preferredgaitspeed
Onestudy14of18participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinpreferredgaitspeedcomparedtousualcareattheendofintervention,
althoughthisdifferencewasnotofclinicalsignificance(VERYLOWCONFIDENCEINEFFECT)
RivermeadMobilityIndex
NationalClinicalGuidelineCentre,2013.
393
Onestudy47of68participantsfoundnosignificantdifferenceinRivermeadMobilityIndexbetween
theparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
intervention(MODERATECONFIDENCEINEFFECT)
Weightbearing(affectedside)
Onestudy14of18participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinweightbearing(affectedside)comparedtousualcareattheendof
intervention(VERYLOWCONFIDENCEINEFFECT)
Stairclimbing
Twostudies136195of61participantsfoundnosignificantdifferenceinstairclimbingbetweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
intervention(VERYLOWCONFIDENCEINEFFECT)
TimedUpandGo(sec)
Onestudy84of24participantsfoundnosignificantdifferenceinTimedUpandGo(sec)betweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
intervention(LOWCONFIDENCEINEFFECT)
HealthrelatedQoL
Onestudy136of20participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Physical
functioningdomain)betweentheparticipantswhoreceivedresistancetrainingandthosewho
receivedusualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Onestudy136of20participantsfoundnosignificantdifferenceinHealthrelatedQoL(SF36Mental
healthdomain)betweentheparticipantswhoreceivedresistancetrainingandthosewhoreceived
usualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
EuroQoL
Onestudy47of67participantsfoundnosignificantdifferenceinEuroQoL(Selfperceivedhealth)
betweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareatthe
endofintervention(MODERATECONFIDENCEINEFFECT)
Mood
Onestudy238of88participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinMood(CentreforEpidemiologyStudiesforDepressionscale)compared
tousualcareattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Endofretentionfollowup:
Kneeextension(Nm)
Onestudy84of24participantsfoundnosignificantdifferenceinkneeextension(Nm)betweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(LOWCONFIDENCEINEFFECT)
Kneeflexion(Newtonmetre)
Onestudy84of24participantsfoundnosignificantdifferenceinkneeflexion(Nm)betweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(LOWCONFIDENCEINEFFECT)
NationalClinicalGuidelineCentre,2013.
394
Maximalgaitspeed
Onestudy84of24participantsfoundnosignificantdifferenceinmaximalgaitspeedbetweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(MODERATECONFIDENCEINEFFECT)
RivermeadMobilityIndex
Onestudy47of51participantsfoundnosignificantdifferenceinRivermeadMobilityIndexbetween
theparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(LOWCONFIDENCEINEFFECT)
TimedUpandGo(sec)
Onestudy84of24participantsfoundnosignificantdifferenceinTimedUpandGo(sec)betweenthe
participantswhoreceivedresistancetrainingandthosewhoreceivedusualcareattheendof
retentionfollowup(LOWCONFIDENCEINEFFECT)
Mood
Onestudy238of86participantsfoundthatresistancetrainingwasassociatedwithstatistically
significantimprovementinMood(CentreforEpidemiologyStudiesforDepressionscale)compared
tousualcareattheendofretentionfollowup(VERYLOWCONFIDENCEINEFFECT)
EuroQoL
Onestudy47of49participantsfoundnosignificantdifferenceinEuroQoL(Selfperceivedhealth)
betweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusualcareatthe
endofretentionfollowup(MODERATECONFIDENCEINEFFECT)
CardiorespiratoryversusResistancetraining:
Preferredgaitspeed
Fourstudies52134177229of221participantsfoundthatcardiorespiratorytrainingwasassociatedwith
statisticallysignificantimprovementinpreferredgaitspeedMobilitycomparedtotheusualcare,
althoughthisdifferencewasnotofclinicalsignificance(LOWCONFIDENCEINEFFECT)
Threestudies14136195of80participantsfoundnosignificantdifferenceinpreferredgaitspeed
Mobilitybetweentheparticipantswhoreceivedresistancetrainingandthosewhoreceivedusual
careattheendofintervention(VERYLOWCONFIDENCEINEFFECT)
Economicevidencestatement
Nocosteffectivenessevidencewasidentified.
13.2.2
Recommendationsandlinkstoevidence
81.Encouragepeopletoparticipateinphysicalactivityafterstroke.
82.Assesspeoplewhoareabletowalkandaremedicallystableaftertheir
strokeforcardiorespiratoryandresistancetrainingappropriatetotheir
individualgoals.
NationalClinicalGuidelineCentre,2013.
395
83.Cardiorespiratoryandresistancetrainingforpeoplewithstrokeshould
bestartedbyaphysiotherapistwiththeaimthatthepersoncontinues
theprogrammeindependentlybasedonthephysiotherapists
instructions(seerecommendation84).
84.Forpeoplewithstrokewhoarecontinuinganexerciseprogramme
independently,physiotherapistsshouldsupplyanynecessary
informationaboutinterventionsandadaptationssothatwherethe
personisusinganexerciseprovider,theprovidercanensuretheir
programmeissafeandtailoredtotheirneedsandgoals.This
informationmaytaketheformofwritteninstructions,telephone
conversationsorajointvisitwiththeproviderandthepersonwith
stroke,dependingontheneedsandabilitiesoftheexerciseprovider
andthepersonwithstroke.
85.Tellpeoplewhoareparticipatinginfitnessactivitiesafterstrokeabout
commonpotentialproblems,suchasshoulderpain,andadvisethemto
seekadvicefromtheirGPortherapistiftheseoccur.
Relativevaluesof
differentoutcomes
TheGDGagreedthatbeingfithasanimpacton:cardiovascularmortality,obesity,
speed,enduranceandmood.
TheGDGconsideredthoseoutcomesmeasuringfitness,mobilityandmoodtobe
important.
Tradeoffbetween
clinicalbenefitsand
harms
Thereisageneralagreementthatphysicalactivityisbeneficial.TheCochrane
reviewdemonstratedimprovementsinphysicalfitness,mobilityandmoodandthat
respiratorytrainingimprovedspeed,toleranceandindependenceinwalking.The
GDGnotedthatallpeopleincludedwithinthestudieshadsomewalkingcapacityat
baseline.Thestudiesdidnotcommentonharmorpotentialsideeffects.Adverse
eventswerenotconsistentlyreportedwithinthesestudiesbutrequireserious
consideration.
TheGDGagreedthatpeopleneedtobecardiovascularlystable(theirtreatmentis
notchangingandtheirsymptomsarenotgettingworse),buthavingsymptomsdoes
notmeanthattheycannotexercise.
TheGDGagreedthatcardiovascularexercisewassafeundersupervisionforcertain
people.Itwasagreedthatanassessmentshouldbeundertakenbyahealth
professionaltoestablishsuitabilityforthistypeofinterventionbutthatthebenefits
ofexerciseinpreventingfurtherdeteriorationwereestablished.Thegroupagreed
thatmostfitnesstrainingisdoneinthecommunityandasoutpatients,andthere
arenowprogrammesavailableofadaptedfitnessprogrammessuitableforastroke
populationthatwouldbesafe.Ideallypeoplewouldstarttheirexerciseprogramme
undersupervisionofacommunityphysiotherapistandthenapersonaltrainerwho
hadknowledgeandexperienceofworkingwithpeoplewithdisabilitieswhowould
continuetheprogramme.
ShoulderpainisthemostlikelyharmpeopleexperienceandthereforetheGDG
agreedthatarecommendationtodirectpeopletoseekmedicaladviceshouldbe
made.
Economic
considerations
Cardiorespiratorytrainingisdeliveredaspartofusualrehabilitationprogrammesby
physiotherapists.Fitnesstrainingisdoneinthecommunityandthecostperhourof
acommunitybasedphysiotherapistis30.TheGDGacknowledgedthatadditional
costswouldbeincurredifpeoplearereferredfortrainingprogrammespost
rehabilitation;theGDGfeltthatthepotentialcostsofcardiorespiratoryandfitness
trainingarelikelytobeoutweighedbythebenefits.
NationalClinicalGuidelineCentre,2013.
396
Qualityofevidence
Theconfidenceintheeffectofspecifiedoutcomesrangefromhightoverylow.
Significanteffectswerefoundinthedomainswheresuchchangesmightbe
anticipatedwithcardiorespiratoryandresistanceinterventions(i.e.physicalfitness
andmobilityoutcomes).Thesefindingsareconsistentwithresultsinotherpatient
groupsforsimilarfitnessinterventionstudies.Inthegroupswhoreceived
cardiorespiratorytrainingsignificantimprovementswerefoundforgaitspeed,gait
enduranceandmeasuresofphysicalfitnessattheendofintervention(timepoint
whenatrainingprogrammefinishes)andwereretainedattheendoffollowup
(anytimepointoccurringaftertheendoftheintervention).Tworecentlargescale
studies125,267contributedtomanymobilityoutcomesandresultedinlarger
improvementsofhightolowqualityfortheseoutcomes.Thisalsoincludes
measuresofstrokeimpacttheevidenceofwhichwasratedasmoderatequality.
Otherconsiderations
Thegroupfeltthatparticipatinginexerciseshouldgenerallybeencouraged.The
GDGalsonotedthatthefindingsofthereviewwereuncontroversialinthatthey
highlightmobilityandfitnessasthemainhealthgains.
Therewasadiscussionaboutmedicalsuitabilityandtheroleofmedicaladvicewith
regardstocardiorespiratory/resistancetraining.Itwasfeltthattherecould
potentiallybeseriousmedicalrisksassociatedwithfitnesstrainingsincemany
peoplewhohavehadastrokealsohavecardiacconditions.Sincetherewouldbea
tradeoffbetweenbenefitsandpotentialharms,cliniciansshouldassessanddiscuss
thesewithpeoplewhohavehadastrokewhoareconsideringtakingpartinsuch
activities.TheGDGalsodebatedtheissueofindividualpreferenceandpersonal
history.Whetherapersonwantstotakepartinexercisetrainingcoulddependon
previousactivitylevelsandwork/lifecommitments.Thegroupacknowledgeda
distinctionbetweenthegoalsofthosewantingtoregainfunctionalabilityandthose
whowanttocommittoprogressivecardiorespiratoryexerciserequiringsignificant
timeandeffort.Whilstfitnesstrainingmaynotbesuitableorwantedbyeveryone,
itwasagreedphysicalactivityatwhateverlevelshouldbeofferedandpromoted.
TheGDGhighlightedtheStartactivestayactivereportpromotingregularphysical
activitythroughoutapersonslife62.Althoughfitnesstrainingtendstobedonemore
commonlyinthecommunitysettingratherthanintheacutesetting,studieshave
beendonesafelyintheacutesetting.
13.3 Handandarmtherapies:orthosesfortheupperlimb
Handorthoses,orsplints,areusuallylightweight,formedsupportsforprovidingprotection,rest,or
alignmentforthefingers,handandwrist.Afterstroke,ifhandfunctiondoesnotreturn,softtissue
tightnessandcontracturesoftenoccurleadingtosecondaryproblemsoffurtherlimitedfunction,
pain,oedemaandpossibly,worseningspasticity.Handsplintsaresometimesprovidedtoaidin
maintainingthelengthofsofttissuesandthustherangeofmotionofthejoints.Theyarealso
thoughttoreducetheeffectsofspasticity.However,thereisdifferingopinionwithregardstothe
design,schedulesandclinicalaimsforupperlimbsplinting,aswellasbothbiomechanicaland
neurophysiologicalclinicalrationales.Additionally,therearerespectedmembersofthetherapy
professionswhobothsupportandcontesttheuseofthisclinicaltool(LanninNA,2003145).
13.3.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
orthosesforpreventionoflossofrangeofmovementintheupperlimbversususualcare?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke.
Intervention:
Orthosesfortheupperlimbincluding:
softandscotchcasts,
splint,brace,lowtemperaturesplints,palmprotector,lycra
splinting
NationalClinicalGuidelineCentre,2013.
397
ClinicalMethodologicalIntroduction
alltheaboveinterventionswithorwithoutbotulinumtoxin,
13.3.1.1
Comparison:
UsualCare
Outcomes:
Rangeofmovementassessedbygoniometry
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingtheeffectivenessofdifferent
typesoforthosesasinterventionsforpreventionoflossofrangeofmovementintheupperlimbfor
adultsandyoungpeopleover16yearswhohadapreviousstroke.Onlystudieswithaminimum
samplesizeof20participants(10ineacharm)wereselected.TwoRCTs18,144wereidentified.Table
97summarisesthepopulation,intervention,comparisonandoutcomesofthestudy.
Table97: Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
Author
Year
POPULATION
INTERVENTION
COMPARISON
OUTCOME
Lannin,
2007144
Adultswhohadastroke
withintheprevious8
weeksandhadnoactive
wristextension.
Twosplintinginterventions
werestudied.Inboth
groups,custommade,
static,palmarmittsplints
forupto12hoursovernight
for4weekswereused.
Neutralsplint;participants
woreahandsplintwhich
positionedthewristin00to
100extension.(N=21)
Extensionsplint:
participantsworeahand
splint,whichpositionedthe
wristinacomfortableend
ofrangeposition(>450wrist
extension)withthe
metacarpophalangealand
interphalangealjoints
extended.(N=21)
Controlgroupdid
notwearahand
splintforthestudy
period.(N=21)
Wrist
extensibility
(indegrees)
at4and6
weeks
Basaran
201218
N=39participantswitha
historyofsinglestroke
andwristMASscore1+;
iftakingantispasticity
drugs,dosagehadtobe
stableduringprevious
month
Twosplintinginterventions
werestudied:staticdorsal
splintwornforupto10
hoursovernightfor5weeks
orstaticvolarsplintworn
forupto10hoursovernight
for5weeks
Allpatientshad
Passive
homeexercise
rangeof
programincluding
motion
motortrainingand
(PROM)of
stretching,reaching
wrist
andgrasping3
extension
timesadayand
(goniometer
advisedtouse
)
handsasmuchas
possibleduringthe
dayfor5weeks
NationalClinicalGuidelineCentre,2013.
398
StrokeRehabilitation
Movement
Comparison:Neutralsplintversususualcare
Table98: NeutralsplintversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Neutral
splint
Mean(sd)
Effect
Usual
care
Mean
(sd)
Mean
difference
(95%ci)
Mean
Difference
Confidence
(MD)(95%CI) (ineffect)
Wristextensibility(indegrees)(4weeksfollowup)(Betterindicatedbyhighervalues)
Lanninet RCT
al,
2007144
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
53.1(14.9)
47.3
(16.9)
1.40(8.20,
5.40)
Wristextensibility(indegrees)(6weeksfollowup)(Betterindicatedbyhighervalues)
Lanninet RCT
al,
2007144
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
Moderate
48.8(14.5)
39.4
(17.8)
4.20(
11.65,3.25)
O
(a) MeandifferencedidnotreachtheagreedMIDof5 .
NationalClinicalGuidelineCentre,2013.
MD1.4lower
(8.2lowerto
5.4higher)
399
MD4.2lower Moderate
(11.65lower
to3.25higher)
StrokeRehabilitation
Movement
Comparison:Extensionsplintversususualcare
Table99: ExtensionsplintversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Extension
splint
Mean(sd)
Effect
Usual
care
Mean
(sd)
Mean
difference
(95%CI)
Mean
Difference
(MD)(95%CI)
Wristextensibility(indegrees)(4weeksfollowup)(Betterindicatedbyhighervalues)
Lanninet RCT
al,
2007144
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
45.5(15.4)
47.3
(16.9)
1.30(2.58,
5.18)
MD1.3higher
(2.58lowerto
5.18higher)
Wristextensibility(indegrees)(6weeksfollowup)(Betterindicatedbyhighervalues)
Lanninet RCT
al,
2007144
(a)
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
42.5(14.9)
39.4
(17.8)
0.80(5.05,
3.45)
MeandifferencedidnotreachtheagreedMIDof5 .
NationalClinicalGuidelineCentre,2013.
Moderate
Confidence
(ineffect)
400
MD0.8lower
(5.05lowerto
3.45higher)
Moderate
StrokeRehabilitation
Movement
Comparison:Dorsal/Volarsplintversususualcare
Table100:Dorsal/VolarsplintversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecisio
n
Static
dorsal/
volarsplint
Mean
change
from
baseline
(SD)
Effectsize
Usual
care
Mean
change
from
baseline
(SD)
Mean
difference
between
interventi
onand
control
(95%CI)
Mean
difference
(95%CI)
Confidence
ineffect
Passiverangeofmotion(PROM)ofwristextensionDorsalsplint(followup5weeks;measuredwith:meanchangescores;Betterindicatedbyhighervalues)
Basaran
2012
randomised
trials
noserious
riskofbias
noserious
inconsistency
noserious
indirectness
very
serious(a)
2.31(8.07)
0.42(4.5) 1.89(
3.18,6.96)
MD1.89higher Low
(3.18lowerto
6.96higher)
Passiverangeofmotion(PROM)ofwristextensionVolarsplint(followup5weeks;measuredwith:Meanchangescores;Betterindicatedbyhighervalues)
Basaran
2012
(a)
(b)
randomised
trials
noserious
riskofbias
noserious
inconsistency
noserious
indirectness
serious(b)
3.46(7.18)
0.42(4.5) 3.04(
1.62,7.70)
The confidence interval reaches from appreciable harm to appreciable benefit of the intervention (i.e. crosses both default MIDs)
The confidence interval ranges from appreciable benefit to no effect (crosses one default MID)
NationalClinicalGuidelineCentre,2013.
401
MD3.04higher Moderate
(1.62lowerto
7.7higher)
StrokeRehabilitation
Movement
13.3.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingorthosesforpreventionoflossofrangeoftheupper
limbwithusualcarewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Thecostofprovidingwrist/handorthoseswasbasedontheRCTincludedinclinicalreview.InLannin
etal(2007)144,theyusedcustommadestatic,palmarmittsplints.AnexpertadvisortotheGDG
providedthecostforaprefabricatedsplint:Restingpanpositionsplint,33.93excludingVAT,
thoughcostswillvaryaccordingtotypeanddesignofprefabricatedsplint.
Custommadeorthoseswouldbemadebyamemberofspecialistmultidisciplinaryorthoticsteam
andwouldincurextracosts.Inaddition,therewouldbepersonnelcostsrelatedtothetimerequired
tomakeandadjusttheULOtotakeintoaccountthespecificpatientsneeds.Adjustmentsmaybe
madebyeitherorthotistsandexperiencedphysiotherapistsoroccupationaltherapists(band6or7),
dependingontherequirements(forexampleorthotiststendtomakepermanentandmorecomplex
adjustments).Theestimatedcostsrangefrom45to59perhourofclientcontactr.
13.3.1.3
Evidencestatements
Clinicalevidencestatements
Neutralsplint
Onestudy144comprising42participantsfoundnosignificantdifferenceonthewristextensibilityat4
and6weeksfollowupforparticipantswearingneutralsplintfor4weekscomparedtoparticipants
whoreceivedusualcare.(MODERATECONFIDENCEINEFFECT)
Extensionsplint
Onestudy144comprising42participantsfoundnosignificantdifferenceonthewristextensibilityat4
and6weeksfollowupforparticipantswearingextensionsplintfor4weekscomparedtoparticipants
whoreceivedusualcare.(MODERATECONFIDENCEINEFFECT)
Dorsalsplint
Onestudy18comprising26participantsfoundnosignificantdifferenceonthepassiverangeof
motion(PROM)ofwristextensionafter5weeksofinterventionforparticipantswearingastatic
dorsalsplintcomparedtoparticipantswhoreceivedusualcare.(LOWCONFIDENCEINEFFECT)
Volarsplint
r EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandAgendaforChangesalaryband6and751(typicalsalarybandidentifiedbyclinicalGDGmembers).Assumed
thatanorthotistiscostedsimilartoaphysiotherapist.
NationalClinicalGuidelineCentre,2013.
402
StrokeRehabilitation
Movement
Onestudy18comprising26participantsfoundnosignificantdifferenceonthepassiverangeof
motion(PROM)ofwristextensionafter5weeksofinterventionforparticipantswearingastaticvolar
splintcomparedtoparticipantswhoreceivedusualcare.(MODERATECONFIDENCEINEFFECT)
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.3.2
Recommendationsandlinktoevidence
86.Donotroutinelyofferwristandhandsplintstopeoplewith
upperlimbweaknessafterstroke.
87.Considerwristandhandsplintsinpeopleatriskafterstroke
(forexample,peoplewhohaveimmobilehandsdueto
weakness,andpeoplewithhightone),to:
maintainjointrange,softtissuelengthandalignment
increasesofttissuelengthandpassiverangeofmovement
facilitatefunction(forexample,ahandsplinttoassistgrip
orfunction)
aidcareorhygiene(forexample,byenablingaccesstothe
palm)
increasecomfort(forexample,usingasheepskinpalm
protectortokeepfingernailsawayfromthepalmofthe
hand).
88.Wherewristandhandsplintsareusedinpeopleafterstroke,
theyshouldbeassessedandfittedbyappropriatelytrained
healthcareprofessionalsandareviewplanshouldbe
established.
89.Teachthepersonwithstrokeandtheirfamilyorcarerhowto
putthesplintonandtakeitoff,careforthesplintand
monitorforsignsofrednessandskinbreakdown.Providea
pointofcontactforthepersonifconcerned.
Recommendations:
Relativevaluesofdifferent
outcomes
Wristsplintsareusedtoensurethatrangeofmovementisnotlost
followingstroke.Shouldlossofrangeofmovementoccurthiswould
haveanimpactonupperlimbfunctionshouldmovementatthehand
andwristrecover.Forthisreasontheoutcomeofinterestincludedin
thereviewwasrangeofmovement.TheGDGnotedthatrangeof
movementisoneofanumberofpotentialoutcomeswhichinclude
functionandamountofmuscletone
Tradeoffbetweenclinical
benefitsandharms
Anassessmentoftheuseofsplintsincludescheckingifitfitsandisworn
properly,andareviewplanestablished.TheGDGagreedthat
informationandtrainingforthepatientandcarerswasimportantfor
themtoensurethesplintwasusedcorrectlyandtorecogniseany
adverseeffectsthatwouldneedprofessionalcareandadvice.
Potentialcontraindicationsmayincludesensoryimpairment,spasticity,
poorskinconditionincludinginflammation,oedema,andpoorvascular
NationalClinicalGuidelineCentre,2013.
403
StrokeRehabilitation
Movement
supply,eachofwhich maycontribute toskinbreakdownafterstroke.
TheGDGagreeditwasessentialthatthesplintisassessed,fittedand
monitoredbystafftrainedinthisarea.
Economicconsiderations
Nocosteffectivenessstudieswereidentifiedforthisquestion.Thecost
ofaprefabricatedsplintwasestimatedataround34excludingVAT
thoughcostsvaryconsiderably.Inaddition,thereissomepersonneltime
requiredtoassessandmakeadjustmentsforthepatientaswellasto
ensureitscorrectuse.
Custommadeorthoseswouldbemadebyamemberofspecialist
multidisciplinaryorthoticsteamandwouldincurextracosts.Inaddition,
therewouldbepersonnelcostsrelatedtothetimerequiredtomakeand
adjusttheULOtotakeintoaccountthespecificpatientsneeds.
Basedontheresultsoftheclinicalreview,theGDGdidnotconsider
wristandhandsplintingtobecosteffectiveasaroutinetreatmentin
themajorityofpatients.
Qualityofevidence
Onlytwosmallstudies,oneinearlypostacutestrokeandoneinlater
strokerehabilitationwereidentified18,144.Confidenceintheresultsseen
inthewristextensibilityoutcomewasmoderatetolowduetonot
reachingtheminimalimportantdifferenceof5degreesaspreviously
agreedwiththeGDG(seethemethodchapter)orthedefaultMID.
HowevertheGDGconsideredbothofthesestudiestobearobust
rehabilitationstudiesandacknowledgedthedifficultiesofdouble
blindinginthistypeofintervention.
Otherconsiderations
TheGDGconsensuswasthatroutinesplintingearlyafterstrokewould
probablynotbeofbenefit,exceptinselectedpatientswheresplinting
maybeusedtohelpmanagetone,reducepainandimprovefunction.
Whetherornottousesplintslateronintherehabilitationpathwayswas
seenasunclearsincetheresultscanbeinterpretedasbeing
inconclusive.
Furtherresearchisneededtoassesswhetherupperlimbsplintingin
conjunctionwithothermodalitiesaidsthemanagementofspasticity.
Todate,thedetailsarenotknownastowhetherupperlimbsplintingis
usefulinreducingproblemsinthepoorlyfunctioninghandafterstroke
eitherasasingleinterventionorincombinationwithotherinterventions
suchasbotulinumtoxininjectionsorelectricalstimulation.
13.4 Electricalstimulation:upperlimb
Functionalelectricalstimulation(FES)andneuromuscularelectricalstimulation(NMES),usedhereto
indicateagenericformoftherapeuticelectricalstimulation(ES)tomuscles,areanadjuncttoa
comprehensiverehabilitationprogramtoimprovearmandhandfunctionafterstroke.Itmaybe
usedfortherapeuticpurposesorforfunctionalpurposes.ESisseldomusedinisolation,butmost
recentlyintandemwithorinadditiontoanactivetaskoriented,exerciseprogram.ES,applied
usuallyviasurfaceelectrodes,butalsooccasionallythroughimplantedelectrodes,activatesmuscle
contractionperipherallyusuallythroughstimulatingnervestomuscles.Withcurrenttechnology,ES
devicesaresmallandeasytouseandcanbepreprogrammedtoprescribedcyclesandduration,
includemultiplemusclegroups,bepassivelyoractivelytriggeredandbeusedinsomefunctional
activities.Oncesetupbytheappropriatehealthprofessional,thetreatmentcanoftenbecontinued
athome,enhancingthepracticeeffect.
13.4.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
electricalstimulation(ES)forhandfunctionversususualcare?
ClinicalMethodologicalIntroduction
NationalClinicalGuidelineCentre,2013.
404
StrokeRehabilitation
Movement
13.4.1.1
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
FunctionalElectricalStimulation(FES)withorwithoutrobotics,
ESwithorwithouttranscranialmagneticstimulationor
neuromuscularelectricalstimulation(NMS)
Comparison
Usualcare
Outcomes
Anyoutcomereportedinthepaper.
UpperLimboutcomesincluding:
ActionResearchArmTest(ARAT),
FuglMeyerAssessment(FMA),
9holepegtest,
gripstrength.
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingtheeffectivenessofelectrical
stimulation(ES)toimprovehandfunctionforpatientsover16yearsoldwithstroke.Eighteen(18)
RCTswereidentified.Table101summarisesthepopulation,intervention,comparisonandoutcomes
foreachofthestudies.
Table101:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Alon,2007
Patientswith
unilateralacute
stroke(24weeks
ago)whowere
medicallystablewith
aparesisofupper
limb(FuglMeyer
score1140)without
limitedrangeof
passivemotionand
atleast60%finger
flexion/extension
withstimulation.
FESplusstandard
rehabilitation
training(usualcare);
electricalstimulation
andinduced
contractionof
wrist/finger
extensors.FESwas
providedinitiallyfor
10min/4timesdaily
in2sessionsaspart
oftheirstandard
exerciseand2
sessionsasnot;the
durationincreased
by5min/dayto1
hourpersession4
timesdaily(FESwas
synchronisedwith
exercisefor2
sessions,eachlasted
30minandtherest
ofFEStrainingwas
notsynchronized).
(N=7)
Usualcare;standard
rehabilitation
exercise(duration3
hours)included
physical,
occupational(30min
twicedaily5daysa
weekduring
hospitalisation)and
speechtherapy
within12daysof
admission.After
discharge,patients
practised30min
twicedaily
unsupervised
rehabilitation
exercisefor12visits
perweek.(N=8)
BoxandBlocks
test
JebsenTaylor
HandFunction
test(lightcans)
ModifiedFugl
Meyer
Assessment
Alon,20087
Patientswith
unilateralacute
stroke(24weeks
ago)whowere
medicallystablewith
FESplusindividually
tailoredexercise
regimen(usualcare).
FESdurationwas
increasedby5min/
Usualcare:
BoxandBlocks
individuallytailored
test,
exerciseregimenfor JebsenTaylor
30minsessionstwice
HandFunction
dailyfor5days/week
test(lightcans),
NationalClinicalGuidelineCentre,2013.
405
StrokeRehabilitation
Movement
STUDY
POPULATION
aparalysis/paresisof
upperlimb(Fugl
Meyerscore210)no
limitedrangeof
passivemotion;at
least60%finger
flexion/extension
withstimulation
INTERVENTION
dayto1hour
practicedfor4times/
day.(N=13)
COMPARISON
during
hospitalisation;after
discharge,usualcare
wasprovidedfor30
mintwicedaily
withoutsupervision
(occupation
therapists/physiother
apistswereinvolved
in12sessionsper
week).(N=13)
OUTCOMES
Cauraugh,
200038
Patientswithchronic
stroke(longerthana
yearago)with
chronicupper
extremity
impairment(upper
limitcutoffof75%
motorrecoveryand
forlowerlimitcutoff
point);subjectshad
tobecapableof
voluntarilyextending
thewrist20against
gravityfroma90
flexionposition.
Electromyography
triggered
Neuromuscular
Stimulation(EMG):
subjectswere
instructedtoinitiate
finger/wrist
extensionsothata
targetthresholdof
EMGactivitywas
voluntarilyachieved;
EMGwasprovidedin
2sessionsof30
movementtrials
(around60min)
during3daysaweek
for2weeks.(N=7)
Usualcare:patients
followedthesame
procedureasthe
experimentalgroup
exceptthattheydid
notreceivethe
neuromuscular
electricalstimulation.
(N=4)
BoxandBlocks
test
Motor
AssessmentScale
FuglMeyer
Assessment
2force
generationtasks
Cauraugh,
200239
Patientswithatleast
onecerebrovascular
accident(CVA)and
nomorethan2CVAs
onthesamesideof
thebrain(upperlimit
cutoffof80%motor
recovery;forlower
limitcutoffpoint;
subjectshadtobe
capableofvoluntarily
extendingthe
fingers/wrist10
againstgravityfroma
90flexionposition.
Unilateraltraining
group:EMG
triggeredstimulation
toassistwristand
fingerextension
(unilateral
movement).(N=10)
Bilateraltraining
group:EMG
triggeredstimulation
plusassistancefrom
unimpairedlimbas
wrist/finger
extensionexecuted
simultaneouslyon
bothlimbs(bilateral
movement).3setsof
30successfulEMG
triggeredtrials
(around1.5hours);
totalof6hoursof
trainingon4days
during2weeks.
(N=10)
Usualcare:subjects
triedtovoluntarily
extendwrist/fingers
for5seconds
followedby25
secondsrest,
repeatedlyfor90
min.persession.
(N=5)
BoxandBlocks
test
Sustained
muscle
contraction
Chae,1998
41
Strokesurvivors
admittedtoanacute
15sessionsof
stimulatingthe
15sessionsofsurface FuglMeyer
stimulation,butthe
Upperextremity
NationalClinicalGuidelineCentre,2013.
406
ModifiedFugl
Meyer
Assessment
StrokeRehabilitation
Movement
STUDY
POPULATION
inpatient
rehabilitationservice
within4weeksof
theirunilateral
stroke.Patientswere
18yrsoldorolder
withmoderateto
severeupper
extremityparesis
(FuglMeyerscore
lessthan44),withno
historyofpotentially
fatalcardiac
arrhythmias.
INTERVENTION
extensordigitorum
communisandthe
extensorcarpi
radialis(ECR)through
circular2.5cm
surfaceelectrodesin
additiontostandard
physical,
occupational,and
speechtherapy
interventions.The
stimulationcurrent
intensitywassetto
producefullwrist
andfingerextension
withadutycycleof
10secondsonand10
secondsoff.
(N=14)
COMPARISON
electrodeswere
placedawayfromall
motorpoints,
producingonly
cutaneous
stimulationjust
beyondsensory
thresholdand
withoutmotor
activationinaddition
tostandardphysical,
occupational,and
speechtherapy
interventions.
(N=14)
OUTCOMES
assessment
Functional
Independence
Measure
Chan,
200942
Patientswithfirst
timestrokelonger
than6weeksago
withascoreof0on
fingermassextension
subitemofFugl
MeyerAssessment.
FESwithmuscle
movement(20min)
plus10min
stretching/passive
mobilization
activitiestofacilitate
activemovement
plus60min
conventional
occupationaltherapy
training.EachFES
sessionlasted20
min,with2training
activitiesoutof4
tasks.(N=10)
Usualcare:
stretching/passive
mobilization
activitiestofacilitate
activemovement(10
min)plusplacebo
electricalstimulation
withsensationonly
(20min)plus
conventional
occupationaltherapy
training(60min).
(N=10)
FunctionalTest
forthe
Hemiplegic
UpperExtremity,
FuglMeyer
Upperextremity
assessment
GripPower
ActiveRangeof
Motionwrist
extension
Functional
Independence
Measure
Modified
AshworthScale
ofshoulder,
elbow,wrist
Forwardreaching
distance
Hara,
2008105
Patientswithchronic
stroke(1or2strokes
onsamesideofbrain
longerthanayear
ago)andchronic
spasticupper
extremity
impairments(Stroke
Impairment
AssessmentSet
[SIAS]scores05)
withpassiverangeof
motioninwrist
extensionto45from
neutralandshoulder
FESforwrist
extensionduring
coordinated
movement(triggered
byvoluntary
movementby
patient)for30
minute/sessionfor5
daysaweekathome.
FEStraining
increasedover10
daystoamaximum
of1hourpersession.
Supervisionwas
providedbyan
occupational
Supervisionbya
rehabilitationtrainer
inextendingthe
impairedwristsand
fingersduring
rehabilitation
sessionsonceaweek
forabout5months.
Eachsessionlasted
approximately40
min.(N=10)
ActiveRangeOf
Movement
(ROM)
electromyographi
cmeasures
(maximum
isometric
contraction)
modified
Ashworthscale
(MAS)
9holepegtest
NationalClinicalGuidelineCentre,2013.
407
StrokeRehabilitation
Movement
STUDY
POPULATION
flexionto140.
INTERVENTION
therapistfor40min
onceaweekfor5
months.(N=10)
COMPARISON
OUTCOMES
Hsu,2010117 Acutestrokepatients
withunilateral
stroke,onsetwithin3
months,Brunnstrom
stageIVwithno
contraindicationfor
NMES.
Patientswere
randomisedto3
groups:high
Neuromuscular
ElectricalStimulation
(NMES),lowNMES,
orcontrol.
(N=22)
4weeksofNMES,5
timesperweekin
additiontoregular
inpatient
rehabilitation.
30minutesper
sessionwaschosen
asthelowdose
NMESand60
minutesasthehigh
doseNMES.
(N=22)
Regularinpatient
rehabilitation
FuglMeyer
Upperextremity
assessment
ActionResearch
ArmTest
MotorActivity
Log
Kimberley,
2004137
Patientswithchronic
stroke(longerthan6
monthsago)
experiencingatleast
10active
flexion/extensionat
metacarpophalangeal
jointofindexfinger;
MiniMentalState
Examinationscore25
ormoreoutof
possible30.
Intensivehomeuse
ofFES:60hourstotal
use;6hoursperday
for10daysover3
weeks;halfofthe
timeusingactive
effortbysubjectto
triggerstimulated
responsethenFES
contractingmuscles;
otherhalfmachine
automatically
stimulatingmuscles
tocontractcyclically
withouttriggerfrom
patient.(N=8)
Shamtreatment;
lightcameonbutno
deliveredby
machine.(N=8)
BoxandBlocks
test
MotorActivity
Log:amountof
usescore(AS),
howwellscored
(HW)
JebsenTaylor
HandFunction
Test:pageturn,
smallobjects,
feeding,stacking,
lightcans,heavy
cans(all
measuredinsec).
Strengthof
fingerextension,
Fingertracking
accuracytest
Mangold,
2009167
Firsttimestroke
patients(218weeks
ago)withsevere
hemiparesistototal
hemiplegiaofarm
and/orhand
(maximumvalueof
ChedokeMcMaster
StrokeAssessment
[CMSA]forarmand
hand3points).
FESlasted45
min/sessionincluding
1520minofputting
onandtakingoffFES
/treatingspasticity
and2530minof
functionaltraining
andwasofferedin3
sessions/weekIf
necessary,therapists
treatedspasticityand
providedmanual
assistance.FES
triggeredbypatient
ortherapist.(N=12)
Conventionaltraining Nottingham
providedfor35
Activitiesofdaily
occupationaltherapy
Living
sessionsperweek;
Chedoke
eachsessionlasted
McMasterStroke
45min.Mobilisation
Assessment(arm,
andexercises
hand)
supportedby
Modified
therapistor
AshworthScale
performed
(fingerflexors,
bimanually)(N=11).
wristflexors)
Mann,
2005168
Firsttimestroke
patients(112
FES:stimulationto
givefullelbow,wrist
Passiveextension
exercisesofelbow,
NationalClinicalGuidelineCentre,2013.
408
ActionResearch
ArmTest
StrokeRehabilitation
Movement
STUDY
POPULATION
monthspoststroke)
withhemiplegia
(medicallystable)
andabletotake
hemiplegichandto
mouth.
INTERVENTION
andfingerextension
withoutdiscomfort;
increasedfrom10to
30mintwiceaday
overaround1week.
(N=11)
COMPARISON
wristandfingersto
bepractisedforthe
sameperiodeach
day.Ondischarge,
continuedwithhelp
ofcaregiveror
independently.
(N=11)
OUTCOMES
Popovic,
2003210
Firsttimestroke
patients(2weeks6
monthspoststroke)
abletounderstand
howtoapplyFESto
controlgrasp.
Patientswere
groupedtoeither
higherfunctioning
group(ableto
activelyextendthe
pareticwristmore
than20oandextend
their
metacarpophalangeal
andinterphalangeal
jointsofalldigits
morethan20o)orto
thelowerfunctioning
group(patientscould
extendtheparetic
wristbetween10o
20o).
FESincluded
reaching,grasping
andusingobjectsand
returningthemto
theirplacesduring
thefirst3weeks.FES
exerciseincluded30
minlongtreatment
sessionsofexercise
withstimulationfor7
days/week.
HighfunctioningFES:
(N=8)
LowfunctioningFES:
(N=6)
Usualcare:
conventionaldaily
therapyfor26weeks.
Forthefirst3weeks,
controlgrouphad30
minlongtreatment
sessionsofexercise
only.
Highfunctioning
usualcare:(N=8)
Lowfunctioning
usualcare:
(N=6)
UpperExtremity
FunctionTest
(UEFT)
DrawingTest(%
ofareaofsquare
correctly
captured)
Modified
Ashworthscale
ReducedUpper
ExtremityMotor
ActivityLog
questionnaire
(RUE/MAL);
maximumscore,
howwellscale
was
Powell,
1999212
Acutestrokepatients
(24weeksafter
stroke)withMedical
ResearchCouncil
powerofwrist
extensiongrade4/5
orworseat.
FESplusstandard
physiotherapy:3
halfhourperiods
daily(total90mins.
daily)for8weeks.
(N=30)
Usualcare:standard
physiotherapy
includingdiscussing
progressin
rehabilitationupto
10minfor3times
weeklytocontrolfor
similarcontact
beforeandafterFES
sessions.
(N=30)
Gripstrength
activeand
passiverangeof
motion
AshworthScale
ActionResearch
ArmTest
Numberofpegs
persecond
Rankinscale
BarthelIndex
Thrasher,
2008257
Hemiplegicpatients
hospitalisedfor
recentstroke(27
weekspoststroke)
withascoreof1or2
forcombinedarm
andhandon
ChedokeMcMaster
StagesofMotor
Recovery(CMSMR,
i.e.spasticorflaccid
paralysisofarmand
FESplusconventional
occupationaltherapy
andphysiotherapyto
shoulder,elbow,
wristandhand5
daysperweekfor12
16weeks;each
sessioncombined
withFESfor45mins.
ofthesession;
stimulatorresponded
topushbutton
Conventional
occupationaltherapy
andphysiotherapyto
shoulder,elbow,
wristandhand5
daysperweekfor12
16weeks;each
sessionlasted45min
(musclefacilitation,
repetitivefunctional
training,
strengtheningagainst
Object
manipulation
Palmargrip
torque
Pinchgrippulling
force
BarthelIndex
UpperExtremity
FuglMeyer
Assessment
UpperExtremity
NationalClinicalGuidelineCentre,2013.
409
StrokeRehabilitation
Movement
STUDY
POPULATION
handwithlittleorno
voluntary
movement).
INTERVENTION
commandby
therapistwhen
patienttried
unsuccessfullyto
performtask;
therapistguidedarm
toensureanormal
movement.Inearly
stages,all
movements
performedwithFES;
inlatertreatments
FESusedless.(N=10)
COMPARISON
resistance,electrical
stimulationfor
isolatedmuscle
strengthening(not
forfunctional
training),activitiesof
dailylivingincluding
selfcareand
caregivertraining.
(N=11)
OUTCOMES
Chedoke
McMasterstages
ofMotion
Recovery
Lin,2011155
Firststrokepatients
(within3months
postonset)with
hemiplegiaofone
upperlimband
shoulderflexor
strengthbefore
treatmentwasgrade
3orless(outof5).
Patientsdidnothave
severecognitive
dysfunction(they
scored7orbetteron
theabbreviated
mentaltest).
Standardtreatment
plusNeuromuscular
ElectricalStimulation
(NES)lastingfor30
minutes,5
days/weekfor3
weeks.The2channel
RespondSelectII
stimulator(Texas,
USA)wasused.
(N=23)
Standardtreatment,
includingphysical
therapyand
occupational
therapy,for30
minuteson5days
/weekfor3weeks.
(N=23)
Modified
AshworthScale
(MAS)
Upperlimb
sectionofthe
FuglMeyer
Assessment
(FMAU)
ModifiedBarthel
Index(MBI)
Sahin,2012
227
Patientsbetween45
65yearsofage,who
haddeveloped
forearmflexor
spasticityfollowinga
stroke.Hemiplegia
waslongerthanone
year;score2or3
spasticityaccording
toModified
AshworthScale
(MAS)andastable
neurologicalstate.
Neuromuscular
ElectricalStimulation
(NMES)and
stretchingwith
Proprioceptive
Neuromuscular
facilitation(PNF)
appliedtotheupper
extremityafterhot
treatmentwith
infrared:5daysa
weekfor20sessions.
(N=21)
Stretchingwith
Proprioceptive
Neuromuscular
facilitation(PNF)
appliedtotheupper
extremityafterhot
treatmentwith
infrared:5daysa
weekfor20sessions.
(N=21)
Wristspasticity
(MAS)
Wristextension
RangeofMotion
(degrees)
Brunnstrom
motorscale
(upper)
Functional
Independence
Measure(FIM)
Electrophysiologi
calevaluation:
Fmax/Mmax,
Hmax/Mmax
Shindo
2011237
Participantswithfirst
timeunilateral
supratentorialstroke;
strokeonsetwithin
60days;age2080
years;muscle
activitiesinthe
affectedextensor
digitorumcommunis
(EDC)detectablewith
surfaceelectrodes;
couldnotfullyextend
3weeksofHybrid
Assistive
Neuromascular
DynamicStimulation
(HANDS)therapy:
neuromuscular
electricalstimulation
withintegrated
volitionalelectrical
stimulator(IVES)plus
wristsplintfor8
hoursadayplus
Worethesamewrist FuglMeyer
splintfor8hoursa
Assessment
dayplusstandard
(proximaland
rehabilitation(1hour
distal),
physicaltherapyand ActionResearch
1houroccupational
ArmTest,
therapyperday,5
MotorActivity
daysaweekplus
LogModified
speechtherapyif
AshworthScale
indicated);instructed
tousetheaffected
handasmuchas
NationalClinicalGuidelineCentre,2013.
410
StrokeRehabilitation
Movement
STUDY
POPULATION
pareticfingersand
couldnotextend
pareticfingers
individually;passive
rangeofmotion>0
fortheaffectedwrist
extensionand10
for
metacarpophalangeal
jointextension;Mini
MentalState
Examinationscore
>23
INTERVENTION
COMPARISON
standard
possibleinactivities
rehabilitation(1hour ofdailyliving(N=12)
physicaltherapyand
1houroccupational
therapyperday,5
daysaweekplus
speechtherapyif
indicated);instructed
tousetheaffected
handasmuchas
possibleinactivities
ofdailyliving(N=12)
OUTCOMES
Rosewilliam
2012220
Adultpatientswitha
firststrokewhohad
noarmfunction
(score0intheGrasp
subsectionofthe
ActionResearchArm
Test)within6weeks
ofonsetandno
contraindicationsto
surface
neuromuscular
electricalstimulation
(sNMES).
sNMESfor6weeks:
electricalstimulators
towristandfinger
extensorsatleast
twiceadayfor30
minutesessionsfor5
daysaweekplusa
definedmoduleof
upperlimb
physiotherapyfor6
weeksinadditionto
routinetreatmenton
thestrokeunit
(N=45)
ARAT
NationalClinicalGuidelineCentre,2013.
411
Adefinedmoduleof
upperlimb
physiotherapyfor6
weeksinadditionto
routinetreatmenton
thestrokeunitonly
(N=45)
StrokeRehabilitation
Movement
Comparison:Electricalstimulationversususualcare
Table102:ElectricalstimulationversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
Confide
/Mean
Difference
nce(in
(MD)(95%CI) effect)
BoxandBlocks(numberofblocksmovedin1minute)Postintervention(Betterindicatedbyhighervalues)usualcare
2
Alon
20076,
Alon
20087
RCTs
unblinded
Veryserious Noserious
limitations( inconsistency
a)
Noserious
indirectness
Noserious
imprecision
Alon2007:
42.3(16.6)
Alon2008:
10.5(12)
Alon
9.53(3.20,
2007:
15.87)
26.3(11)
Alon
2008:2.5
(4.9)
MD9.53higher Low
(3.2to15.87
higher)
BoxandBlocks(numberofblocksmovedin1minute)Postintervention(Betterindicatedbyhighervalues)shamintervention
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
27(4.8)
24.3(6.1) 2.70(
2.68,8.08)
MD2.70(2.68
lowerto8.08
higher)
Low
MD5.22lower
(9.66to0.78
lower)
Low
JebsenTaylorHandFunctiontest(timetomove5lightcans)(posttreatmenteffect)(Betterindicatedbylowervalues)usualcare
2
Alon
20076,
Alon
20087
RCTs
unblinded
Veryserious Noserious
limitations( inconsistency
a)
Noserious
indirectness
Noserious
imprecision
Alon2007:
6.7(2.9)
Alon2008:
40.5(22.8)
Alon
2007:
11.8(5.4)
Alon
2008:
52.9
NationalClinicalGuidelineCentre,2013.
412
5.22(
9.66,
0.78)
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
(17.3)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
JebsenTaylorHandFunctiontest(timetomovefivelightcans)(posttreatmenteffect)(Betterindicatedbylowervalues)shamintervention
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
10.8(2.1)
34.9
(25.8)
24.10(
42.04,
6.16)
MD24.10
lower(42.04to
6.16lower)
Modera
te
Alon
2007:
40.6(8.2)
Alon
2008:
14.5
(10.3)
8.85(3.35,
14.36)
MD8.85higher Very
(3.35to14.36
low
higher)
ModifiedFuglMeyerAssessment(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Alon
20076,
Alon
20087
RCTs
unblinded
Veryserious Serious
limitations( inconsistency
a)
Serious
indirectness(n)
Noserious
imprecision
Alon2007:
49(5.1)
Alon2008:
24.2(13.7)
FunctionalTestfortheHemiplegicUpperExtremity(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Chan
200942
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
3.7(0.5)
3.1(0.6)
0.60(0.12,
1.08)
MD0.6higher
(0.12to1.08
higher)
Modera
te
11.9
(12.4)
8.50(
1.30,
18.30)
MD8.5higher
(1.3lowerto
18.3higher)
Low
Forwardreachingdistance(cm)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Chan
200942
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
20.4(9.8)
Rangeofmotionwristextension(O)(posttreatmenteffect)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
413
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
2
Chan
200942,
Sahin
2012227
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
RCTdouble
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Chan:21
(28.5)
Sahin:25
(6.2)
Chan:6.5 1.57(
(18.9)
1.96,5.09)
Sahin:
23.8(5.6)
MD1.57higher
(1.96lowerto
LOW
5.09higher)
Serious
imprecision(f)
2.2(2)
2.2(2.1)
MD0.2higher
(1.6lowerto2
higher)
Chae:11.3
(3.0)
Chan:80.2
(6.8)
Sahin:109.8
(18.8)
Chae:
1.40(
10.6(5.9) 1.69,4.50)
Chan:
77.6(12)
Sahin:
102.7
(19.6)
MD1.40higher Modera
(1.69lowerto
te
4.50higher)
13.9(5.5)
13.6(6.5) 0.30(
4.16,4.76)
MD0.30higher Modera
(4.16lowerto
te
4.76higher)
15.8(5.8)
16.1(6.7) 0.30(
MD0.30lower
Grippower(kg)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Chan
200942
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
0.20(
1.60,2.00)
Low
FunctionalIndependenceMeasure(FIM)(postintervention)(Betterindicatedbyhighervalues)
3
Chae
199841,
Chan
200942,
Sahin
2012227
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
FunctionalIndependenceMeasure(FIM)(4weeksfollowup)(Betterindicatedbyhighervalues)
1Chae
199841
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
FunctionalIndependenceMeasure(FIM)(12weeksfollowup)(Betterindicatedbyhighervalues)
1Chae
RCTdouble
Serious
Noserious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
414
Modera
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
199841
Design
blinded
Riskofbias
limitations(
b)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
4.94,4.34)
Absoluteeffect
/Mean
Difference
(MD)(95%CI)
(4.94lowerto
4.34higher)
Confide
nce(in
effect)
te
ModifiedAshworthScaleofshoulder(posttreatmenteffect)(Betterindicatedbylowervalues)
1
Chan
200942
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
0.3(0.5)
0.5(0.6)
0.20(
0.68,0.28)
MD0.2lower
(0.68lowerto
0.28higher)
Low
1.3(0.8)
1.6(1.1)
0.30(
1.14,0.54)
MD0.30lower
(1.14lowerto
0.54higher)
Low
0.9(0.9)
1.4(1)
0.50(
1.33,0.33)
MD0.50lower Low
(1.33lowerto
0.33higher)
12.9(7.9)
8.9(2.3)
4.00(
1.70,9.70)
MD4higher
(1.7lowerto
9.7higher)
Low
1.9(0.82)
1.3(0.71) 0.60(
0.15,1.35)
MD0.6higher
(0.15lowerto
Low
ModifiedAshworthScaleelbowPostintervention(Betterindicatedbylowervalues)
1
Chan
200942
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
ModifiedAshworthScalewristPostintervention(Betterindicatedbylowervalues)
1
Chan
200942
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
Strengthoffingerextension(Newtons)Postintervention(Betterindicatedbyhighervalues)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
MotorActivityLogAmountofusescorePostintervention(Betterindicatedbyhighervalues)
1
Kimberl
RCTdouble
blinded
Serious
limitations(
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
NationalClinicalGuidelineCentre,2013.
415
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
ey
2004137
Design
Riskofbias
b)
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
1.35higher)
MotorActivityLogHowwellusedscorePostintervention(Betterindicatedbyhighervalues)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
2.1(0.84)
1.4(0.73) 0.70(
0.07,1.47)
MD0.7higher
(0.07lowerto
1.47higher)
Low
Serious
imprecision
0.6(1)
0.11
(0.29)
0.49(0.05,
0.93)
MD0.49higher Low
(0.05to0.93
higher)
Serious
imprecision
0.56(0.87)
0.11
(0.29)
0.45(0.07,
0.83)
MD0.45higher Low
(0.07to0.83
higher)
0.74(1.22)
0.12
(0.32)
0.62(0.09,
1.15)
MD0.62higher Low
(0.09to1.15
higher)
0.69(1.1)
0.12
(0.32)
0.57(0.09,
1.05)
MD0.57higher Low
(0.09to1.05
MotorActivityLog:Amountofuse:Lowdose(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
MotorActivityLog:Amountofuse:Highdose(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
MotorActivityLog:Qualityofmovement:Lowdose(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
MotorActivityLog:Qualityofmovement:Highdose(Betterindicatedbyhighervalues)
1
Hsu,
RCTSingle
blinded
Serious
limitations(
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
NationalClinicalGuidelineCentre,2013.
416
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
2010117
Design
Riskofbias
b)
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
higher)
JebsenTaylorpageturntest(s)Postintervention(Betterindicatedbylowervalues)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
17.1(5.7)
19.5(4.3) 2.40(
7.35,2.55)
MD2.4lower
(7.35lowerto
2.55higher)
Low
25(5.3)
41.4
(12.6)
16.40(
25.87,
6.93)
MD16.4lower
(6.93to25.87
lower)
Modera
te
6.7(2.52)
27.9(6.9) 21.20(
26.29,
16.11)
MD21.2lower
(16.11to26.29
lower)
Modera
te
25.3(7.6)
56.7
(26.6)
MD31.4lower
(12.23to50.57
lower)
Modera
te
JebsenTaylorsmallobjectstest(s)Postintervention(Betterindicatedbylowervalues)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
JebsenTaylorfeedingtest(s)Postintervention(Betterindicatedbylowervalues)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
JebsenTaylorstackingtest(s)Postintervention(Betterindicatedbylowervalues)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
JebsenTaylorheavycanstest(s)Postintervention(Betterindicatedbylowervalues)
NationalClinicalGuidelineCentre,2013.
417
31.40(
50.57,
12.23)
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
1
Kimberl
ey
2004137
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(e)
42.4(35)
30.8(21) 11.60(
16.68,
39.88)
MD11.6higher Very
(16.68lowerto low
39.88higher)
Noserious
indirectness
Serious
imprecision(c)
30(2.1)
6.2(16)
23.80(
42.09,
5.51)
MD23.80
lower(42.09to
5.51lower)
Low
9.6(6.3)
9.2(0.48,
17.92)
MD9.2higher
(0.48to17.92
higher)
Modera
te
0.2(0.1)
1.7(0.82,
2.58)
MD1.7higher
(0.82to2.58
higher)
Modera
te
Fingertrackingaccuracytest(posttreatmenteffect)
1
Kimberl
ey
2004137
RCTdouble
blinded
Serious
limitations(
b)
Noserious
inconsistency
UpperExtremityFunctionTestPostinterventionhighfunctiongroup(Betterindicatedbyhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
18.8(10.9)
UpperExtremityFunctionTestPostinterventionlowfunctiongroup(Betterindicatedbyhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
1.9(1.1)
UpperExtremityFunctionTestFollowup(26weeks)highfunctiongroup(Betterindicatedbyhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
29.9(9.7)
15.4(7.6) 14.5(5.96,
23.04)
UpperExtremityFunctionTestFollowup(26weeks)lowfunctiongroup(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
418
MD14.5higher Modera
(5.96to23.04
te
higher)
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
1
Popovic
2003210
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
4.9(3.1)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
1.5(0.9)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
3.4(0.82,
5.98)
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
MD3.4higher
(0.82to5.98
higher)
Modera
te
Drawingability(%areacomparedwithtargetsquare)Postinterventionhighfunctiongroup(rangeofscores:0100;Betterindicatedbyhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
83.8(5.6)
68.7
(11.7)
15.10
(6.11,
24.09)
MD15.1higher Low
(6.11to24.09
higher)
Drawingtest(%areacomparedwithtargetsquare)(higherfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedbyhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
39.9(8.1)
25.9(8.6) 14.00
(5.81,
22.19)
MD15.1higher Low
(6.11to24.09
higher)
Drawingability(%areacomparedwithtargetsquare)Postinterventionlowfunctiongroup(rangeofscores:0100;Betterindicatedbyhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
24(9.9)
16(2.8)
8.0(0.23,
16.23)
MD8.0higher
(0.23lowerto
16.23higher)
Low
Drawingability(%areacomparedwithtargetsquare)Followup(26weeks)highfunctiongroup(followupmean26weeks;rangeofscores:0100;Betterindicated
byhighervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c
49.8(9.6)
36.7(6.9) 13.10
(3.64,
22.56)
MD13.1higher Low
(3.64to22.56
higher)
ModifiedAshworthScale(higherfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedbylowervalues)
1
RCTsingle
Serious
Noserious
Noserious
Noserious
1.25(0.5)
2.25
NationalClinicalGuidelineCentre,2013.
419
1.00(
MD1lower
Modera
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Popovic
2003210
Design
blinded
Riskofbias
limitations(
b)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
(0.75)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
1.62,
0.38)
Absoluteeffect
/Mean
Difference
(MD)(95%CI)
(0.38to1.62
lower)
Confide
nce(in
effect)
te
ModifiedAshworthScale(lowerfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedbylowervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(f)
2.5(0.75)
2.25
(0.75)
0.25(
0.60,1.10)
MD0.25higher Low
(0.6lowerto
1.1higher)
ReducedUpperExtremityMotorActivityLogquestionnaireAmountscale(higherfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedbyhigher
values)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
59.7(2.5)
28.7
(11.7)
31.00
(19.14,
42.86)
MD31higher
(19.14to42.86
higher)
Low
ReducedUpperExtremityMotorActivityLogquestionnaireAmountscale(lowerfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedbyhigher
values)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
16.7(8.3)
3.3(1.7)
13.40
(6.62,
20.18)
MD
13.40
higher
(6.62to
20.18
higher)
Low
ReducedUpperExtremityMotorActivityLogquestionnaireHowwellscale(higherfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedby
highervalues)
1
Popovic
RCTsingle
blinded
Serious
limitations(
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
66.7(11.4)
32.5
(10.6)
NationalClinicalGuidelineCentre,2013.
420
34.20
(23.41,
MD
34.20
Low
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
2003210
Design
Riskofbias
b)
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
44.99)
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
higher
(23.41to
44.99
higher)
ReducedUpperExtremityMotorActivityLogquestionnaireHowwellscale(lowerfunctioninggroupversususualcare)(26weeksfollowup)(Betterindicatedby
highervalues)
1
Popovic
2003210
RCTsingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(c)
11.5(6.1)
2.3(1.2)
9.20(4.23,
14.17)
MD9.20
higher
(4.23to
14.17
higher)
Low
5.10(
0.52,
10.72)
MD5.1
higher
(0.52
lowerto
10.72
higher)
Low
5.00[
1.39,
11.39]
MD5.0
higher
(1.39
lowerto
11.39
Low
ChangeinActionResearchArmTest(totalscore)(4weeksfollowup:LowdoseFES)(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(h)
8.6(11.3)
3.5(8.2)
ChangeinActionResearchArmTest(totalscore)(4weeksfollowup:HighdoseFES)(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(h)
8.5(13.2)
3.5(8.2)
NationalClinicalGuidelineCentre,2013.
421
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
higher)
ChangeinActionResearchArmTest(totalscore)(12weeksfollowup:LowdoseFES)(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(h)
17.2(19.1)
4(9.5)
13.20
(5.19,
21.21)
MD5.1
higher
(0.52
lowerto
10.72
higher)
Low
Hsu:4
(9.5)
Mann:
24.4
10.68
(5.68,
15.68)
MD10
higher
(3.77to
16.23
higher)
Low
34.4
24.7
9.70(2.35,
17.05)
MD9.70
higher
(2.35to
17.05
higher)
Verylow
3(1,3.3)
1.5(0.8,
2)
P=0.19(j)
(i)
Low(i)
ChangeinActionResearchArmTest(totalscore)(12weeksfollowup)(Betterindicatedbyhighervalues)
2
RCT
Hsu,
2010117;
Mann
2005168
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(h)
Hsu(high
dose):15.9
(18.4)
Mann:34.4
ChangeinActionResearchArmTest(totalscore)(24weeksfollowup)(Betterindicatedbyhighervalues)
1Mann
2005168
RCT
Veryserious Noserious
limitations( inconsistency
g)
Noserious
indirectness
Serious
imprecision(h)
ActivitiesofDailyLiving(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Mangold
Veryserious Noserious
limitations( inconsistency
Noserious
indirectness
(i)
NationalClinicalGuidelineCentre,2013.
422
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
2009167
Design
Riskofbias
g)
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
0(0,1)
P=0.57(j)
(i)
Low(i)
0.3(0,
0.5)
P=1.0(j)
(i)
Low(i)
0.5(0,
1.1)
P=0.17(j)
(i)
Low(i)
0.5(0,1.3)
0.5(0,
1.1)
P=0.68(j)
(i)
Low(i)
2(0,3)
1(0,4)
(i)
(i)
Moderate(i)
GaininChedokeMcMasterStrokeAssessment(arm)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Mangold
2009167
Veryserious Noserious
limitations( inconsistency
g)
Noserious
indirectness
(i)
1(0,1)
GaininChedokeMcMasterStrokeAssessment(hand)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Mangold
2009167
Veryserious Noserious
limitations( inconsistency
g)
Noserious
indirectness
(i)
0(0,1)
GaininModifiedAshworthScale(fingerflexors)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Mangold
2009167
Veryserious Noserious
limitations( inconsistency
g)
Noserious
indirectness
(i)
0(0.5,0.8)
GaininModifiedAshworthScale(wristflexors)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Mangold
2009167
Veryserious Noserious
limitations( inconsistency
g)
Noserious
indirectness
(i)
Changeingripstrength(kg)(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
(i)
Changeingripstrength(kg)(32weeksfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
423
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
(i)
2(0,8)
4(0,10)
(i)
(i)
Moderate(i)
(i)
0(0,1)
0(0,1)
(i)
(i)
Moderate(i)
(i)
1(0,1.5)
1(0,1)
(i)
(i)
Moderate(i)
(i)
10(0,29)
2(0,14) (i)
(i)
Moderate(i)
(i)
6(0,31)
1(0,16) (i)
(i)
Moderate(i)
(i)
0(0,13)
0(0,
0.8)
(j)
Moderate(i)
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
Ashworthscore(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Asworthscore(32weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
ARAT(totalscore)(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
ARAT(totalscore)(32weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Noofpegspersecond(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
424
(i)
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
(i)
0(0,0.16)
(i)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
0(0,
0.11)
(i)
(i)
Moderate(i)
5(3,7)
4(1,6)
(i)
(i)
Moderate(i)
(i)
7(5,10)
4(2,9)
(i)
(i)
Moderate(i)
(i)
1(1,0)
1(1,0)
(i)
(i)
Moderate(i)
(i)
1(2,0)
1(1,0)
(i)
(i)
Moderate(i)
Chae:13.1
Chae:
5.72(2.79,
MD5.72
Moderate
Noofpegspersecond(32weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Rankinscale(8weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
Rankinscalescore(32weeksfollowup)(Betterindicatedbyhighervalues)
1Powell
1999213
RCTsingle
blinded
Serious
limitations
(k)
Noserious
inconsistency
Noserious
indirectness
FuglMeyerAssessmentUpperLimb(posttreatmenteffect)(Betterindicatedbyhighervalues)
3
RCTSingle
Serious
Noserious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
425
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Chae
199841,
Chan
200942;
Lin,
2011155
Design
blinded
Riskofbias
limitations(
b)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
(10.3)
Chan:25.9
(8.9)
Lin:20.3
(5.4)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
6.5(6.1)
Chan:
22.1
(9.9)
Lin:14.5
(5.8)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
8.65)
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
higher
(2.79to
8.65
higher)
FuglMeyerAssessmentUpperLimb(total)(1weekfollowup)(Betterindicatedbyhighervalues)
4
RCTSingle
blinded
Chae
41
1998 ;C
han
200942;
Lin,
2011155;
Shindo
20011237
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Chae:13.1
(10.3)
Chan::25.9
(8.9)
Lin:20.3
(5.4)
Shindo:12.2
(5.3)
Chae:
9.7(7.7)
Chan:
22.1
(9.9)
Lin:14.5
(5.8)
Shindo:
5.5(6)
5.98(3.45,
8.50)
MD5.98
higher
(3.45to
8.50
higher)
Moderate
Chae:17.8
(12.6)
Hsu(high
dose):25.5
(20)
Lin:22.6
Chae:
9.7(7.7)
Hsu:
14.2
(14.5)
Lin:17.7
6.11(2.85,
9.38)
MD6.11
higher
(2.85to
9.38
higher)
Moderate
FuglMeyerAssessmentUpperLimb(total)(12weeks)(Betterindicatedbyhighervalues)
3
RCTSingle
blinded
Chae
199841;
Hsu,
2010117;
Lin,
2011155
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
NationalClinicalGuidelineCentre,2013.
426
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
(5.7)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
(6.2)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
FuglMeyerAssessmentUpperLimb(total)(1monthfollowup:LowdoseFES)(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(d)
Hsu:28.1
(18)
Hsu:
14.2
(14.5)
13.90
(4.24,
23.56)
MD
13.90
higher
(4.24to
23.56
higher)
Low
Chae:
11.2
(8.7)
Hsu:17
(15.4)
Lin:18.5
(6.7)
8.45(5.05,
11.85)
MD8.45
higher
(5.05to
11.85
higher)
Moderate
19.70
(9.32,
30.08)
MD19.7
higher
(9.32to
30.08
higher)
Low
FuglMeyerAssessmentUpperLimb(total)(3monthsfollowup)(Betterindicatedbyhighervalues)
3
RCTSingle
blinded
Chae
41
1998 ;
Hsu,
2010117;
Lin,
2011155
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Chae:20.6
(15.1)
Hsu(high
dose):32.8
(23.7)
Lin:26.0
(5.1)
FuglMeyerAssessmentUpperLimb(total)(3monthsfollowup:LowdoseFES)(Betterindicatedbyhighervalues)
1
Hsu,
2010117
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(d)
Hsu:36.7
(19.5)
Hsu:17
(15.4)
FuglMeyerAssessmentUpperLimb(total)(6monthsfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
427
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(d)
29.8(3.6)
20.3
(12.3)
9.5(3.59
to15.41)
MD9.5
higher
(3.59to
15.41
higher)
Low
Serious
imprecision(f)
1.16(0.50)
0.78
(0.55)
0.38(0.04
to0.72)
MD0.38
higher
(0.04to
0.72
higher)
Low
Serious
imprecision(f)
1.42(0.51)
1.11
(0.32)
0.31(0.04
to0.58)
MD0.31
higher
(0.04to
0.58
higher)
Low
1.56(0.53)
1.50
(0.53)
0.06(0.28
to0.40)
MD0.06
higher(
0.28
lowerto
0.40
higher)
Verylow
ModifiedAshworthScale(3Weeksfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
ModifiedAshworthScale(1monthfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
ModifiedAshworthScale(3monthsfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(n)
NationalClinicalGuidelineCentre,2013.
428
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
ModifiedAshworthScale(6monthsfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(f)
1.67(0.52)
1.86
(0.38)
0.19(
0.48to
0.1)
MD0.19
lower
(0.48
lowerto
0.1
higher)
Low
Serious
imprecision(l)
57.0(10.7)
49.7
(11.4)
7.3(0.17
to14.43)
MD7.3
higher
(0.17to
14.43
higher)
Low
Serious
imprecision(l)
64.5(10.4)
55.7
(12.1)
8.80(1.51
to16.09)
MD8.8
higher
(1.51to
16.09
higher)
Low
Noserious
72.4(8.5)
59.3
13.10(6.37
MD13.1
Moderate
ModifiedBarthelIndex(3weeksfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
ModifiedBarthelIndex(1monthfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
ModifiedBarthelIndex(3monthsfollowup)(Betterindicatedbyhighervalues)
Lin,
RCTSingle
Serious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
429
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
2011155
Design
blinded
Riskofbias
limitations(
b)
Inconsistency
inconsistency
Indirectness
indirectness
Imprecision
imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
(12.0)
Effect
79.2(5.2)
66.1
(11.3)
13.1(7.38
to18.82)
MD13.1
higher
(7.38to
18.82
higher)
Moderate
3.6(3.0)
3.5(2.9)
0.10(
1.68,1.88)
MD0.10
higher
(1.68
lowerto
1.88
higher)
Low
0.27(0.15)
0.25
(0.19)
0.02(
0.08,0.12)
MD0.10
higher
(1.68
lowerto
1.88
higher)
High
Relative
Risk
Mean
difference
(95%CI)/P
values
to19.83)
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
higher
(6.37to
19.83
higher)
ModifiedBarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
Lin,
2011155
RCTSingle
blinded
Serious
limitations(
b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
ElectrophysiologicalevaluationFmax/Mmax(%)Postintervention(Betterindicatedbyhigher)
Sahin
2012227
RCTDouble Noserious
blinded
limitation
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(o)
ElectrophysiologicalevaluationHmax/Mmax(%)Postintervention(Betterindicatedbyhigher)
Sahin
2012227
RCTDouble Noserious
blinded
limitation
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
NationalClinicalGuidelineCentre,2013.
430
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies
Design
Riskofbias
Inconsistency
Indirectness
Imprecision
ES
Number
ofevent/
TotalN
Mean
(SD)/
Median
(IQR)
Usualcare
Numberof
event/
TotalN
Mean
(SD)/Medi
an(IQR)
Effect
Relative
Risk
Mean
difference
(95%CI)/P
values
Absoluteeffect
/Mean
Confide
Difference
nce(in
(MD)(95%CI) effect)
WristspasticityModifiedAshworthScalePostintervention(Betterindicatedbyhigher)
Sahin
2012227
RCTDouble Noserious
blinded
limitation
Noserious
inconsistency
Noserious
indirectness
(i)
1.8
(i)
(i)
High
(i)
4.5
(i)
(i)
High
Brunnstrommotorscale(upper)Postintervention(Betterindicatedbyhigher)
Sahin
2012227
RCTDouble Noserious
blinded
limitation
Noserious
inconsistency
Noserious
indirectness
(a)Bothstudieswereunblindedwithunclearrandomizationandallocationconcealment.
(b)Unclearrandomizationandallocationconcealment.
(c)ConfidenceintervalcrossedoneendofdefaultMID.
(d)MeandifferencedidnotreachtheagreedMID(10%differenceinthescale).
(e)ConfidenceintervalcrossedbothendsofdefaultMID.
(f)MeandifferencedidnotreachthedefaultMID.
(g)Unclearblinding,randomizationandallocationconcealment.
(h)MeandifferencedidnotreachtheagreedMIDof12points.
(i)Imprecisioncouldnotbeassessedasauthorsreportedonlymedian(IQR).Resultscouldnotbemetaanalysedandrelative/absoluteeffectcouldnotbeestimated.
(j)Pvalueasreportedbyauthors.
(k)Inadequateallocationconcealment.
(l)MeandifferencedidnotreachtheagreedMID(9.25).
(m)ConfidenceintervalcrossedbothendsofthedefaultMID.
(n)Itemsoftheoriginalscalenotincluded.
(o)ConfidenceintervalcrossesbothendsofdefaultMID
Narrativesummaries
ThefollowingstudiesaresummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
NationalClinicalGuidelineCentre,2013.
431
StrokeRehabilitation
Movement
Cauraughetal,200038foundthattheexperimentalgroupwhoreceivedtheFEStrainingmovedsignificantlymoreblocksanddisplayedahigherisometric
forceimpulseaftertherehabilitationtreatmentcomparedtousualcaregroup.NeitherMotorAssessmentScalenorFuglMeyertestsweresignificantly
differentbetweenthetwogroups.
Cauraughetal,200239foundsignificantfindingsfavouringthecoupledbilateralmovementtrainingandEMGtriggeredneuromuscularstimulationgroup.
Inaddition,theunilateralmovement/stimulationgroupexceededthecontrolacrossthecategoriesoftasks.
Thrasheretal,2008257foundthattheFESgroupimprovedsignificantlymorethanthecontrolgroupintermsofobjectmanipulation,palmergriptorque,
andpinchgrippullingforce,BarthelIndex,UpperExtremityFuglMeyerscoresandUpperExtremityChedokeMcMasterstagesofMotorRecovery.
Haraetal,2008104reportedthattheFESgroupdisplayedsignificantlygreaterimprovementsintheactiveRangeofMovementofwristandfinger
extensionandshoulderflexion,modifiedAshworthscale(MAS)andfunctionalhandtestsandwasabletosmoothlyperformactivitiesofdailylifeusing
thehemiplegicupperextremities.
NationalClinicalGuidelineCentre,2013.
432
StrokeRehabilitation
Movement
13.4.1.2
Economicevidence
Literaturereview
NorelevanteconomicevaluationscomparingESwithusualcarewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
TypicalcostsforFESwereobtainedfromOdstockMedicalLimitedatSalisburyDistrictHospital(by
email,2021stDecember2010)whosupplytheFESsystemdescribedintheRCTreportedbyMannet
al(2005)168includedintheclinicalreview(theMicrostim2[MS2v2],aselfcontainedtwochannel
exercisestimulator).ThecostoftheMS2v2kitis267(excludingVAT).Thedeviceisguaranteedfor
2yearsandsparepartsandservicemaintenanceareofferedforaminimumof5years.The
electrodesaresinglepatientuseandlastaroundfourweeks.Electrodescostbetween6and10
perpackoffour(excludingVAT)dependingonsizeandquality.Thedevicewillrunonstandardor
rechargeablePP3batteries(suppliedinkit).Thecostofastandard6monthtreatmentpackageusing
theMS2v2systemconsistingofoneinitialassessmentandfivetreatmentsessionsischargedat
840;eachsessionis140.Thisincludesthecostofallequipment,consumables,physiotherapyand
hospitaloverheadsandisdeliveredasanoutpatientservice.PatientscanalsousetheMS2v2dailyin
theirownhomes.Basedonthestandardtreatmentpackagecost,forFEStobejudgedcosteffective
itwouldneedtoprovidebenefitstopatientsthattranslatedtoatleastanadditional0.042QALYsper
person.
13.4.1.3
Evidencestatements
Clinicalevidencestatements
Twostudies6,7comprisingof41participantsfoundthatparticipantswhoreceivedtheElectrical
StimulationexperiencedastatisticallysignificantimprovementintheBoxandBlocstestattheendof
thetrialcomparedtoparticipantswhoreceivedusualcare(LOWCONFIDENCEINEFFECT).
Onestudy137comprisingof16participantsfoundnosignificantdifferenceintheBoxandBlockstest
attheendofthetrialbetweenparticipantswhoreceivedtheElectricalStimulationandthosewho
receivedshamtreatment(LOWCONFIDENCEINEFFECT).
Twostudies6,7comprisingof41participantsfoundthatparticipantswhoreceivedtheElectrical
StimulationexperiencedastatisticallysignificantimprovementintheJebsenTaylorHandFunction
test(lightcans)attheendofthetrialcomparedtoparticipantswhoreceivedusualcare(LOW
CONFIDENCEINEFFECT).
Onestudy137comprisingof16participantsfoundastatisticallysignificantimprovementinthe
JebsenTaylorHandFunctiontest(lightcans)attheendofthetrialforparticipantswhoreceivedthe
ElectricalStimulationcomparedtothosewhoreceivedshamtreatment(MODERATECONFIDENCEIN
EFFECT).
Twostudies6,7comprisingof41participantsfoundthatparticipantswhoreceivedtheElectrical
StimulationexperiencedastatisticallysignificantimprovementintheModifiedFuglMeyer
NationalClinicalGuidelineCentre,2013.
433
StrokeRehabilitation
Movement
Assessmentattheendofthetrialcomparedtoparticipantswhoreceivedusualcare(VERYLOW
CONFIDENCEINEFFECT).
Onestudy42comprisingof20participantsfoundthatparticipantsreceivedtheElectricalStimulation
hadsignificantlyhigherscoresintheFunctionalTestfortheHemiplegicUpperExtremityattheend
ofthetrialcomparedtotheusualcaregroup(MODERATECONFIDENCEINEFFECT).
Onestudy42comprisingof20participantsfoundnosignificantdifferenceonthefollowingoutcomes
betweentheElectricalStimulationandtheusualcaregroupsattheendofthetrial:
forwardreachdistance(cm)(LOWCONFIDENCEINEFFECT),
activerangeofmotioninwristextension(VERYLOWCONFIDENCEINEFFECT),
grippower(kg)(LOWCONFIDENCEINEFFECT),
FunctionalIndependenceMeasure(LOWCONFIDENCEINEFFECT),
ModifiedAshworthScaleofshoulder(LOWCONFIDENCEINEFFECT),
ModifiedAshworthScaleofelbow(LOWCONFIDENCEINEFFECT),
ModifiedAshworthScaleofwrist(LOWCONFIDENCEINEFFECT)
Twostudies42,227comprising62participantsfoundnosignificantdifferencewiththerangeofmotion
inwristextensionbetweentheElectricalStimulationandtheusualcaregroupsattheendofthe
intervention(HIGHCONFIDENCEINEFFECT)
Threestudies41;42;227comprisingof90participantsfoundnosignificantdifferenceintheFunctional
IndependenceMeasurebetweentheElectricalStimulationgroupandtheusualcaregrouppost
treatment(MODERATECONFIDENCEINEFFECT)
Onestudy41comprisingof28participantsfoundnosignificantdifferenceintheFunctional
IndependenceMeasurebetweentheElectricalStimulationgroupandtheusualcaregroupat4and
12weeksfollowup(MODERATECONFIDENCEINEFFECT)
Onestudy137comprisingof16participantsfoundnosignificantdifferenceonthefollowingoutcomes
betweentheElectricalStimulationandtheusualcaregroupsattheendofthetrial:
Strengthoffingerextension(LOWCONFIDENCEINEFFECT),
MotorActivityLog;amountofusescore(LOWCONFIDENCEINEFFECT),
MotorActivityLog;howwellusedscore(LOWCONFIDENCEINEFFECT),
JebsenTaylorHandFunctiontest(pageturn)(LOWCONFIDENCEINEFFECT),
JebsenTaylorHandFunctiontest(heavyscans)(VERYLOWCONFIDENCEINEFFECT).
Onestudy117comprisingof66participantsfoundstaticallysignificantimprovementinthefollowing
outcomesbetweentheElectricalStimulationandtheusualcaregroups
MotorActivityLog:amountofusescorelowdose(LOWCONFIDENCEINEFFECT),
MotorActivityLog:amountofusescorehighdose(LOWCONFIDENCEINEFFECT),
MotorActivityLog:qualityofmovementlowdose(LOWCONFIDENCEINEFFECT),
MotorActivityLog:qualityofmovementhighdose(LOWCONFIDENCEINEFFECT)
Onestudy137comprisingof16participantsfoundastatisticallysignificantimprovementinthe
followingoutcomesattheendofthetrialforparticipantswhoreceivedtheElectricalStimulation
comparedtothosewhoreceivedshamtreatment:
JebsenTaylorHandFunctiontest(smallobjects)(MODERATECONFIDENCEINEFFECT),
JebsenTaylorHandFunctiontest(feeding)(MODERATECONFIDENCEINEFFECT),
JebsenTaylorHandFunctiontest(stacking)(MODERATECONFIDENCEINEFFECT),
NationalClinicalGuidelineCentre,2013.
434
StrokeRehabilitation
Movement
Fingertrackingaccuracytest(LOWCONFIDENCEINEFFECT).
Onestudy210comprisingof16participantsfoundastatisticallysignificantimprovementinthe
followingoutcomesforthehigherfunctioningparticipantswhoreceivedtheElectricalStimulation
comparedtothosewhoreceivedusualcare:
UpperExtremityFunctionTest(attheendofthetrialandat26weeksfollowup)(MODERATE
CONFIDENCEINEFFECT),
Drawingtest(attheendofthetrialandat26weeksfollowup)(LOWCONFIDENCEINEFFECT),
Ashworthgrade(at26weeksfollowup)(MODERATECONFIDENCEINEFFECT),
ReducedUpperExtremityMotorActivityLogQuestionnaireamountscale(at26weeksfollowup)
(LOWCONFIDENCEINEFFECT),
ReducedUpperExtremityMotorActivityLogQuestionnairehowwellscale(at26weeksfollow
up)(LOWCONFIDENCEINEFFECT).
Onestudy210comprisingof12participantsfoundastatisticallysignificantimprovementinthe
followingoutcomesforthelowerfunctioningparticipantswhoreceivedtheElectricalStimulation
comparedtothosewhoreceivedusualcare:
UpperExtremityFunctionTest(attheendofthetrialandat26weeksfollowup)(MODERATE
CONFIDENCEINEFFECT),
Drawingtest(at26weeksfollowup)(LOWCONFIDENCEINEFFECT),
ReducedUpperExtremityMotorActivityLogQuestionnaireamountscale(at26weeksfollowup)
(LOWCONFIDENCEINEFFECT),
ReducedUpperExtremityMotorActivityLogQuestionnairehowwellscale(at26weeksfollow
up)(LOWCONFIDENCEINEFFECT).
Onestudy117comprising66participantsfoundnodifferenceinthechangescoresofActionResearch
ArmTest(totalscore)betweenthegroupthatreceivedlowdoseESandtheusualcaregroupat4
weeksfollowup(LOWCONFIDENCEINEFFECT).
Onestudy117comprising66participantsfoundnodifferenceinthechangescoresofActionResearch
ArmTest(totalscore)betweenthegroupthatreceivedhighdoseESandtheusualcaregroupat4
weeksfollowup(LOWCONFIDENCEINEFFECT).
Onestudy117comprising66participantsfoundastatisticallysignificantimprovementinthechange
scoresofActionResearchArmTest(totalscore)betweenthegroupthatreceivedlowdoseESand
theusualcaregroupat12weeksfollowup(LOWCONFIDENCEINEFFECT).
Twostudies168;117comprisingof88participantsfoundastatisticallysignificantimprovementinthe
changescoresofActionResearchArmTest(totalscore)at12weeksfollowupfortheparticipants
whoreceivedtheElectricalStimulationcomparedtothosewhoreceivedusualcare(LOW
CONFIDENCEINEFFECT).
Onestudy168comprisingof22participantsfoundastatisticallysignificantimprovementinthechange
scoresofActionResearchArmTest(totalscore)at24weeksfollowupfortheparticipantswho
receivedtheElectricalStimulationcomparedtothosewhoreceivedusualcare(VERYLOW
CONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
435
StrokeRehabilitation
Movement
Threestudies41,42;155comprisingof85participantsfoundthatparticipantswhoreceivedthe
ElectricalStimulationexperiencedastatisticallysignificantimprovement(posttreatment)intheFugl
MeyerAssessmentcomparedtoparticipantswhoreceivedusualcare.Thisdifferencewasofclinical
importance(MODERATECONFIDENCEINEFFECT).
Threestudies41;117;155comprisingof109participantsfoundthatparticipantswhoreceivedthe
ElectricalStimulationexperiencedastatisticallysignificantimprovementintheFuglMeyer
Assessmentcomparedtoparticipantswhoreceivedusualcareatonemonthfollowup.This
differencewasofclinicalimportance(MODERATECONFIDENCEINEFFECT).
Onestudy117comprisingof66participantsfoundthatparticipantswhoreceivedlowdoseElectrical
StimulationexperiencedastatisticallysignificantimprovementintheFuglMeyerAssessment
comparedtoparticipantswhoreceivedusualcareatonemonthfollowup.Thisdifferencewasnotof
clinicalimportance(LOWCONFIDENCEINEFFECT).
Threestudies41;117;155comprisingof109participantsfoundthatparticipantswhoreceivedthe
ElectricalStimulationexperiencedastatisticallysignificantimprovementintheFuglMeyer
Assessmentcomparedtoparticipantswhoreceivedusualcareat3monthsfollowup.Thisdifference
wasofclinicalimportance(MODERATECONFIDENCEINEFFECT).
Onestudy117comprisingof66participantsfoundthatparticipantswhoreceivedlowdoseElectrical
StimulationexperiencedastatisticallysignificantimprovementintheFuglMeyerAssessment
comparedtoparticipantswhoreceivedusualcareat3monthsfollowup.Thisdifferencewasnotof
clinicalimportance(LOWCONFIDENCEINEFFECT).
Onestudy155comprisingof46participantsfoundthatparticipantswhoreceivedtheElectrical
StimulationexperiencedastatisticallysignificantimprovementintheFuglMeyerAssessmentat6
monthsfollowupcomparedtoparticipantswhoreceivedusualcare.Thisdifferencewasnotof
clinicalimportance(LOWCONFIDENCEINEFFECT).
Onestudy155comprisingof46participantsfoundthatparticipantswhoreceivedtheElectrical
Stimulationexperiencedastatisticallysignificantimprovementposttreatmentand1monthfollow
upwiththemodifiedAshworthscalecomparedtoparticipantswhoreceivedusualcare(LOW
CONFIDENCEINEFFECT).
Onestudy155comprisingof46participantsfoundnosignificantimprovementwiththemodified
Ashworthscaleat3monthsfollowupbetweentheElectricalStimulationandtheusualcaregroups
(VERYLOWCONFIDENCEINEFFECT).
Onestudy155comprisingof46participantsfoundnosignificantimprovementwiththemodified
Ashworthscaleat6monthsfollowupbetweentheElectricalStimulationandtheusualcaregroups
(LOWCONFIDENCEINEFFECT).
Onestudy155comprisingof46participantsfoundthatparticipantswhoreceivedtheElectrical
StimulationexperiencedastatisticallysignificantimprovementinthemodifiedBarthelIndexatpost
treatmentand1monthfollowupcomparedtoparticipantswhoreceivedusualcare.Thisdifference
wasnotofclinicalimportance(LOWCONFIDENCEINEFFECT).
Onestudy155comprisingof46participantsfoundthatparticipantswhoreceivedtheElectrical
StimulationexperiencedastatisticallysignificantimprovementinthemodifiedBarthelIndexat3and
6monthsfollowupcomparedtoparticipantswhoreceivedusualcare(MODERATECONFIDENCEIN
EFFECT).
Onestudy227comprising42participantsfoundnosignificantdifferenceinelectrophysiological
evaluation(Fmax/Mmax)betweentheElectricalStimulationandtheusualcaregroup(LOW
CONFIDENCEINEFFECT)
NationalClinicalGuidelineCentre,2013.
436
StrokeRehabilitation
Movement
Onestudy227comprising42participantsfoundnosignificantdifferenceinelectrophysiological
evaluation(Hmax/Mmax)betweentheElectricalStimulationandtheusualcaregroup(HIGH
CONFIDENCEINEFFECT)
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.4.2
Recommendationsandlinktoevidence
90.Donotroutinelyofferpeoplewithstrokeelectrical
stimulationfortheirhandandarm.
91.Consideratrialofelectricalstimulationinpeoplewhohave
evidenceofmusclecontractionafterstrokebutcannotmove
theirarmagainstresistance.
92.Ifatrialoftreatmentisconsideredappropriate,ensurethat
electricalstimulationtherapyisguidedbyaqualified
rehabilitationprofessional.
93.Theaimofelectricalstimulationshouldbetoimprove
strengthwhilepractisingfunctionaltasksinthecontextofa
comprehensivestrokerehabilitationprogramme.
94.Continueelectricalstimulationifprogresstowardsclear
functionalgoalshasbeendemonstrated(forexample,
maintainingrangeofmovement,orimprovinggraspand
release).
Relativevaluesofdifferent
outcomes
AwiderangeofmeasureswereusedinthesestudiesandtheGDGnoted
thattherewasnopsychometricallyrobustpatientreportedoutcome
measureusedforassessmentofreducedupperlimbfunction.
Thewiderangeofmeasuresreportedwithinthetrialsreviewedmakes
interpretingthedatadifficult.
Tradeoffbetweenclinical
benefitsandharms
TherearefewrisksassociatedwithElectricalStimulation.The
commonestisaskinreactionwhenselfadhesiveelectrodesareused.
BenefitsarisefromtheincreasedrangeofmovementsproducedbyES
withtheassociatedincreasedeaseofperformanceoffunctionaltask.ES
wastypicallytargetedatfingerandwristextensorsbutwasalsousedfor
elbowextensionandshoulderflexion.TheGDGwerealsoawareofthe
useofelectricalstimulationformanagementofspasticityhoweverthis
wasnotincludedinreview.
Economicconsiderations
Nocosteffectivenessstudieswerefoundforthisquestion.
ESforhandfunctionsarenotroutinelyusedintheUKNHScurrently.
AtypicalcostperpatientofdeliveringESwasestimatedtobearound
840(oneinitialassessmentfollowedbyfivesessionsinhospital).Based
onthesecosts,forEStobejudgedcosteffectiveitwouldneedto
providebenefitstopatientsthattranslatedtoatleastanadditional
0.042QALYsperperson.TheGDGconsideredthisadditionalbenefit
achievableinaselectedpopulationforwhomthetreatmentis
consideredappropriate(forexamplepeoplewhohaveevidenceof
NationalClinicalGuidelineCentre,2013.
437
StrokeRehabilitation
Movement
musclecontractionafterstrokebutcannotmovetheirarmagainst
resistance).
Qualityofevidence
Themajorityofthestudiesreportedbenefitbutthiswasnotalways
significant.TheGDGconsideredthattheresultsshouldbeinterpreted
withcautionduetothesmallsamplesizeofthestudies.Veryfewstudies
reportedfollowupresultsbutPopovic210,Mann168,Powell213,Lin155did
reportstatisticalsignificanceinfavourofESbetween1andsixmonths
followupforarangeofoutcomesincludingFuglMeyerAssessmentand
modifiedBarthelIndex.
Thestudiescouldbedividedintothosethatlookedatearlyafterstroke,
lateafterstrokeandthosethatincorporatedphysiotherapyguided
functionalexerciseandthosethatdidnot.TheGDGobservedthatthe
Kimberleystudy137wastheonlyonethatdidnothavephysiotherapyas
thecomparatoranditwasnotedthatgenerallythepatientsincludedin
thestudiestendedtobetheyoungeragegroup(between4570years
old).
TheGDGconsideredthatwhenusedearlyafterstrokeinhigh
functioningpeoplethereappearedtobelimitedevidenceofbenefit6,210
butalargerstudyisneeded.
TheGDGalsonotedthatthestudiesbyAlon6,7usedamodifiedFugl
MeyerAssessmentoutcomeandthereforetheresultsshownwould
needtoberegardedwithcaution.
Otherconsiderations
TheGDGnotedthatthestudybyAlon6,7waspartiallysponsoredbythe
manufacturersofthedeviceandthattheresultswereconsistentwith
otherstudieswhichwerepubliclyorcharityfunded.Electrical
stimulationisnotwidelyavailable,andifatrialoftreatmentisofferedto
apatientitshouldonlybedeliveredbyahealthprofessionalwiththe
appropriateskillset.TheGDGagreedthatanassessmentofthosewho
maybenefitfromtheinterventionshouldbecarriedoutandatrialof
useconductedtoestablishifanimprovementinrangeofmovementor
functionofthehandorwristisclearlydemonstrated.
13.5 Constraintinducedmovementtherapy
Constraintinducedmovementtherapyisanapproachtopromoteincreasedactivityintheimpaired
upperlimbinpatientsafterstroke.Inordertoovercomelearnednonuseintheaffectedlimbthe
unaffectedlimbisrestrainedusuallybyahandmittenorarmslingforlongperiodsoftheday,
therebypromotingtheuseoftheaffectedlimbineverydaysituations.Inadditiontotherestraint,
treatmentincludesperiodsofintensivefocusedexerciseoractivityusuallyundertheguidanceofa
therapist.Becauseofthenatureoftheinterventionconstraintinducedmovementtherapyisnot
suitablefor,oracceptableto,allpatientsafterstroke.
13.5.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
constraintinducedtherapyversususualcareonimprovingfunctionandreducing
disability?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
Constraintinducedmovementtherapy(CIMT)forupperlimb
Subgroupanalysis
Lessthan5hours
Morethan5hours
Anyconstraintegslings
Comparison
Usualcare
NationalClinicalGuidelineCentre,2013.
438
StrokeRehabilitation
Movement
ClinicalMethodologicalIntroduction
Outcomes
FunctionalIndependenceMeasure(FIM)
BarthelIndex
FuglMeyerAssessment
ActionResearchArmTest(ARAT)
WolfMotorFunctionTest(WMFT)
9holepegtest
Anyadverseevent
13.5.1.1
Clinicalevidencereview
SearcheswereconductedforsystematicreviewsandRCTscomparingConstraintInducedMovement
Therapies(CIMTs)withusualcareforimprovingupperlimbfunctionandreducingdisabilityinpeople
afterstroke.Onlystudieswithaminimumsamplesizeof20participants(10ineacharm)and
includingatleast50%ofparticipantswithstrokewereselected.Fifteen(15)RCTswereidentified.
Table1summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudies.
Table103:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
54
Dahl,2008
Dromerick,
200072
POPULATION
Inpatientsafter
stroke(2weeks
8yearspost
stroke)with
unilateralhand
impairmentbut
morethan
20degrees
activewrist
extensionand
10degrees
activefinger
extension.
INTERVENTION
COMPARISON
OUTCOMES
CIMT:amitten
immobilisedthenon
paretichandfortarget
90%ofwakinghours
(actually13
hours/day).Training
providedingroupsof4
participantsledby
physicaland
occupationaltherapists
andassistedbytrained
nursesfor6hours/day
for10consecutive
weekdays;exercises
andactivitieschosen
from150in10fields
(includingpersonal
care,
kitchen/household
etc.).(N=18)
Usualcare:
communitybased
followupaccording
topatient'sneeds,
involvingbothupper
andlowerlimb
trainingandcould
includeinpatient
rehabilitation
(physiotherapyplus
occupational
therapy)following2
outpatientsessions
perweek.(N=12)
Inpatientsin
acutestroke&
braininjury
rehabilitation
service
(admission
within14days
ofischemic
stroke)with
persistent
hemiparesis
leadingto
CIMT:apaddedmitten
immobilisedthenon
paretichandforat
least6hours/day
duringthe14day
treatmentperiod.
Treatmentwas
directedtowards
subjectattentionand
efforttowardthe
hemipareticupper
extremityand
Usualcare:
traditional
occupationaltherapy
plusacircuittraining
programallowing
patientstoperform
bilateralselfrangeof
motionand
functionalactivities
inasupervised
settingfor2
hours/day,5
NationalClinicalGuidelineCentre,2013.
439
WolfMotor
FunctionTest
(WMFT)
Functional
Independence
Measure(FIM)
ActionResearch
ArmTest(ARAT)
BarthelIndex
Functional
Independence
Measure(FIM)
StrokeRehabilitation
Movement
STUDY
POPULATION
impairedupper
extremity
function.Details
onpatients
wristandfinger
extensionability
werenot
reported.
INTERVENTION
minimisedtheuseof
theuninvolvedupper
extremityduring
functionalactivities.All
subjectsalsoreceived
routine
interdisciplinarystroke
rehabilitation;
individualisedcircuit
trainingtechniques2
hours/day,5
days/weekfor2
consecutiveweeks.
(N=11)
COMPARISON
days/weekfor2
consecutiveweeks.
(N=9)
Hammer,
2009102
Patientsafter
stroke(16
monthspost
stroke)with
abilitytomove
theshoulder
andelbow
voluntarilyand
extend20
degreesinthe
wristand10
degreesinthe
fingersofthe
pareticarmand
hand.
Forcedusedsling:
arestrainingsling
immobilisedthe
unaffectedarmwitha
targetof6hours/day
(actuallyachieved3.7
hours/day)for5
days/weekfor2
weeks.Patientsalso
receivedindividualised
physicaland
occupationaltherapy
fortheupperand
lowerlimbtraining
basedonandtask
orientatedapproach.
(N=15)
Usualcare:standard
interdisciplinary
rehabilitationofdaily
training5days/week
(withoutforced
use).(N=15)
Lin,2009153
Chronic
unilateralstroke
patients(>6
monthspost
onsetof
ischemicor
hemorrhagic
stroke)with
Brunnstrom
abovestageIII
forproximal
anddistalpart
ofupperlimb;
anamountof
usescore<2.5
onthemotor
Activitylogof
theupperlimb
andModified
Ashworthscale
score2inany
jointofthe
shoulder,elbow
wrist,orfingers.
DistributedCIMT:
amittenrestrictedthe
movementof
unaffectedhandfor6
hours/day.Patients
alsoreceivedintensive
trainingofaffected
upperlimbin
functionaltasksfor2
hours/weekdayfor3
weeks.(N=20)
Usualcare:training
inhandfunction,
coordination,
balance,movements
ofaffectedupper
limbandpracticeon
functionaltaskswith
unaffectedorboth
limbs.(N=20)
NationalClinicalGuidelineCentre,2013.
440
OUTCOMES
FuglMeyer
Assessment
ActionResearch
ArmTest(ARAT)
FuglMeyer
Assessment(FMA)
Functional
Independence
Measure(FIM)
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
Lin,2007
Chronicstroke
patients(1326
monthspost
first
cerebrovascular
incident)with
Brunnstrom
abovestageIII
forproximal
anddistalpart
ofarm;an
amountofuse
score<2.5on
themotor
Activitylogof
theaffected
armand
Modified
Ashworthscale
score=2inany
jointofthe
shoulder,elbow
wrist,orfingers.
ModifiedCIMT:amitt
restrictedthe
movementof
unaffectedhandfora
targetof6hours/day
(actual6.2hours/day)
for3weeksplus
intensivetrainingof
affectedarm
supervisedbytrained
occupationaltherapists
for2hours/weekday.
(N=17)
Usualcare:involving
strength,balanceand
finemotordexterity
training,functional
taskpracticewhen
possible,and
stretching/weight
bearingbythe
affectedarm.(N=17)
Functional
Independence
Measure(FIM)
Myint,
2008182
Strokepatients
(216weeks
afterstroke)
with
hemiparesisof
theaffected
limbandwith
minimal
movementof
20degreeswrist
extensionand
10degrees
extensionofall
digits
CIMT:patientswearing
apaddedshoulder
slingfor90%ofwaking
hours).Patientsalso
received4hours/day
for5days/weekfor2
weekssupervised
activitiesincluding
shaping(abehavioural
methodtoimprove
motorperformancein
smallstepsand
encouragingpositive
feedback).(N=28)
Usualcare:
conventional
occupationaland
physicaltherapy
involving
neurodevelopmental
techniques,bimanual
tasks,compensatory
techniques,strength,
rangeofmotion,
positioning,mobility
for4hours/dayfor5
days/weekfor2
weeks.(N=20)
ActionResearch
ArmTest(ARAT)
BarthelIndex
9holepegtest
Page,
2008197
Chronicstroke
patients(>12
monthspost
stroke)ableto
selectively
activelyextend
atleast10
degreesatthe
metacarpophala
ngealand
interphalangeal
jointsand20
degreesatthe
wrist.
ModifiedCIMT:sling
andhandinmesh
polystyrenefilledmitt
restrictedtheuseof
unaffectedarmfor5
hours/weekdayduring
atimeoffrequentarm
use.Patientsalso
receivedtrainingofthe
moreaffectedarm
therapyusingshaping
techniquesforhalf
houronetoone
sessionsfor3daysper
weekfor10weeks
assistedbytherapists.
(N=13)
Usualcare:time
matched
rehabilitation
focusingon
proprioceptive
neuromuscular
facilitation
techniques,
stretchingand
compensatory
techniques.(N=12)
Inorout
ForceduseCIMT:
Usualcare:
154
Ploughman,
NationalClinicalGuidelineCentre,2013.
441
OUTCOMES
ActionResearch
ArmTest(ARAT)
FuglMeyer
assessment(FMA)
ActionResearch
StrokeRehabilitation
Movement
STUDY
2004206
POPULATION
patientswith
firststroke(less
than16weeks
poststroke)
showing
minimal
movementof
thearmand
hand(intertiary
mixed
rehabilitation
centre).Motor
controlofthe
upperextremity
ofmorethan
stage2butnot
morethanstage
6onthe
Chedoke
McMaster
Impairment
Inventory(CMII)
ofthearmand
hand.
INTERVENTION
patientswearingthick
constraintknitted
acrylicthumbless
mittentodiscourage
useofunaffectedarm
andhand;wornfor1
hourperdayincreasing
to6hoursby2weeks
and6hourstowards
theendoftreatment.
(N=13)
COMPARISON
OUTCOMES
conventionaltherapy
ArmTest(ARAT)
byfacilitatingthe
Functional
proximalmotor
Independent
controlprogressing
Measure(FIM)
toskilledtask
training,strengthand
endurancetraining,
functionalelectrical
stimulation,gait
training,education.
(N=14)
Taub,
2006254
Chronicstroke
patients
(mean=4.5
yearsafter
stroke)with
motordeficit;
abilityto
activelyextend
10degreesat
metacarpophala
ngealand
interphalangeal
jointsand20
degreesat
wrist.
CIMT:arestinghand
splint/slingonthe
unaffectedupper
extremityprevented
useofthatarmfora
targetof90%of
wakinghoursfor6
hours/dayfor10
consecutiveweekdays.
Trainingontheparetic
armconsistedof
'shaping'(a
behaviouralmethodto
improvemotor
performanceinsmall
stepsandencouraging
positivefeedback).
(N=21)
Usualcare:program
ofphysicalfitness,
cognitive,and
relaxationexercises
forthesamelength
oftimeandwiththe
sameamountof
interactionwiththe
therapistsasthe
interventiongroup.
(N=20)
vanderLee,
1999270
Patientsafter
singlestroke(at
least1yearpost
stroke)with
hemiparesison
thedominant
sideandwitha
minimumof20
degreesof
activewrist
extensionand
10degreesof
fingerextension
Forcedusetreatment:
unaffectedarmwas
immobilisedusing
splint(wornathome)
for6hours/dayfor5
days/weekduring12
daysoftreatmentplus
aclosedarmslingwas
attachedtothewaist
duringthetreatment
hours.(N=33)
Usualcare:involving
neurodevelopmental
bimanualtraining
providedingroupsof
4participants
(housekeeping
activities,
handicrafts,games)
(N=33)
FuglMeyer
Assessment(FMA)
ActionResearchArm
Test(ARAT)
Wolf,
Patientswith
CIMT:mittrestricted
Usualcare:ranged
NationalClinicalGuidelineCentre,2013.
442
WolfMotor
FunctionTest
(WMFT)
WolfMotor
StrokeRehabilitation
Movement
STUDY
2006286
POPULATION
firsttime
clinicalischemic
orhemorrhagic
cerebrovascular
accident(36
monthsafter
stroke).Lower
functioning
participantshad
atleast10
degreesof
activewrist
extension,at
least10degrees
ofthumb
abduction/exte
nsion,andat
least10degrees
ofextensionin
atleast2
additional
digits.These
movementshad
toberepeated
3timesin1
minute.
INTERVENTION
theuseofunaffected
handforagoalof90%
ofwakinghoursfora
totalof14days
(treatmentdaysplus
weekends);the
patientsalsoreceived
adaptivetaskpractice
andstandardtask
trainingofpareticlimb
for6hours/dayon
weekdays(N=106)
COMPARISON
OUTCOMES
fromnotreatmentto
FunctionTest
applicationof
(WMFT)
mechanical
interventions
(orthotics)orvarious
occupationaland
physicaltherapy
approachesathome
asadaypatientoras
anoutpatient.
(N=116)
patients(3
theuseoflessaffected
weeksto37
handfor6hours/day
monthspost
for5days/weekfor3
onsetwith
weeksandpatients
Brunnstrom
alsoreceivedtypical
abovestageIII
trainingactivitiesfor
forproximal
dailytasksinvolving
partofupper
theuseofmore
limb;anamount affectedlimb(2
ofusescore
hours/day).Also
<2.5onthe
received
MotorActivity
interdisciplinary
logoftheupper rehabilitation(1.5
limb;noserious hours/dayfor5
cognitive
days/week).(N=24)
deficits.
Usualcare:
neurodevelopmental
therapyemphasising
functionaltask
practice,stretching
andweightbearing
withmoreaffected
armandfinemotor
dexterityfor2
hours/day.Also
received
interdisciplinary
rehabilitation(1.5
hours/dayfor5
days/week).(N=23)
FuglMeyer
Assessment(FMA)
Wu,2007(b) Elderlystroke
292
patients(mean
age72years)
with0.531
monthspost
onsetofafirst
everstrokewith
considerable
nonuseofthe
affectedlimb
(anamountof
usescore<2.5
ontheMotor
Usualcare:
2hourtherapy
sessionwith75%of
timespenton
neurodevelopmental
techniques
emphasising
functionaltask
practice,stretching,
weightbearing,fine
motordexterityand
25%on
compensatory
FuglMeyer
Assessment(FMA)
Functional
Independence
Measure(FIM)
ModifiedCIMT:amitt
wasappliedinthe
unaffectedhandfor6
hours/weekdayat
timeoffrequentarm
usefor3weeks.
Patientsalsoreceived
individualised2hour
therapysessionsfor5
times/weekinvolving
shapingandadaptive
repetitivetasks
focusingondaily
NationalClinicalGuidelineCentre,2013.
443
StrokeRehabilitation
Movement
STUDY
POPULATION
Activitylog);
Modified
Ashworthscale
score2inany
joint
INTERVENTION
activitiesand15
minutesoftherapyon
normalisingmuscle
tone.(N=13)
COMPARISON
techniques.(N=13)
OUTCOMES
Wu,2007(c
)293
Poststroke
patients(1236
monthspost
strokeofafirst
ever
cerebrovascular
accident)with
anabilityto
activelyextend
atleast10
degreesatthe
metacarpophala
ngealand
interphalangeal
jointsand20
degreesatthe
wrist.
ModifiedCIMT:amitt
wasappliedinthe
unaffectedhandfor6
hours/weekdayat
timeoffrequentarm
usefor3weeks.
Trainingadministered
intensively2hoursper
day,5daysperweek,
for3weeks.Training
tookplaceduring
scheduled
occupationaltherapy
sessions,andother
routine
interdisciplinarystroke
rehabilitation
proceededasusual.
(N=15)
Usualcare:patients
receivedtraining
matchedtothe
mCIMTinduration
andintensityof
occupationaltherapy
activities.(N=15)
Functional
Independence
Measure(FIM)
Wu,2011294
Patientsafter
stroke(mean
poststroke16.2
months)and
mildto
moderate
motor
impairment
with
Brunnstrom
abovestageIII
forproximal
partofupper
extremity;an
amountofuse
score<2.5on
theMotor
Activitylog.
1)DistributedCIMT:
amittrestrictedthe
unaffectedhandfor6
hours/dayand
intensivelytrainedthe
affectedupper
extremityinfunctional
tasks.(N=22)
2)Bilateralarm
trainingfocusingon
thesimultaneous
movementsin
symmetricor
alternatingpatternsof
bothupperextremities
infunctionaltasks.
(N=22)
Studyduration:2
hours/day,5
days/weekfor3weeks
Usualcare:
patientsreceived
compensatory
practiceon
functionaltaskswith
theunaffectedupper
extremityorboth
upperextremities.
(N=22)
WolfMotor
FunctionTest
(WMFT)
NationalClinicalGuidelineCentre,2013.
444
StrokeRehabilitation
Movement
Comparison:constraintInducedmovementtherapiesversususualcare
Table104:Constraintinducedmovementtherapy(CIMT)versususualcareclinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
Confidence
(ineffect)
ActionResearchArmTest(posttreatment)(Betterindicatedbyhighervalues)
4
Dromerick
200072;
Myint
2008182;
Page
2008197;
VanderLee
1999270
RCTssingle
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
Dromerick
2000:52.8
(5.9)
Myint
2008:47.1
(10.2)
Page2008:
40.54
(8.18)
Vander
Lee:39.2
(13.1)
Dromerick
2000:44.3
(11.1)
Myint
2008:33.6
(12.5)
Page2008:
29.17(10)
Vander
Lee:30
(13.9)
10.78
(7.27,
14.30)
MD10.78
higher
(7.27to
14.30
higher)
Low
38(12.3)
30.8(13.6)
MD7.20
(0.94,
13.46)
MD7.20
higher
(0.94to
13.46
higher)
Moderate
ActionResearchArmTest(4weeksfollowup)(Betterindicatedbyhighervalues)
1Vander
Lee1999270
RCTsingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
ActionResearchArmTest(12weeksfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
445
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Effect
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
1Myint
2008182
RCTsingle
blinded
Serious
limitations(a
)
Noserious
indirectness
Noserious
indirectness
Noserious
imprecision
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
Confidence
(ineffect)
49.6(9.9)
39.9(14.1)
MD9.7
(2.51,
16.89)
MD9.7
higher
(2.51to
16.89
higher)
Moderate
38.5(13.6)
30.7(14.2)
MD7.80
(1.09,
14.51)
MD7.80
higher
(1.09to
14.51
higher)
Moderate
Dahl2008:
2.03(0.82)
Taub
2010:4.6
(4.4)
Wu
2011:5.83
(4.65)
MD0.53(
0.91,0.16
)
MD0.53
lower(0.91
to0.16
lower)
Low
1.77(0.92)
MD0.05(
0.59,0.69)
MD0.05
higher
(0.59lower
to0.69
Low
ActionResearchArmTest(10monthsfollowup)(Betterindicatedbyhighervalues)
1Vander
Lee1999270
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(b)
WolfMotorFunctionTest(performancetime)(posttreatment)(Betterindicatedbylowervalues)
3Dahl
RCTssingle
200854;
blinded
Taub
2006254;Wu
2011294
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
e)
Dahl2008:
1.56(0.57)
Taub
2010:3
(1.1)
Wu
2011:4.02
(2.49)
WolfMotorFunctionTest(performancetime)(6monthsfollowup)(Betterindicatedbylowervalues)
1Dahl
200854
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
e)
1.82(0.8)
NationalClinicalGuidelineCentre,2013.
446
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
higher)
(f)
(f)
MD34.00
(14.50,
53.50)
MD34.00
higher
(14.50to
53.50
higher)
High
Dahl
2008:3.47(
0.6)
Taub
2010:2.9(0.
5)
Wu
2011:3.66
(0.87)
MD0.46
(0.08,0.
83)
MD
0.46higher
(0.08to0.
83higher)
Low
3.73(0.58)
MD0.22(
0.21,0.65)
MD0.22
higher
(0.21lower
to0.65
higher)
Low
Confidence
(ineffect)
ChangeinWolfMotorFunctionTest(performancetime)(12monthsfollowup)
1Wolf
2006286
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
WolfMotorFunctionTest(functionalability)(posttreatment)(Betterindicatedbyhighervalues)
3Dahl2008 RCTssingle
54
;Taub
blinded
2006254;Wu
2011294
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
e)
Dahl
2008:3.85
(0.5)
Taub2010:
3.2(0.4)
Wu
2011:3.78
(0.71)
WolfMotorFunctionTest(functionalability)(6monthsfollowup)(Betterindicatedbyhighervalues)
1Dahl
200854
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
e)
3.95(0.61)
ChangeinWolfMotorFunctionTest(functionalability)(12monthsfollowup)
NationalClinicalGuidelineCentre,2013.
447
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Effect
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
Confidence
(ineffect)
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
1Wolf
2006286
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
e)
(f)
(f)
MD0.11(
0.05,0.27)
MD0.11
higher
(0.05lower
to0.27
higher)
Moderate
Serious
imprecision(
e)
(f)
(f)
MD0.67(
1.51,2.85)
MD0.67
higher
(1.51lower
to2.85
higher)
Moderate
Serious
imprecision(
e)
(f)
(f)
MD2.64(
6.27,0.99)
MD2.64
lower
(6.27lower
to0.99
higher)
Moderate
Dahl
2008:107.3
3(8.8)
Lin
2007:113.0
6(10.55)
Lin
Dahl
2008:111.6
7(6.49)
Lin
2007:105.6
7(15.85)
Lin
MD2.87( MD2.87
0.12,5.87) (0.12lower
to5.87
higher)
ChangeinWolfMotorFunctionTest(weight)(12monthsfollowup)
1Wolf
2006286
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
ChangeinWolfMotorFunctionTest(grip)(12monthsfollowup)
1Wolf
2006286
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
FunctionalIndependenceMeasure(posttreatment)(Betterindicatedbyhighervalues)
5
Dahl200854;
Lin2007154;
Lin2009153;
Wu
2007(b)292
Wu2007
RCTS4
Serious
single
limitations(c
blinded1
)
double
blinded
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
NationalClinicalGuidelineCentre,2013.
448
Low
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
(c)293
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
CIMT
Mean
(SD)/
Frequencie
s(%)
2009:122.0
5(5.6)
Wu
2007:104.8
5(12.13)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
2009:116.6
5(8.34)
Wu2007:
100.85
(20.08)
Confidence
(ineffect)
FunctionalIndependenceMeasure(totalscore)(6monthsfollowup)(Betterindicatedbyhighervalues)
1Dahl
200854
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
109.56
(8.25)
112.92
(6.75)
MD3.36
MD3.36
(8.762.04) lower(8.76
lowerto
2.04
higher)
Low
6.27(0.78)
6(0.92)
MD0.27(
0.49,1.03)
MD0.27
higher
(0.49lower
to1.03
higher)
Low
5.62(0.52)
MD0.47
(0.01,0.93)
MD0.47
higher
(0.01to
0.93
higher)
Low
FunctionalIndependencemeasure(eating)(posttreatment)(Betterindicatedbyhighervalues)
1Dromerick RCTsingle
200072
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
FunctionalIndependencemeasure(grooming)(posttreatment)(Betterindicatedbyhighervalues)
1Dromerick RCTsingle
200072
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
6.09(0.53)
FunctionalIndependencemeasure(bathing)(posttreatment)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
449
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
1Dromerick RCTsingle
200072
blinded
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
Confidence
(ineffect)
5.27(1.1)
5.25(0.46)
MD0.02(
0.70,0.74)
MD0.02
(0.70lower
to0.74
higher)
Low
FunctionalIndependencemeasure(upperextremitydressing)(posttreatment)(Betterindicatedbyhighervalues)
1Dromerick RCTsingle
200072
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
6(0.77)
5.25(0.71)
MD0.75
(0.10,1.40)
MD0.75
(0.10to
1.40
higher)
Low
Serious
imprecision(l
)
Lin2009:
52.3(7.17)
Page2008:
48.23
(8.06)Wu
2007(a):
46.75
(11.58)Wu
2007(b:
49.54
(12.84))
Lin2009:
51.25(12.59
)
Page2008:
42.42(12)
Wu2007
(a);:44.78
(13.08)Wu
2007(b):
49.38
(10.18)
MD2.15(
1.56,5.86)
MD2.15
higher
(1.56lower
to5.86
higher)
Low
Noserious
imprecision
51.6(8)
45(10.6)
MD6.60
(2.07,11.13
)
MD6.60
higher
(2.07to
High
FuglMeyerAssessment(posttreatment)(Betterindicatedbyhighervalues)
4Lin
RCTssingle
2009153;
blinded
Page
2008197;Wu
2007(a)291;
Wu2007
(b)292
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
FuglMeyerAssessment(3weeksfollowup)(Betterindicatedbyhighervalues)
1Vander
Lee1999270
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
450
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
11.13
higher)
Confidence
(ineffect)
FuglMeyerAssessment(6weeksfollowup)(Betterindicatedbyhighervalues)
1Vander
Lee1999270
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(l
)
52.3(8.3)
46.7(9.6)
MD5.60
MD5.60
(1.27,9.93) higher
(1.27to
9.93
higher)
Moderate
Serious
imprecision(l
)
50.9(9.9)
45.5(9.7)
MD5.4
(0.67,
10.13)
MD5.4
higher
(0.67to
10.13
higher)
Moderate
Serious
imprecision
(h)
Dromerick
2000:100
(1.1)
Myint
2008:92.6
(8.5)
Dromerick
2000:98.5
(3.77)
Myint
2008:85.3
(13.6)
MD3.53(
1.89,8.95)
MD3.53
higher
(1.89lower
to8.95
higher)
Low
Serious
imprecision
97.6(4.2)
93.4(7.7)
MD4.2
MD4.2
(0.48,7.92) higher
FuglMeyerAssessment(1yearfollowup)(Betterindicatedbyhighervalues)
1Vander
Lee1999270
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(posttreatment)(Betterindicatedbyhighervalues)
2Dromerick RCTssingle
200072;
blinded
Myint
2008182
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(12weeksfollowup)(Betterindicatedbyhighervalues)
1Myint
2008182
RCTsingle
blinded
Serious
limitations(a
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
451
Low
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
)
Inconsistency
Indirectness
Imprecision
(i)
CIMT
Mean
(SD)/
Frequencie
s(%)
Absolute
Effect/
Usualcare Mean
Mean
Mean(SD)/ difference/ Difference
Frequencie RiskRatio (MD)(95%
s(%)
(95%CI)
CI)
(0.48to
7.92
higher)
Confidence
(ineffect)
9holepegtest(posttreatment)(Betterindicatedbyhighervalues)
1Myint
2008182
RCTsingle
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(j)
16/28
(57.1%)
9/20(45%)
RR1.27
(0.71to
2.27)
121more
per1000
(from131
fewerto
572more)
VeryLow
Serious
imprecision(
k)
18/28
(64.2%)
10/20
(50%)
RR1.29
(0.77to
2.16)
145more
per1000
(from115
fewerto
580more)
Low
4/18
(22.2%)
0/12
RR6.16
(0.36to
104.90)
145more
per1000
(from115
fewerto
580more)
VeryLow
9holepegtest(3monthsfollowup)(Betterindicatedbyhighervalues)
1Myint
2008182
RCTsingle
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Adverseevent(muscletendernessintheaffectedarm)(Betterindicatedbylowervalues)
1Dahl
200854
(a)
(b)
(c)
(d)
(e)
RCTsingle
blinded
Serious
limitations(d
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(j)
Unclearallocationconcealmentandunclearallocationconcealment
MeandifferencedidnotreachtheagreedMIDof12points.
Unclearrandomizationandunclearallocationconcealment
unclearrandomization
MeandifferencedidnotreachtheagreedMIDof19points.
NationalClinicalGuidelineCentre,2013.
452
StrokeRehabilitation
Movement
(f)
(g)
(h)
(i)
(j)
(k)
(l)
Nomeans(SD)pergroupwerereportedbyauthors.
MeandifferencedidnotreachtheagreedMIDof22pointsforthetotalscore,17pointsonthemotorscaleand3pointsforthecognitivescale.
MeandifferencedidnotreachtheagreedMIDof6.6points.
MeandifferencedidnotreachtheagreedMIDof9.25points.
ConfidenceintervalcrossedbothendsofdefaultMID.
ConfidenceintervalcrossesoneendofdefaultMID.
MeandifferencedidnotreachagreedMID(10%differenceoftotalscore)
Narrativesummaries
ThefollowingstudiesaresummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Onestudy102randomisedaconvenientsampleof30participantsintoforcedusetraining(N=15)andstandardrehabilitationprogramme(N=15).Thestudy
foundthatthechangesintheforcedusegroupdidnotdifferfromthechangesinthestandardrehabilitationgroupforanyoftheoutcomemeasures
(FuglMeyerAssessmentandActionResearchArmTest).Bothgroupsimprovedovertime(posttreatment3monthsfollowup),withstatistically
significantchangesintheFuglMeyerAssessment(meanscorechangedfrom52to57).Thetrialwasunblindedandofasmallsamplesize.
Inonestudy206,30participantswererandomlyallocatedtoforcedusetherapyandconventionaltherapy.Participantsintheforcedusegrouphadan85%
improvement(baseline=20.7(15.49))inARATscore,whereasthosewhoreceivedconventionaltherapyhada74%improvement(baseline=16.0(13.64))
(p=0.20).NosignificantdifferenceinFIMwasobserved(datanotpresented).Noneoftheparticipantsintheforcedusegroupachieved6hoursof
constraintwearingaday(averagetime=2.7hours/day).Datawerepresentedasgraphsandtheycouldnotbeextracted/usedformetaanalysis.The
analysiswasnotdonebasedonITTandthestudyhadunclearrandomisationandallocationconcealment.
NationalClinicalGuidelineCentre,2013.
453
StrokeRehabilitation
Movement
13.5.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingconstraintinducedmovementtherapywithusualcare
wereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Lookingatresourcesusedinthestudiesincludedintheclinicalreview,themaindifferencein
resourcesusedbetweeninterventionandusualcarewasoftheconstraintused,withnosubstantial
differenceinpersonneltime.TheGDGadvisedthatthecostofconstraintwasminimalforexample
itmayinvolveusingbandaging.However,thecostsattributabletoCIMTwilldependonhowand
whenitisoffered.IfCIMTactivitiesareincorporatedaspartoftheusualrehabilitation,costsmay
notbesubstantiallyhigherthanusualcare;if CIMTisofferedinadditiontousualrehabilitationcare
thatpatientsreceive,additionalcostswouldbeincurredduetoadditionalresourceuse(forexample,
stafftime).
13.5.1.3
Evidencestatements
Clinicalevidencestatements
Fourstudies72,182,197,270of159participantsfoundthatpatientswhoreceivedconstraintinduced
movementtherapyshowedstatisticallysignificantimprovementinActionResearchArmtest
comparedtopatientswhoreceivedusualcareatpostintervention,althoughitwasnotofclinical
significance(LOWCONFIDENCEINEFFECT).
Onestudy270of66participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinActionResearchArmtestcomparedto
patientswhoreceivedusualcareat4weeksfollowup,althoughitwasnotofclinicalsignificance
(MODERATECONFIDENCEINEFFECT).
Onestudy182of66participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinActionResearchArmtestcomparedto
patientswhoreceivedusualcareat12weeksfollowup(MODERATECONFIDENCEINEFFECT).
Onestudy270of66participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinActionResearchArmtestcomparedto
patientswhoreceivedusualcareat10monthsfollowup,althoughitwasnotofclinicalsignificance
(MODERATECONFIDENCEINEFFECT).
Threestudies54,254,294of115participantsfoundthatpatientswhoreceivedconstraintinduced
movementtherapyshowedstatisticallysignificantimprovementinWolfMotorFunctiontest
performancetimecomparedtopatientswhoreceivedusualcareatpostintervention,althoughit
wasnotofclinicalsignificance(LOWCONFIDENCEINEFFECT).
Onestudy54of30participantsshowedthattherewasnosignificantdifferenceinperformancetime
oftheWolfMotorFunctiontestbetweenthosepatientswhoreceivedconstraintinducedmovement
therapyandthosewhoreceivedusualcareat6months(LOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
454
StrokeRehabilitation
Movement
Onestudy286of222participantsshowedthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinchangeintheWolfMotorFunctiontest
performancetimecomparedtopatientswhoreceivedusualcareat12months(HIGHCONFIDENCE
INEFFECT).
Threestudies54,254,294of115participantsfoundthatpatientswhoreceivedconstraintinduced
movementtherapyshowedstatisticallysignificantimprovementinfunctionalabilityofWolfMotor
Functiontestcomparedtopatientswhoreceivedusualcareatpostintervention,althoughitwasnot
ofclinicalsignificance(LOWCONFIDENCEINEFFECT).
Onestudy54of30participantsshowednosignificantdifferenceinthefunctionalabilityoftheWolf
MotorFunctiontestbetweenthosewhoreceivedconstraintinducedmovementtherapyandthose
whoreceivedusualcareat6months(LOWCONFIDENCEINEFFECT).
Onestudy286of222participantsshowednosignificantdifferenceinthechangeoftheWolfMotor
Functiontestbetweenthosewhoreceivedconstraintinducedmovementtherapyandthosewho
receivedusualcareat12monthsforthefollowingscales:
Functionalability(MODERATECONFIDENCEINEFFECT)
Weight(MODERATECONFIDENCEINEFFECT)
Grip(MODERATECONFIDENCEINEFFECT)
Fivestudies153,197,154,292(Lin2009,Page2008,Lin2007Wu2007(b))of160participantsshowedthat
therewasnostatisticallysignificantdifferenceintheFunctionalIndependenceMeasure(totalscore)
betweenthosepatientswhoreceivedconstraintinducedmovementtherapyandthosewhoreceived
usualcareatpostintervention(LOWCONFIDENCEINEFFECT).
Onestudy54of30participantsshowednosignificantdifferenceintheFunctionalIndependence
Measure(totalscore)betweenthosewhoreceivedconstraintinducedmovementtherapyandthose
whoreceivedusualcareat6months(LOWCONFIDENCEINEFFECT).
Onestudy72of20participantsshowednosignificantdifferenceinthefollowingscalesofthe
FunctionalIndependenceMeasurebetweenthosewhoreceivedconstraintinducedmovement
therapyandthose197whoreceivedusualcareatpostintervention:
eating(LOWCONFIDENCEINEFFECT)
Bathing(LOWCONFIDENCEINEFFECT).
Onestudy72of20participantsshowedthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinthefollowingscalesoftheFunctional
IndependenceMeasurecomparedtopatientswhoreceivedusualcareatpostintervention,although
thesedifferenceswerenotofclinicalsignificance:
grooming(LOWCONFIDENCEINEFFECT)
upperextremitydressing(LOWCONFIDENCEINEFFECT
NationalClinicalGuidelineCentre,2013.
455
StrokeRehabilitation
Movement
Fourstudies153,197,291,292of138participantsshowedthattherewasnostatisticallysignificant
differenceinFuglMeyerassessmentbetweenthosepatientswhoreceivedconstraintinduced
movementtherapyandthosewhoreceivedusualcareatpostintervention(LOWCONFIDENCEIN
EFFECT).
Onestudy270of66participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinFuglMeyerassessmentcomparedto
patientswhoreceivedusualcareat3weeksfollowup(HIGHCONFIDENCEINEFFECT).
Onestudy270of66participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinFuglMeyerassessmentcomparedtopatients
whoreceivedusualcareat6weeksfollowup,althoughitwasnotofclinicalsignificance(MODERATE
CONFIDENCEINEFFECT).
Onestudy270of66participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinFuglMeyerassessmentcomparedtopatients
whoreceivedusualcareat1yearfollowupalthoughitwasnotofclinicalsignificance(MODERATE
CONFIDENCEINEFFECT).
Twostudies72,182of68participantsshowedthattherewasnostatisticallysignificantdifferencein
BarthelIndexbetweenthosepatientswhoreceivedconstraintinducedmovementtherapyandthose
whoreceivedusualcareatpostintervention(LOWCONFIDENCEINEFFECT).
Onestudy182of48participantsfoundthatpatientswhoreceivedconstraintinducedmovement
therapyshowedstatisticallysignificantimprovementinBarthelIndexbetweenthosepatientswho
receivedconstraintinducedmovementtherapyandthosewhoreceivedusualcareat12weeks
followup,althoughitwasnotofclinicalsignificance(LOWCONFIDENCEINEFFECT).
Onestudy182of48participantsshowedthattherewasnostatisticallysignificantdifferenceinNine
holePegtestbetweenthosepatientswhoreceivedconstraintinducedmovementtherapyandthose
whoreceivedusualcareatpostintervention(VERYLOWCONFIDENCEINEFFECT).
Onestudy182of48participantsshowedthattherewasnostatisticallysignificantdifferenceinNine
holePegtestbetweenthosepatientswhoreceivedconstraintinducedmovementtherapyandthose
whoreceivedusualcareat3monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy54of30participantsshowedthattherewasnostatisticallysignificantdifferenceinthe
experienceofmuscletendernessintheaffectedarmbetweenpatientswhoreceivedconstraint
inducedmovementtherapyandthosewhoreceivedusualcare(VERYLOWCONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.5.2
Recommendationsandlinktoevidence
95.Considerconstraintinducedmovementtherapyforpeople
withstrokewhohavemovementof20degreesofwrist
extensionand10degreesoffingerextension.Beawareof
potentialadverseevents(suchasfalls,lowmoodand
fatigue).
Relativevaluesofdifferent
outcomes
TheoutcomesofinterestincludedtheFunctionalIndependence
Measure(FIM),BarthelIndex,FuglMeyerscore,ActionResearchArm
test(ARAT),WolfMotorFunctionTest(WMFT)and9holepegtest.The
9holepegtestmaybeinsensitiveasitisameasureoffinefinger
NationalClinicalGuidelineCentre,2013.
456
StrokeRehabilitation
Movement
movementsaswellastheabilitytoreach.SimilarlytheFIMandBarthel
Indexasmeasuresofdependencemaybeunresponsivetochangesin
upperlimbfunction.TheGDGnotedthatthereisnopsychometrically
robustpatientreportedoutcomemeasuresfocussingonupperlimb
activity.Anyadverseeventwasalsoincludedwherereported.
Tradeoffbetweenclinical
benefitsandharms
Participantswhoreceivedconstraintinducedmovementtherapy(CIMT)
demonstratedaclinicallysignificantimprovementinFuglMeyerscores
atashorttermof3weeksandinARATscoresandfunctionalabilityof
WMFTat12weeksfollowup.Althoughtheimprovementin
performancetimeofWMFTforthosewhoreceivedCIMTwasnotof
clinicalsignificanceatpostinterventionandat6monthsfollowup,the
differenceinthisoutcomebecameclinicallysignificantbetweenthe
participantsintheCIMTandtheusualcaregroupsat1yearfollowup.
Onlyonestudyreportedadverseevents,theexperienceofmuscle
tendernessintheaffectedarm(Dahl,2008)54duringconstraintinduced
movementtherapy,thoughitsprevalencewasnotsignificantlydifferent
betweenthetwogroups.HowevertheGDGconsideredtherewere
possibleharmsassociatedwiththistherapyandagreedthatwhen
selectingpatientsforCIMT,attentionneedstobemadetopotential
adverseeventssuchasfallinganddeteriorationinmood.
Economicconsiderations
Nocosteffectivenessstudieswereidentifiedforthisquestion.IfCIMT
activitiesareincorporatedaspartoftheusualrehabilitation,costsmay
notbesubstantiallyhigherthanusualcare.However,offeringCIMT
mightrepresentachangeintheusualactivitiesthatarepartofthe
rehabilitation.TheGDGagreedthatitisunlikelythatCIMTisofferedin
additiontousualrehabilitationcareandthereforenoadditionalcosts
wouldbeincurredduetoadditionalresourceuse.
Qualityofevidence
TheconfidenceintheeffectfortheoutcomesofFunctional
IndependenceMeasure,Barthelindex,WolfMotorFunctionTest,and9
holepegtestrangedfromhightoverylowduetolimitationsinstudy
design(unclearallocationconcealmentandunclearrandomisation)and
imprecisionaroundtheeffectestimate.TheGDGacknowledgedthat
duetothenatureoftheinterventionitwasdifficulttorecruitpeople
intostudies.Themeanageofstrokesurvivorsis7374245anditwas
notedthatthepatientswithinthesestudies,withtheexceptionofthe
Wustudy2007(b)292,wererelativelyyoungforastrokepopulationand
arelikelytoreflectthosewhoareadmittedintospecialistrehabilitation
units.
Inpatientswithmovementof20degreeswristand10degreesinfingers
ConstraintInducedMovementTherapywithrepetitivetaskpracticemay
beofbenefitforpatientsbothearly(2weeksafteronset)andlateafter
stroke.
Otherconsiderations
TheGDGagreedthattheactiveelementofconstrainttherapyisthe
amountofpracticeperformedbytheweakarmandthisneedstobe
carefullystructuredandtailoredtotheindividualpatientneeds.The
GDGwereunsurewhatvaluepatientsplaceonsmallimprovementsin
upperlimbfunction.Whilstthistypeofinterventionmaynotbesuitable
ortoleratedbysomepatients,theGDGagreedthatitisanintervention
thattendstobeusedwiththosepatientswhoarehighlymotivatedto
gettheirmovementsbackanditisthesewhowouldvaluethistypeof
interventionmost.
NationalClinicalGuidelineCentre,2013.
457
StrokeRehabilitation
Movement
13.6 Shoulderpain
Therewasalackofdirectevidenceforthetreatmentofshoulderpain.Thereforerecommendations
inthissectionwerebasedonmodifiedDelphiconsensusstatementsderivedfrompublishednational
andinternationalguidance.ThissectionoftheDelphisurveywasaimedatthoseDelphipanel
memberswhofelttheyhadtherelevantexperiencetocommentonshoulderpain.OtherDelphi
panelmemberscouldoptoutofthissection.Responserateswerethereforelowerinthissection.
Belowweprovidetablesofstatementsthatreachedconsensusandstatementsthatdidnotreach
consensusandgiveasummaryofhowtheywereusedtodrawuptherecommendations.Fordetails
ontheprocessandmethodologyusedforthemodifiedDelphisurveyseeAppendixF.
13.6.1
Howshouldpeoplewithshoulderpainafterstrokebemanagedtoreducepain?
Population
Adultsandyoungpeople16orolderwhohavehadastrokeandhavesymptomsof
shoulderpain
Components
Assessment
Painmanagement
FES
Physicaltherapies
Outcomes
13.6.2
Mobility
Function
pain
Delphistatementswhereconsensuswasachieved
Table105:Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
1.
Results
%
Statement
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Informationshouldbeprovidedbythe
healthcareprofessionalonhowto
preventpain/traumatotheshoulder.
77.6
7/49(14%)panelmembers
commented
Mostpanelmemberswho
commentedonthisquestionqueried
whotogivetheinformationto
(patient,carer,otherstaff)and
underwhichconditions(ifthereis
weaknessintheshoulder).
Itwasstatedinonecommentthat
therewasnoinformationavailable
onthistopic.
Whenmanagingshoulderpainthe
followingtreatmentsshouldbe
considered:
Positioning
70.7
Inround223/49(47%)panel
memberscommented;13/42(31%)
inround3
Noneoftheothertreatment
NationalClinicalGuidelineCentre,2013.
458
StrokeRehabilitation
Movement
Number
Statement
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
optionsgainedconsensus the
optionswere:
Armsslings,
Shouldersupport,
Highintensitytranscutaneous
nervestimulation,and
FunctionalElectricalStimulation
Analgesics
Physicaltherapies
Strapping
Commentsweredivided:
Anumberofpanelmembers
statedthatshoulderpainhastobe
treatedinaflexiblemannerand
accordingtoindividualneeds.
Somestatedthattreatmentshould
beevidencebased.
Othersstatedthattheevidence
formostoftheoptionswaspoor
13.6.3
Delphistatementwhereconsensuswasnotreached
Table106:Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
Statement
Results
%
Amountandcontentofpanel
commentsorthemes
1.
Thepersonwhohashadastroke
shouldbeassessedforshoulderpain
63.6
Inround213/48(27%)panel
memberscommented;7/42(17%)in
round3
Therewasageneralopinionthat
thisshouldbeeasilyascertainedand
thereforeafullassessmentisnot
needed.
2.
Thereisaneedforanalgorithmto
assessandtreatshoulderpain
31.0
Inround223/49(47%)panel
memberscommented;13/42(31%)
inround3
Somecommentsweremadethat
therearealgorithmsalreadyin
existence.
Otherscommentedthatthe
evidencefortreatmentswaspoor
andthereforethereisnotenough
informationtocreateanalgorithm.
Therewerealsocommentsthatthis
NationalClinicalGuidelineCentre,2013.
459
StrokeRehabilitation
Movement
Number
13.6.4
Results
%
Statement
Amountandcontentofpanel
commentsorthemes
wouldbeuseful.
RecommendationsandlinkstoDelphiconsensussurvey
Statements
Recommendation
32.Informationshouldbeprovidedbythehealthcareprofessionalon
howtopreventpain/traumatotheshoulder.
33.Whenmanagingshoulderpainthefollowingtreatmentsshouldbe
Whenmanagingshoulderpainthefollowingtreatmentsshouldbe
considered:
Positioning
96.Provideinformationforpeoplewithstrokeandtheirfamiliesand
carersonhowtopreventpainortraumatotheshoulderiftheyare
atriskofdevelopingshoulderpain(forexample,iftheyhaveupper
limbweaknessandspasticity).
97.Manageshoulderpainafterstrokeusingappropriatepositioningand
othertreatmentsaccordingtoeachpersonsneed.
98.ForguidanceonmanagingneuropathicpainfollowNeuropathicpain
(NICEclinicalguideline96).
Economicconsiderations Thereisaminorcostofstafftimeassociatedwiththeprovisionof
information.HowevertheGDGconsideredthesetobelargelyoffsetby
thebenefits.
Otherconsiderations
TheGDGagreedthiswasacommonproblemamongstpeopleafter
strokeandthatpreventionshouldbehighlighted.However,themeansof
preventingshoulderpainisnotuniversallyagreedandthismaybedue,
tothelargearrayofidentifiedcauses,includingspasticity,thalamic
(central)pain,complexregionalpainsyndromes(CRPS),(forexample:
shoulderhandsyndrome),fractureandsofttissueproblems.Itis
generallyagreedthatoneofthemajorcausesofinjuringtheshoulderis
poormanualhandlingandsupportoftheatriskarmbyhealth
professionals,carers,orthepatientthemselves.
Inthesurveyconsensuswasreachedonlyforprovidinginformationto
preventshoulderpain.TheGDGclarifiedthisstatementbyindicatingthe
peoplelikelytodevelopshoulderpainwerethosewithchangesintone
orpowerintheirarms.Algorithmsandassessmentsdidnotreach
consensusintheDelphiandtheGDGdiscussedtheprosandconsof
includingthisinalistofassessmentsroutinelycarriedout.Itwasfeltthat
askingpeoplewhodisplaydiscomfortwhenmovingtheirarmswouldbe
sufficientinthemajorityofcases.
Whilsttherewasconsensusthatpositioningtheshouldermayhelpto
alleviatesymptoms,overalltheviewfromthesurveyshowedtherewas
noevidencebasetorecommendanyparticulartreatment.TheGDG
NationalClinicalGuidelineCentre,2013.
460
StrokeRehabilitation
Movement
agreedthiswasanareawherefurtherresearchwasneededtoassessthe
effectivenessofthevariousmanagementstrategiescurrentlyused,and
agreedthataresearchrecommendationbeincludedintheguideline.
TheGDGweresurprisedthattherewasnoagreementabouttheuseofa
simpletreatmentsuchasanalgesicstoalleviatepain.Thegroupagreed
thatwhilstitwasnotpossibletomakearecommendation,health
professionalsshouldconsiderothertreatmentsaccordingtoindividual
need.TheGDGacknowledgedthatavariedrangeoftherapieswere
currentlybeingusedinpracticethatinclude:upperlimbsupport
includingslingsandorthotics,strappingoftheshoulder,rangeofmotion
exercises,ultrasound,electricalstimulation,steroidandbotulinumtoxin
injections,acupunctureandmassagetherapy.Theuseofshoulderslings
maybeassociatedwithsomerisks,includingholdingthelimbinapoor
positionthatislikelytocausesofttissuecontracture,inhibitinguseofa
recoveringlimb,andhaveanadverseeffectonsymmetryandbalance,
makingfallsmorelikely.
13.7 Repetitivetasktraining
Rehabilitationisintegraltothecarepathwayafterstroke.Howevertheoptimumcomponentsof
physicalrehabilitationareuncertain.Repetitivetasktrainingpromotestherepetitionofmotor
movementrelatedtopurposefultasks.Thismightforexampleincludereachingforacuporcombing
hair.Thefocusisgenerallyontheimpairedlimbandisoneapproachtoincreasetheamountof
physicalrehabilitation.
13.7.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
repetitivetasktrainingversususualcareonimprovingfunctionandreducingdisability?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
Repetitivetasktraining
Lowerlimbfunctionaltasksand/or
Upperlimbfunctionaltasks
Comparison
Usualcare
Outcomes
Lowerlimb
Anytimedwalk;6minutewalktest,5metre,10metretimedwalk
Changeinwalkingdistance
Rivermeadmobilityindex
Upperlimb
Arm:
FuglMeyerAssessment,
ActionResearchArmTest(ARAT)
Hand:
Anypegholetest,
FrenchayArmTest,
MotorAssessmentScale(MAS)
NationalClinicalGuidelineCentre,2013.
461
StrokeRehabilitation
Movement
13.7.1.1
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingtheclinicaleffectivenessof
repetitivetasktrainingwithusualcaretoimprovefunctionandreducedisabilityforadultsandyoung
people16orolderwhohavehadastroke.Onlystudieswithaminimumsamplesizeof20
participants(10ineacharm)wereselected.FiveRCTswereidentified.
Table107summarisesthepopulation,intervention,comparisonandoutcomesforeachofthe
studies.
Table107:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Blennerhassett,
200427
Inpatientswith
aprimary
diagnosisof
strokeandwho
areabletowalk
10metreswith
supervision
(withorwithout
walkingaids)
Lowerlimbtraining:
therapistassistedfunctional
tasksincludedsittostand,
stepups,obstaclecourse,
plus
stretching/strengthening
exercise,andsome
endurancetraining
(stationary,bikesand
treadmill)for1hour/day,5
days/weekfor4weeks.
(N=15)
Therapist
assisted
functionaltasks
toimprove
reachand
grasp,handeye
coordination,
stretchingand
strengthening
for1hour/day,
5days/weekfor
4weeks.(N=15)
Higgins,
2006111
Patientswitha
firstor
recurrentstroke
(lessthan1year
poststrokeat
studyentry);
whohadthe
abilitytowalk
10m
independently
usinganaideor
orthosis
with/without
supervision.
Armtraining:
90minutepersession(total
18sessions)3timesaweek
for6weekswithatherapist
torepetitivelyperformtasks
thatthepatientsfound
difficult.Taskswere
changedortheirlevelof
difficultywasincreased
whenpatientsmaximised
theirperformance.
(N=47)
18sessionsof
walking
intervention
consistedof10
functionaltasks
3timesaweek
for6weeks.
(N=44)
Kwakkel,
1999140
Severely
disabled
patientswith
primaryfirst
everstroke
(within14days
afterstroke
onset);withan
inabilitytowalk
atfirst
assessment.
Upperlimbtraining:
functionalexercisesthat
facilitatedforcedarmand
handactivity.(N=33)
Lowerlimbtraining:
emphasisonachieving
stabilityandimprovinggait
velocity(N=31).
Bothinterventionswere
assistedbytherapistsfor30
minutes,5days/weekfora
totalof20weeks,plus
1.5hour/weekactivitiesof
dailylivingtrainingbyan
occupationaltherapist.
Armandleg
were
immobilised
withan
inflatable
pressuresplint
(30min,5
days/week).
(N=37)
NationalClinicalGuidelineCentre,2013.
462
Motor
assessment
Scale(MAS)
hand
6minute
walktest(m)
Timeupand
gotest(sec)
9hole
pegtest
Action
Research
Armtest
(ARAT)
10metre
timed
walkingtest:
comfortable
and
maximum
walking
speed
(m/sec)
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
Salbach,
2004229
Patientswith
walkingdeficit
withinoneyear
ofafirstor
recurrent
stroke;ableto
walk
independently
usinganaideor
orthosis
with/without
supervision.
Lowerlimbtraining:10
walkingrelatedtasks(total
18sessions)supervisedbya
physicaloroccupational
therapist,for3times/week
for6weeks,inrehabilitation
orhospitalsetting.(N=44)
Functional
upperextremity
tasks(total18
sessions/3
timesperweek
for6weeks)
thatweredone
whilesittingand
patientswere
recommended
topractiseat
home.(N=47)
Patientswith
recentfirsttime
stroke(2to35
dayspostonset)
frominfarction
intheanterior
circulation.
Upperlimbtaskfunctional
trainingfor1hour/day,5
days/weekfor4weeks:
repetitivepracticeoftasks
withinthelevelofavailable
voluntarymotion.Alltasks
aredesignedtobestandard,
repeatable,andtohave
somefunctionalgoal(for
examplepointing,grasping
andstirring).(N=22)
Winstein,2004285
NationalClinicalGuidelineCentre,2013.
463
Muscle
facilitation
exercises,
neuromuscular
electric
stimulation
applied
primarilyfor
shoulder
subluxation,
stretching
exercises,
activitiesof
dailyliving(self
carewherethe
upperlimbwas
usedasan
assist)and
caregiver
training.(N=21)
OUTCOMES
6minute
walktest
(m)
5metre
timed
walk:
comfortable
and
maximum
walking
speed(m/s)
Timedup&
gotest(sec)
FuglMeyer
Assessment:
rangeofmotion
pain
sensory
motorfunction
StrokeRehabilitation
Movement
Comparison:lowerlimbtraining(repetitivetaskorfunctional)versususualcare
Table108:Lowerlimbtraining(repetitivetaskorfunctional)versususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Limitation
s
Inconsistenc
y
Indirectnes
s
Imprecision
Lowerlimb
training
Mean(SD)or
median(IQR)
Effect
Usual
care
Mean
(SD)or Mean
median differenc
(IQR)
e(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
6minutewalktest(m)(posttreatmenteffect)(Betterindicatedbyhighervalues)
Blennerhassett
200427;Salbach
2004229
RCTs
single
blinded
Serious
limitations
(a,b)
Noserious
Noserious Serious
inconsistency indirectness imprecision(
c)
Blennerhassett
:404(101)
Salbach:
249(136)
Blenne
rhasset
t:288
(124)
Salbach
:
209
(132)
64.09
(18.52,
109.65)
MD64.09
higher
(18.52to
109.65
higher)
Low
416(171)
313
(154)
103(
13.46,
219.46)
MD103
higher
(13.46
lowerto
219.46
higher)
Low
Blennerhassett
:11.5(3.8)
Salbach:
Blenne
rhasset
t:19.1
6.23(
12.22,
0.25)
MD6.23
lower
(12.22low
Low
6minutewalktest(m)(6monthsfollowup)(Betterindicatedbyhighervalues)
Blennerhassett
200427
RCT
single
blinded
Serious
limitations
(a)
Noserious
Noserious Serious
inconsistency indirectness imprecision(
c)
Timedupandgotest(sec)(posttreatmenteffect)(Betterindicatedbylowervalues)
Blennerhassett
200427;Salbach
2004229
RCTs
single
blinded
Serious
limitations
(a)
Noserious
Noserious Serious
inconsistency indirectness imprecision(
d)
NationalClinicalGuidelineCentre,2013.
464
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitation
s
Inconsistenc
y
Indirectnes
s
Imprecision
Lowerlimb
training
Mean(SD)or
median(IQR)
23.2(20.6)
Usual
care
Mean
(SD)or Mean
median differenc
(IQR)
e(95%CI)
(14.4)
Salbach
:
27.1
(27.1)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
erto0.25
lower)
21.3
(30.3)
10.50(
26.0,5.0)
MD10.50
lower
(26.0
lowerto
5.0
higher)
Low
Kwakke
l:0.37
(0.41)
Salbach
:
0.64
(0.37)
0.48
(0.16,
0.79)
SMD0.48
higher
(0.16
higherto
0.79
higher)
Moderate
Kwakke
l:0.52
(0.58)
0.45
(0.13,
0.77)
SMD0.45
higher
(0.13
higherto
Moderate
Confidence
(ineffect)
Timedupandgotest(sec)(6monthsfollowup)(Betterindicatedbylowervalues)
Blennerhassett
200427
RCT
single
blinded
Serious
limitations
(a)
Noserious
Noserious Serious
inconsistency indirectness imprecision(
d)
10.8(4.5)
5and10metretimedwalk:comfortablespeed(m/sec)(posttreatmenteffect)(Betterindicatedbyhighervalues)
Kwakkel1999140
Salbach2004229
RCTs
single
blinded
Serious
limitations
(a,b)
Noserious
Noserious Noserious
inconsistency indirectness imprecision
Kwakkel:
0.65(0.46)
Salbach:
0.78(0.40)
5and10metretimedwalk:maximumspeed(m/sec)(posttreatmenteffect)(Betterindicatedbyhighervalues)
Kwakkel1999140
Salbach2004229
RCTs
single
blinded
Serious
limitations
(a,b)
Noserious
Noserious Noserious
inconsistency indirectness imprecision
Kwakkel:
0.88(0.66)
Salbach:
NationalClinicalGuidelineCentre,2013.
465
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitation
s
Inconsistenc
y
Indirectnes
s
Imprecision
Lowerlimb
training
Mean(SD)or
median(IQR)
0.99(0.56)
Usual
care
Mean
(SD)or Mean
median differenc
(IQR)
e(95%CI)
Salbach
:
0.80
(0.49)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
0.77
higher)
0.44
(0.44)
0.19(
0.03,
0.41)
MD0.19
higher
(0.03
lowerto
0.41
higher)
Low
0.57
(0.60)
0.28(
0.02,
0.58)
MD0.28
higher
(0.02
lowerto
0.58
higher)
Low
Confidence
(ineffect)
10metretimedwalking:comfortablespeed(m/sec)(6monthsfollowup)(Betterindicatedbyhighervalues)
Kwakkel1999140
RCT
single
blinded
Serious
limitations
(a)
Noserious
Noserious Serious
inconsistency indirectness imprecision
(e)
0.63(0.47)
10metretimedwalking:maximumspeed(m/sec)(6monthsfollowup)(Betterindicatedbyhighervalues)
Kwakkel1999140
RCT
single
blinded
Serious
limitations
(a)
Noserious
Noserious Serious
inconsistency indirectness imprecision
(e)
0.85(0.65)
(a)
Unclearallocationconcealment
7patientswithdrawnfromtrial;4patientswithmissingbaseline/followupdata(Salbach2004).
(c)
ConfidenceintervalcrossedthelowerlimitofagreedMID(28m)
(d)
MeandifferencedidnotreachtheagreedMIDof10sec.
(e)
ConfidenceintervalscrossedoneendofdefaultMID.
(b)
NationalClinicalGuidelineCentre,2013.
466
StrokeRehabilitation
Movement
Comparison:Upperlimbtraining(repetitivetaskorfunctional)versususualcare
Table109:
Upperlimbtraining(repetitivetaskorfunctional)versususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Absolute
effect/
Mean
Differenc
e(MD)
Confidence
(95%CI)
orpvalue (ineffect)
Upperlimb
training
Mean
(SD)/median
(IQR)
Usualcare
Mean
(SD)/
median
(IQR)
1(1)
1(1)
0.00(
0.41,0.41)
MD0.00
(0.41
lowerto
0.41
higher)
Moderate
6(56)
6(56)
(c)
(d)
Moderate
(b)
6(4.26.0)
6(36)
(c)
(d)
Moderate
(b)
6(4.26.0)
6(36)
(c)
p<0.01(e) Moderate
(b)
Mean
difference
(95%CI)
9holepegtest(1monthsfollowup)(Betterindicatedbyhighervalues)
Higgins
2006111
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Motorassessmentscale(MAS)hand(posttreatmenteffect)(Betterindicatedbyhighervalues)
Blennerhasset
t200427
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
(b)
Motorassessmentscale(MAS)hand(6monthsfollowup)(Betterindicatedbyhighervalues)
Blennerhasset
t200427
RCT
single
blinded
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
(b)
ActionResearchArmtest(posttreatmenteffect)(Betterindicatedbyhighervalues)
Kwakkel
1999140
RCT
single
blinded
Serious
limitations(
e)
Noserious
inconsistency
Noserious
indirectness
(b)
NationalClinicalGuidelineCentre,2013.
467
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
orpvalue (ineffect)
Upperlimb
training
Mean
(SD)/median
(IQR)
Usualcare
Mean
(SD)/
median
(IQR)
Mean
difference
(95%CI)
4(038)
0(02.25)
(c)
p<0.001(e
)
Moderate
(b)
1.9(2.02)
0.6(1.93)
1.30(
2.48,
0.12)
MD1.30
lower
(2.48
lowerto
0.12
lower)
Verylow
1.6(2.8)
0.6(1.79)
1.00(
2.40,0.40)
MD1.00
lower
(2.40
lowerto
0.40
higher)
Verylow
0.75(2.99)
0.75(1.33)
0.00(
1.37,1.37)
MD0.00
(1.37
lowerto
1.37
higher)
Verylow
ActionResearchArmtest(6monthsfollowup)(Betterindicatedbyhighervalues)
Kwakkel
1999140
RCT
single
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
(b)
FuglMeyerAssessmentrangeofmotion(posttreatmenteffect)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(g)
FuglMeyerAssessmentpain(posttreatmenteffect)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(g)
FuglMeyerAssessmentsensory(posttreatmenteffects)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(g)
NationalClinicalGuidelineCentre,2013.
468
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Imprecision
Upperlimb
training
Mean
(SD)/median
(IQR)
Usualcare
Mean
(SD)/
median
(IQR)
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
orpvalue (ineffect)
FuglMeyerAssessmentmotorfunction(posttreatmenteffect)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(h)
16.5(13.74)
9.05(7.6)
7.45(0.85,
14.05)
MD7.45
higher
(0.85
lowerto
14.05
higher)
Low
0.46(2.76)
0.33(1.45) 0.13(
1.44,1.18)
MD0.13
lower
(1.44
lowerto
1.18
higher)
Verylow
1.23(2.42)
1.00(2.88) 0.23(
1.82,1.36)
MD0.23
lower
(1.82
lowerto
1.36
higher)
Verylow
0.69(2.36)
0.07(1.03)
MD0.62
higher
Verylow
FuglMeyerAssessmentrangeofmotion(9monthsfollowup)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
,i)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(g)
FuglMeyerAssessmentpain(9monthsfollowup)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
,i)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(g)
FuglMeyerAssessmentsensory(9monthsfollowup)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
NationalClinicalGuidelineCentre,2013.
469
0.62(
0.46,1.70)
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
,i)
Inconsistency
Indirectness
Imprecision
(g)
Upperlimb
training
Mean
(SD)/median
(IQR)
Usualcare
Mean
(SD)/
median
(IQR)
Mean
difference
(95%CI)
8.33
(11.26)
2.56(
7.73,2.61)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
orpvalue (ineffect)
(0.46
lowerto
1.70
higher)
FuglMeyerAssessmentmotorfunction(9monthsfollowup)(Betterindicatedbyhighervalues)
Winstein
2004285
RCT
unblinded
Serious
limitations(f
,i)
(a)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(g)
5.77(4.49)
Unclearallocationconcealment(Blennerhassett2004)
Imprecisioncouldnotbeassessedasresultswerepresentedinmediananditsinterquartilerangeandcouldnotbemetaanalysed.
(c)
Resultswerepresentedasmedian(IQR)andcouldntestimaterelative/absoluteeffect.
(d)
AuthorsreportedthattherewasnosignificantdifferenceontheMAS(hand)betweenthelowerlimbtraininggroupandtheusualgroup.
(e)
Pvalueasreportedbyauthors
(f)
Studywasnotblinded,unclearrandomizationandinadequateallocationconcealment.
(g)
ConfidenceintervalcrossedbothendsofagreedMID(differenceby10%).
(h)
MeandifferenceanditsconfidenceintervalsdidnotreachtheagreedMID(differenceby10%ofthescale).
(i)
Highrateoflosstofollowupat9months(7inthefunctionaltaskgroupand5instandardcaregroup).
(b)
NationalClinicalGuidelineCentre,2013.
470
MD2.56
lower
(7.73
lowerto
2.56
higher)
Verylow
StrokeRehabilitation
Movement
13.7.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingrepetitivetasktrainingwithusualcarewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
IntheRCTsincludedintheclinicalreviewasubstantialdifferenceintermsofpersonneltimewasnot
seenbetweenusualcareandrepetitivetasktraining.Somenegligiblecostswouldbelinkedwiththe
useofcardsthatpatientswereaskedtomanipulate(forexample,inHiggins2002111).However,the
GDGnotedthatthiswasduetothestudiesmatchingtheintensityofinput;inreallifeitwas
expectedthatrepetitivetasktrainingmightinvolvesomeadditionaltherapytimeorthatcarers
wouldbetrainedtoassist.
13.7.1.3
Evidencestatements
Clinicalevidencestatements
Twostudies27,229of121participantsshowedthattherewasastatisticallysignificantimprovementin
locomotorperformanceassessedbythe6minutewalktest(m)inthegroupthatreceivedmobility/
lowerlimbtraining,comparedwiththeusualcaregroupattheendofthetreatment(LOW
CONFIDENCEINEFFECT).
Onestudy27of30participantsshowednosignificantdifferenceinlocomotorperformanceassessed
bythe6minutewalktest(m)betweenthemobility/lowerlimbtraininggroupandtheusualcare
groupat6monthsfollowup(LOWCONFIDENCEINEFFECT).
Twostudies27,229of121participantsshowedastatisticallysignificantimprovementintheTimedUp
andGoTest(sec)forthegroupreceivedthelowerlimbtrainingcomparedtotheusualcaregroupat
theendofthetreatment(LOWCONFIDENCEINEFFECT).
Onestudy27of30participantsshowednosignificantdifferenceinlocomotorperformanceassessed
bytheTimedUpandGoTest(sec)betweenthemobility/lowerlimbtraininggroupandtheusual
caregroupat6monthsfollowup(LOWCONFIDENCEINEFFECT).
Twostudies140,229of159participantsfoundasignificantdifferenceincomfortableandmaximum
walkingspeedmeasuredby5and10mtimedwalk(m/sec)betweenthosewhoreceived
mobility/lowerlimbtrainingandtheusualcaregroupattheendofthetreatment(MODERATE
CONFIDENCEINEFFECT).
Onestudy140of68participantsshowednosignificantdifferenceincomfortableandmaximum
walkingspeedmeasuredby5and10mtimedwalk(m/sec)betweenthosewhoreceived
mobility/lowerlimbtrainingandtheusualcaregroupat6monthsfollowup(LOWCONFIDENCEIN
EFFECT).
Onestudy111of91participantsfoundnosignificantdifferenceinthe9holepegtestscoresbetween
thearmtraininggroupandtheusualcaregroupattheendofthetreatment(MODERATE
CONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
471
StrokeRehabilitation
Movement
Onestudy285of33participantsshowedthatthestandardcaregroupwasassociatedwitha
statisticallysignificantimprovementintheFuglMeyerassessment(rangeofmotion),comparedwith
thosewhoreceivedfunctionaltaskarmtrainingattheendoftreatment(VERYLOWCONFIDENCEIN
EFFECT).
Onestudy285of33participantsshowednosignificantdifferencebetweenthefunctionaltaskarm
traininggroupandtheusualcaregroupattheendofthetreatmentonthefollowingoutcomes:
FuglMeyerassessment(pain)(VERYLOWCONFIDENCEINEFFECT)
FuglMeyerassessment(sensory)(VERYLOWCONFIDENCEINEFFECT)
Onestudy285of33participantsshowedthatthefunctionaltaskarmtraininggroupwasassociated
withastatisticallysignificantimprovementintheFuglMeyerassessment(motorfunction)compared
withthosewhoreceivedusualcareattheendofthetreatment(LOWCONFIDENCEINEFFECT).
Onestudy285of33participantsfoundnosignificantdifferencebetweenthefunctionaltaskarm
traininggroupandtheusualcaregroupat9monthsfollowuponthefollowingoutcomes:
FuglMeyerassessment(rangeofmotion(VERYLOWCONFIDENCEINEFFECT)
FuglMeyerassessment(pain)(VERYLOWCONFIDENCEINEFFECT)
FuglMeyerassessment(sensory)(VERYLOWCONFIDENCEINEFFECT)
FuglMeyerassessment(motorfunction)(VERYLOWCONFIDENCEINEFFECT)
Evidencestatementscouldnotbeproducedforthefollowingoutcome(s)asresultswerenot
presentedinawaythatenabledthesizeoftheinterventionseffecttobeestimated:
Actionresearcharmtest(upperlimbtrainingversusstandardcare)140
Motorassessmentscale(hand)(upperlimbtrainingversusstandardcare)27
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.7.2
Recommendationsandlinktoevidence
99.Offerpeoplerepetitivetasktrainingafterstrokeonarangeof
tasksforupperlimbweakness(suchasreaching,grasping,
pointing,movingandmanipulatingobjectsinfunctional
tasks)andlowerlimbweakness(suchassittostand
transfers,walkingandusingstairs).
Recommendations:
Relativevaluesofdifferent
outcomes
Alltherepetitivetasksconsideredinthereviewcomprisedofcircuittype
tasksforupperandlowerlimb.Othertypesofrepetitivetasksuchas
dressingpracticeortreadmillwerenotidentifiedbythesearch.
Theoutcomesofinterestforthelowerlimbwereanytimedwalk,
changeinwalkingdistanceandRivermeadMobilityIndex.Astherewere
avarietyofdifferenttimedwalkmeasurestheGDGrequestedthatthe
resultsbepresentedtogetherasnogreateremphasiswouldbeplaced
ononeoveranother.
TheGDGagreedtousethefollowingminimalimportantdifferences
(MIDs)publishedintheliteratureforthefollowingoutcomesreported;
20cm/secforthewalkingspeed,12and17pointsfortheaffected
dominantandnondominantsidesrespectivelywhenassessingthe
outcomeofActionResearchArm(ARAT)anddifferenceby10%ofthe
totalscalefortheFuglMeyerassessment(FMA)(pleaserefertoTable
5inthemethodologychapterformoredetailsonthepublishedsources
NationalClinicalGuidelineCentre,2013.
472
StrokeRehabilitation
Movement
oftheagreedminimalimportantdifferences).
Tradeoffbetweenclinical
benefitsandharms
TheGDGagreedtherewerenosignificantharmsassociatedwiththese
interventions.TheconsensusoftheGDGwasthatrepeatedpracticefor
bothupperandlowerlimbfunctionswaslikelytobebeneficialinterms
ofpatientsqualityoflifeandsocialinclusion.
Economicconsiderations
Norelevantcosteffectivenessevidencewasidentified.Theclinical
studiesincludedinthereviewdidnotindicateadifferenceinresource
usebetweenrepetitivetasktrainingandusualcare;however,itwas
consideredthatinrealitytheremaybesomeadditionalpersonneltime,
thereforecostsassociatedwithrepetitivetasktraining.TheGDG
consideredthesecoststobeoffsetbythebenefits.
Qualityofevidence
TheGDGnotedthattwoofthestudiesincludedinthereview(Salbach,
2004andHiggins2002)arethesamestudywithonereportingupper
extremityinterventionresultsandtheothermobility.
Twostudiesshowedasignificantimprovementinthe6minutewalktest
(Blennerhassett,2004andSalbach,200527,229)andtwointhetimed
metreswalks(Kwakkel,1999andSalbach,2004140,229).TheGDGnoted
thatthisimprovementwasfoundattheendofthestudy(post
treatmenteffect)butnotat6monthsfollowup(Kwakkel,1999140),
howeveritwouldusuallybeexpectedthatoncepatientswerewalking
thiswouldbemaintained.Thelowerlimboutcomesweregraded
betweenlowandmoderateduetostudylimitationsandimprecision
aroundtheeffectestimate.
TheWinstein(2004)studydemonstratedthatfunctionaltaskarm
trainingwasassociatedwithasignificantimprovementwithmotor
functioncomparedtousualcaregroupattheendoftreatment.
Confidenceintheresultsfortheseoutcomeswasgradedasverylowdue
tolimitationsinstudydesign(inadequateallocationconcealmentand
randomisation)andtheeffectestimatenotreachingtheminimal
importantdifferenceof10%ofthescale.Howeverthisimprovementin
themotorabilityoutcomeswasnotpreservedat9monthsfollowup.It
wasnotpossibletoestimatethesizeofeffectoftheupperlimb
interventionswithintheKwakkelandBlennerhassettstudiesasresults
werepresentedonlyasmedians(IQR).
Otherconsiderations
TheGDGconsideredthattheinterventionsusedforupperlimbwhich
includedtaskssuchasmanipulatingplayingcardsandhandwritingare
notrepresentativeofusualtherapeuticinterventions.HowevertheGDG
believedsuchtasksareimportantintermsofenablingthepatientto
undertakeactivitiesthemselvesandpromotingparticipationandself
esteem.TheGDGagreedthatalthoughusefulforsomepatientsthese
arehighleveltasks.TheGDGagreedthatthetrialsincludedthose
peoplewhoalreadyhadsomeupperlimbfunction,andthatthisisthe
groupwhoaremostlikelytobenefitfromtheinterventions.
13.8 Walkingtherapies:treadmillandtreadmillwithbodyweight
support
Therearetwotypesoftreadmilltrainingthatarecurrentlyusedtoassistwiththereeducationof
gaitfollowingastroke.Thefirstisaconventionaltreadmillthatrequiresthestrokesurvivorto
mobilisebearingthefullweightoftheirbody.Thesecondisatreadmillwithbodyweightsupport
thatallowsthestrokesurvivortomobilisewithoutrequiringthattheycarrythefullweightoftheir
body.Astheybecomestrongertheyareabletograduallyreducethebodyweightsupport.Theuse
oftreadmilltrainingbothwithandwithoutbodyweightsupporthasbeenshowntoassistwiththe
NationalClinicalGuidelineCentre,2013.
473
StrokeRehabilitation
Movement
reeducationofgaitfollowingstroke,asanadjuncttoconventionalphysiotherapy.Ithasnotbeen
demonstratedtobeofbenefitinsteadofroutinephysiotherapyintervention.
13.8.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessofall
treadmillversususualcareonimprovingwalking?
13.8.2
Evidencereview:Inpeopleafterstrokewhocanwalk,whatistheclinicalandcost
effectivenessoftreadmillplusbodysupportversustreadmillonlyonimprovingwalking?
13.8.2.1
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
Anytreadmilltraining(withorwithoutbodysupport)
Comparison
Usualcare(otherphysiotherapy)
Treadmillwithoutbodysupport
Outcomes
Walkingspeeds(5m/10m/30m)
Anytimedwalk
Walkingendurance
FunctionalIndependenceMeasure(FIM)
BarthelIndex
RivermeadMobilityIndex
Clinicalevidencereview
Searcheswereconductedforsystematicreviews(ofrandomizedcontrolledtrials(RCTs)andcohort
studies)andRCTsthatcomparedtheeffectivenessofalltreadmilltherapieswithusualcareto
improvewalkingforadultsandyoungpeople16orolderwhohavehadastroke.Onlystudieswitha
minimumsamplesizeof20participants(10ineacharm)andincludingatleast50%ofparticipants
withstrokewereselected.Sixteen(16)RCTswereidentified.Onestudy190includedtreadmill
trainingexercisewithbodysupportcomparedtousualcareandthreestudies15,115,272compared
treadmilltrainingexercisewithbodysupportwithtreadmilltrainingexercisewithoutbodysupport.
Alltheotherstudiescomparedtreadmillwithoutbodyweightsupportversususualcare.Table110
summarisesthepopulation,intervention,comparisonandoutcomesforeachofthestudiesincluded
intheclinicalevidencereview.
Table110:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Barbeau
200315
Strokepatients
(onset<6
monthsago)
referredina
hospitalfor
physical
rehabilitation
whocouldnt
walkinanormal
gaitpattern.
Treadmilltrainingwith
bodywhilean
overheadharness
supporteda
percentageoftheir
bodyweightwiththe
assistanceof1or2
therapistsasneeded.
Patientswere
providedupto40%
bodyweightsupport
Treadmilltraining
withoutabody
supportwiththe
assistanceof1or2
therapistsasneeded
(N=50).
Proportionof
participants
achievedover
groundwalking
speedover0.2m/s
Proportionof
participants
achievedover
groundwalking
enduranceover20m
NationalClinicalGuidelineCentre,2013.
474
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
atthebeginningofthe
training,andthat
percentagewas
progressively
decreasedastheir
abilitytowalk
improved(N=50)
OUTCOMES
Duncan
201173
Inpatientswho
hadstroke
within45days
beforestudy
entryandthe
abilityto
undergo
randomisation
within2months
afterstroke;
patients
experienced
residualparesis
inthelegbut
wereableto
walk3mwith
assistance.
Earlytreadmilltraining
withpartialbody
weightsupportand
manualassistanceas
neededfor2030
minutesat3.2kmper
hour,followedbya
progressiveprogram
ofwalkingoverground
for15minutes.Study
duration:1216weeks.
(N=139)
Homeexercise
programtask
specificwalking
program,managed
byaphysical
therapistinthe
home,withthegoals
ofenhancing
flexibility,rangeof
motioninjoints,
strengthofarmsand
legs,coordination,
andstaticand
dynamicbalance.
(N=126)
10metrewalktime
(m/sec)
6minutewalktest
(m)
Eich,200474
Patientswith
firsttimestroke
(<6weeksof
strokeonset)
referredto
inpatient
rehabilitation
centre.
Followuptime=
3months
Treadmilltrainingwith
nobodyweight
supportfor30minutes
andotherindividual
physiotherapyfor30
minutesfor6weeks
(N=25).
Usualcareincluded
60minutesof
individual
physiotherapydaily
for6weeks(N=25).
10metretimedwalk
(m/sec)
6minutewalktest
(m)
RivermeadMotor
AssessmentScore
Franceschini
,200986
Patientswith
subacutestroke
(<6weeksof
strokeonset)
whowere
unabletowalk.
Followuptime=
6months.
Treadmilltrainingwith
bodyweightsupport
for20minutes
followedby40
minutesof
conventionaltraining
for5timesaweekfor
20sessions.The
trainingshouldhave
beencompleted
within5weeksof
inclusioninthestudy.
(N=52)
Usualcareincluded
20sessionsof
conventional
treatment(consisting
ofovergroundgait
training)of60
minuteseach
session.5timesper
weekfor20sessions,
whichshouldhave
beencompleted
within5weeksof
inclusioninthe
study.(N=50)
10metretimedwalk
test(m/sec)
6minutewalktest
(m)
BarthelIndex
Kosak,
2000138
Patients
admittedto
inpatientstroke
rehabilitation
unit.(timesince
Partialbodyweight
supporttreadmill
training(with
overheadmotorised
hoistattachedtoa
Aggressivebracing
assistedwalking
(usinghemibarand
kneeanklefoot
orthosisifnecessary)
Walkingspeed(2
minutetestperiod)
(m/minute)
Walkingendurance
(m)
NationalClinicalGuidelineCentre,2013.
475
StrokeRehabilitation
Movement
STUDY
POPULATION
stroke(mean
days);39forthe
intervention
groupand40for
thecontrol
group).The
sampleincluded
someseverely
affectedpatients
with
hemiparesis
hemisensory
hemianopic
visualdeficits
INTERVENTION
parachutetypebody
harness;assistedby
physicaltherapistand
physicaltherapyaide)
forupto45minutes
astolerated,and45
minutesoftraditional
physicaltherapy5
daysaweek.(N=22)
COMPARISON
forupto45minutes
astolerated,and45
minutesof
traditionalphysical
therapy5daysa
week.(N=34)
Kuys2011139 Patientswith
firsttimestroke
(timesince
strokemean
days52forthe
intervention
groupand48for
thecontrol
group)
Patientswere
followedupat6
and18weeks
post
intervention.
Treadmilltrainingfor
30minutes(excluding
rests),threetimesa
weekfor6weeks,at
anintensityof4060%
heartratereserveora
BorgRatingof
PerceivedExertionof
1114.Alsoreceived
usualphysiotherapyof
approximatelyone
hourperday.(N=15)
Usualphysiotherapy Walkingendurance
ofapproximatelyone
(m)
hourperdayof
Comfortablewalking
comprehensive
speed(m/sec)
therapyusingatask
Fastwalkingspeed
orientedapproach
(m/sec)
targeting
impairmentsand
activitylimitations
specifictoeach
participant.(N=15)
Langhamme
r,2010143
Patientswith
strokeata
private
rehabilitation
centre.(<8
weeksofstroke
onset)
Nofollowup.
Treadmilltraining
exercisewithnobody
supportforupto30
minutesforfivedays
perweek.
Thetreadmillhad
handrailingstohold
onto,otherwisethere
werenosafety
precautions(N=21).
Outdoorwalking
6minutewalktest
exerciseincluded
(m)
walkingata
10metretimedwalk
comfortablespeed
(m/sec)
fivedaysaweekwith
theuseofordinary
assistivedevices
whennecessary.(N
=18)
Laufer,20011
47
Firststroke
patients(onset
ofstroke<=90
day)withan
abilitytowalkon
treadmillata
speedofatleast
0.2km/hr.with
minimalto
moderate
assistancefor2
minuteswithout
rest.
Nofollowup.
Treadmilltraining
exerciseconsistedof
ambulatingona
motordriventreadmill
whichwasadjustedto
thesubjects
comfortablewalking
speedfor5sessionsa
weekfor3weeks.
Actualwalkingtime
duringtraining
sessionsincluded4
minperdaythefirst
week,6minperdayin
2ndweekand8min
perdayin3rdweek.
(N=13)
Floorwalking
exerciseconsistedof
ambulatingonfloor
surfaceata
comfortablespeed
usingwalkingaids,
assistance,and
restingperiodsas
needed.Actual
walkingtime
included4minper
daythefirstweek,6
minperdayin2nd
weekand8minper
dayin3rdweek.
(N=12).
NationalClinicalGuidelineCentre,2013.
476
OUTCOMES
10metretimedwalk
(m/sec)
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
Luft,2008 Patientswith
chronic
hemipareticgait
(>=6months
aftertheirfirst
stroke).
Nofollowup.
Treadmilltraining
exerciseincluded
threeexercisesof40
minuteseachper
weekatanaerobic
intensityof60%of
heartratereserve,for
6months.(N=57)
Usualcareincluded
10metretimedwalk
13supervised
(m/sec)
traditionalstretching 6minutestimed
movements(actively
walk(m)
ifpossibleor
passively)onaraised
mattablewitha
therapistsassistance
for6months(N=56)
Nilsson,
2001190
Patientswitha
firststroke(<8
weeksfrom
onsetofstroke)
withresidual
hemiparesis
afterstroke.
Followuptime=
10months.
Treadmilltrainingwith
bodyweightsupport
wasprovidedfor30
minutesforfivedaysa
week.Thebodyweight
supportwasgradually
reducedasfastas
possibleasthegoal
wastoattainwalking
onthetreadmillwith
fullweightbearing.
(N=36)
Usualcareincluded
individualwalking
trainingbya
physiotherapistfor
30minutesfivedays
aweek.(N=37)
FIM
10metretimedwalk
(m/sec)
Olawale
2011193
Patients(from
anAfrican
population)with
chronicstroke
(324months)in
anoutpatient
stroke
rehabilitation
unitinatertiary
hospital;who
wereableto
walk10m
independently
withorwithouta
walkingaid.
Treadmillwalking
trainingand
conventional
physiotherapyfor12
weeks.(N=20)
Group1:
Overgroundwalking
exerciseand
conventional
physiotherapyfor12
weeks(N=20)
Group2:
Usualcare
conventional
physiotherapyonly
for12weeks(N=20)
10metretimedwalk
(m/sec)
6minutewalktest
(m)
Pohl,
2002208
Patientswith
hemiparesis
causedby
stroke,admitted
toapoststroke
inpatient
rehabilitation
centre(>4weeks
poststroke).
Nofollowup.
Structuredspeed
dependenttreadmill
training(STT);
12x30minofSTT,8x
45minconventional
physiotherapy(gait
trainingallowed)in
total12hoursof
treatment.
Thetotalwalking
distancevariedfrom
sessiontosession.
(N=20)
Usualcareincluded
12sessionsof45min
conventionalgait
training,8sessionsof
45minconventional
physiotherapy(gait
trainingallowed).
Totaldurationof
training:15hoursof
treatmentfor4
weeks.(N=20)
10metretimedwalk
(m/sec)
160
NationalClinicalGuidelineCentre,2013.
477
OUTCOMES
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
Visintin1998
272
Patients
admittedto
hospitalwith
stroke(<6
monthsago),not
walkingwitha
normalgait
patternandnot
ambulating
beforestroke.
Followuptime=
3months.
Treadmillgaittraining
whileanoverhead
harnesssupporteda
percentageoftheir
bodyweight.
Individualsinthebody
weightsupportgroup
wereprovidedupto
40%bodyweight
supportatthe
beginningoftraining,
andthepercentageof
bodyweightsupport
wasprogressively
decreasedasthe
subjectsgaitpattern
andabilitytowalk
improved.Body
weightsupport
treadmilltrainingwas
givenatafrequencyof
fourtimesperweek
for6weeks(nomore
than20minutesper
session).(N=50)
Thecontrolgroup
10metretimedwalk
receivedgaittraining
(m/sec)
onatreadmillwith
Walkingendurance
nobodyweight
(m)
support,i.e.while
bearingfullweight
ontheirlower
extremities.
Treadmillwithout
bodysupportwas
givenfor6weeksat
afrequencyoffour
timesperweek(no
morethan20
minutespersession).
(N=50)
Hoyer,2012
115
Patientsmainly
<6monthsafter
onsetofstroke,
useofwheel
chair,
dependenceon
assistancefor
walkingwithor
withoutwalking
aids,medically
stable,no
neurologicalor
orthopaedic
contraindication
sforwalking.
Treadmilltrainingwith
bodyweightsupport
(TTBWS),plus
conventionalgait
trainingandfunctional
trainingforaperiodof
minimum10weeks.
TTBWSwasdailyfor
the1st4weeks(20
sessions),andthen12
timesaweek(10
sessions)forthe
remaining6weeks.
(N=30)
Intensivegait
training(30min)and
functionaltraining
(30min)dailyfor
minimum10weeks.
(N=30)
Globas
201294
38patientswith
stroke(>6
months)aged
>60yearswith
residual
hemipareticgait
(atleast1clinical
signforparesis,
spasticityor
circumductionof
affectedleg
whilewalking);
abilitytowalkon
treadmillat
0.3km/hr.for3
minuteswith
Highintensityaerobic
treadmillexercise
(TAEX)for3months
(39sessions)starting
with1020minutesat
4050%heartrate
reserve(HRR)building
upto3050minutesat
6080%HRR
ConventionalCare
Sustainedwalking
Physiotherapy(13
ability(6minute
sessionsof1hour
walk).
each/week)including 10mtimedwalkat
passivemuscletone
comfortableand
regulatingexercises
maximalspeeds;
forupperandlower
extremity,balance
training
NationalClinicalGuidelineCentre,2013.
478
OUTCOMES
Functional
Independence
Measure(FIM)
10mwalktest(m/s)
6minwalktest(m)
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
handrailsupport.
COMPARISON
OUTCOMES
Kang
2012131
30Patientswith
hemiparetic
stroke6months
afterdiagnosis;
whocouldwalk
unaidedfor>15
minutes;without
visualdisability
orhemianopia;
MiniMental
State
Examination
score21or
higher;
Brunnstrum
stage>4.
Generalstretching
addingrangeof
motiontobothsides
oftrunk,armsand
legs.Allpatients
received
conventional
physiotherapy5
timesaweekfor4
weeks
10mwalktest,
6minutewalktest
Treadmilltraining3
timesaweekfor4
weeks,30minutes
eachdayspeed
increasedby0.1kmhr
eachtimepatients
couldwalkstablyfor
20seconds;2x15
minuteswith5minute
break
NationalClinicalGuidelineCentre,2013.
479
StrokeRehabilitation
Movement
Comparison:Treadmilltraining(withorwithoutbodysupport)versususualcare
Table111:Alltreadmilltraining(withorwithoutbodysupport)versususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness Imprecision
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
6minutewalktest(m)(acutestrokepatients)(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Eich200474
Langhamme
r2010143
RCTssingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Eich2004:
198.8
(81.1)
Langhamm
er2010:
320.6
(153.8)
Eich
2004:
164.4
(69.3)
Langham
mer
2010:
310.1
(164.4)
31.9
(8.18,
70.56)
MD31.19 Moderate
higher
(8.18
lowerto
70.56
higher)
Olawale
155.27
(66.37)
Globas
265.9
(189)
Kang
240.9
1.07
(17.60,
19.75)
MD1.07
lower
(17.60
lowerto
19.75
higher)
6minutewalktest(m)(chronicstrokepatients)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Olawale
2011193
Globas
201294Kang
2012131
RCTunclear
blinding
Veryserious
limitations(b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Olawale
145.32
(74.97)
Globas
332.1
(138)Kang
242.3(26)
NationalClinicalGuidelineCentre,2013.
480
Low
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
(22.4)
(c)
Franceschi
ni2009:
160(118
231)
Luft2008:
0.63(0.52
0.73)
Noserious
imprecision
(c)
Indirectness Imprecision
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Francesc
hini
2009:
170
(90.5
250)
Luft
2008:
0.57
(0.46
0.69)
(c)
(c)
Low(c)
224.8(90)
163
(70.2)
61.8
(16.48,
107.12)
MD61.8
higher
(16.48to
107.12
higher)
Moderate
217
(108.8
332.5)
210
(140
335)
(c)
(c)
High(f)
Confidence
(ineffect)
6minutewalktest(m)(posttreatment)(e)(Betterindicatedbyhighervalues)
2
Franceschini
200986
Luft2008160
RCTssingle
blinded
Veryserious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
6minutewalktest(m)(18weeksfollowup)(Betterindicatedbyhighervalues)
1
Eich200474
RCTssingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
6minutewalktest(m)(6monthsfollowup)(d)(Betterindicatedbyhighervalues)
1
Franceschini
200986
RCTssingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
481
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness Imprecision
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
10metretimedwalk(m/sec)(acutestrokepatients)(posttreatmenteffect)(Betterindicatedbyhighervalues)
4
Eich200474
Langhamme
r2010143
Laufer
2001147
Pohl2002208
RCTssingle
blinded
Veryserious
limitations(a,h)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Eich2004:
0.7(0.3)
Langhamm
er2010:
1.0(0.4)
Laufer
2001:
0.47(0.4)
Pohl2002:
1.63(0.8)
Eich
2004:
0.6
(0.22)
Langham
mer
2010:
0.9(0.5)
Laufer
2001:
0.33
(0.24)
Pohl
2002:
0.97
(0.64)
0.14(0.03,
0.26)
MD0.14
higher
(0.03
lowerto
0.26
higher)
Low
Olawale
9.28
(0.4)
Globas
0.87(0.6
2)
Kang0.5
0.41
(0.29,
0.53)
MD0.41
higher
(0.29to
0.53
higher)
Moderate
10metretimedwalk(m/sec)(chronicstrokepatients)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Olawale
2011193
Globas
201294Kang
2012131
RCTunclear
blinding
Serious
limitations(b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Olawale
6.71(0.6)
Globas
1.02(0.38)
Kang0.5
(0.2)
NationalClinicalGuidelineCentre,2013.
482
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
Indirectness Imprecision
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
(0.1)
Franceschi
ni2009:
0.5(0.3
0.9)
Luft2008:
0.82(0.69
0.95)
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Francesc
hini
2009:
0.6(0.3
0.9)
Luft
2008:
0.71
(0.58
0.84)
(g)
(g)
Moderate
(f)
Eich2004:
0.77(0.35)
Nilsson
2001:
0.7(0.3)
Eich
2004
0.58
(0.22)
Nilsson
2001:
0.8(0.4)
0.07
(0.06,
0.19)
MD0.07
higher
(0.06
lowerto
0.19
higher)
Moderate
0.7(0.3
1.0)
0.8(0.5
1.1)
(g)
(g)
High(f)
Confidence
(ineffect)
10metretimedwalktest(m/sec)(posttreatmenteffect)(e)(Betterindicatedbyhighervalues)
2
Franceschini
200986
Luft,2008160
RCTssingle
blinded
Serious
limitations(e)
Noserious
inconsistency
Noserious
indirectness
(f)
10metretimedwalk(m/sec)(mean29weeksfollowup)(Betterindicatedbyhighervalues)
2
Eich200474
Nilsson
2001190
RCTssingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
10metretimedwalk(m/sec)(6monthsfollowup)(c)(Betterindicatedbyhighervalues)
1
Franceschini
200986
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
(f)
NationalClinicalGuidelineCentre,2013.
483
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
Serious
imprecision
(i)
11(9.38)
11(5.83)
0.00(
4.38,4.38)
MD0.00
higher
(4.38
lowerto
4.38
higher)
Verylow
Serious
imprecision(i
)
74(70.36)
72
(64.14)
2.00(
34.46,
38.46)
MD2.00
higher
(34.46
lowerto
38.46
higher)
Verylow
Veryserious
imprecision(
k)
284(139)
279
(163)
5.00(
106.86,11
6.86)
MD5
higher
(106.86
lowerto
116.86
higher)
Low
Veryserious
imprecision(
k)
291(157)
293
(180)
2.00(
137.14,
133.14)
MD2
lower
(137.14
Low
Indirectness Imprecision
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
Walkingspeed(2minutes)(m/min)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Kosak
2000138
RCTssingle
blinded
Veryserious
limitations(b)
Noserious
inconsistency
Serious
indirectness
(j)
Walkingendurance(m)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Kosak
2000138
RCTsingle
blinded
Veryserious
limitations(b)
Noserious
inconsistency
Serious
indirectness
(j)
Walkingendurance(m)(6weeksfollowup)(Betterindicatedbyhighervalues)
1Kuys
2011139
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Walkingendurance(m)(18weeksfollowup)(Betterindicatedbyhighervalues)
1Kuys
2011139
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
484
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Effect
Limitations
Inconsistency
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
0.63(0.3)
0.68
(0.37)
0.05(0.3, MD0.05
0.2
lower
(0.3
lowerto
0.2
higher)
Low
Veryserious
imprecision(
k)
0.72(0.35)
0.66
(0.41)
0.06(
0.25,0.37)
MD0.06
higher
(0.25
lowerto
0.37
higher)
Low
Veryserious
imprecision(
k)
0.86(0.43)
0.86
(0.47)
0(0.33,
0.33)
MD0
higher
(0.33
lowerto
0.33
higher)
Low
Indirectness Imprecision
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
lowerto
133.14
higher)
Confidence
(ineffect)
Comfortablewalkingspeed(m/sec)(6weeksfollowup)(Betterindicatedbyhighervalues)
1Kuys
2011139
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision(
k)
Comfortablewalkingspeed(m/sec)(18weeksfollowup)(Betterindicatedbyhighervalues)
1Kuys
2011139
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Fastwalkingspeed(m/sec)(6weeksfollowup)(Betterindicatedbyhighervalues)
1Kuys
2011139
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Fastwalkingspeed(m/sec)(18weeksfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
485
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Effect
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
Mean
difference
(95%CI)
Veryserious
imprecision(
k)
0.91(0.46)
0.82
(0.49)
0.09(
0.29,0.47)
MD0.09
higher
(0.29
lowerto
0.47
higher)
Low
Serious
imprecision(l
)
81.9(9.6)
80.3
(15.9)
1.6(4.96,
8.16)
MD1.6
higher
(4.96
lowerto
8.16
higher)
Low
Serious
imprecision(
m)
32(5)
31.5
(6.6)
0.50
(2.44,
3.44)
MD0.50
higher
(2.44
lowerto
3.44
higher)
Low
(f)
3(24)
2(13.5)
(g)
(g)
High(f)
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
1
Kuys2011139
RCTsingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
FIMmotoritems(10monthsfollowup)(Betterindicatedbyhighervalues)
1
Nilsson
2001190
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
FIMcognitiveitems(10monthsfollowup)(Betterindicatedbyhighervalues)
1
Nilsson
2001190
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Franceschini
200986
RCTssingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
486
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
1
Franceschini
200986
Effect
All
treadmill
training
Mean(SD)
ormedian
(range)
Usual
care
Mean
(SD)or
median
(range)
(f)
4(45)
(f)
Design
Limitations
Inconsistency
Indirectness Imprecision
RCTssingle
blinded
Noserious
limitation
Noserious
inconsistency
Noserious
indirectness
Mean
difference
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
4(35)
(g)
(g)
High(f)
11(11)
11(10
11)
(g)
(g)
Low(f)
11(11)
11(10
11)
(g)
(g)
Low(f)
Rivermeadgrossfunction(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Eich200474
RCTsingle
blinded
Veryserious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
RivermeadMotorAssessmentScore(18weeksfollowup)(Betterindicatedbyhighervalues)
1
Eich200474
RCTsingle
blinded
Veryserious
limitations(a)
Noserious
inconsistency
(a)
Noserious
indirectness
(f)
Inadequaterandomisation(Eich2004;Langhammer2010;Laufer2001;Nilsson2001;Kang2012)
Unclearblinding(Globas2012),randomisationmethodandallocationconcealment(Olawale2011;Kossak2000)
(c)
Thisoutcomeispresentedseparatelyforthestudiesthatcouldnotbemetaanalysedastheresultswerenotpresentedwiththemeanandstandarddeviation
(d)
Highdropoutrate35%inexperimentalgroupand39%incontrolgroup(Luft2008)
(e)
Imprecisioncouldnotbeassessedbecauseonlymedianandinterquartilerangesofdatawerereported.
(f)
Relativeandabsoluteeffectcouldnotbeassessedbecausemedianandinterquartilerangesofdatareported
(g)
Unclearallocationconcealment(Laufer2001;Pohl2002)
(h)
ConfidenceintervalcrossedoneendofdefaultMID.
(i)
Controlgroupreceivedaggressivebracingassistedwalkingexercise.
(j)
ConfidenceintervalcrossedbothendsofdefaultMID.
(k)
ConfidenceintervalcrossedoneendofagreedMID(17points).
(l)
ConfidenceintervalcrossedoneendofagreedMID(3points).
(b)
NationalClinicalGuidelineCentre,2013.
487
StrokeRehabilitation
Movement
Comparison:Earlytreadmilltrainingexercise(2monthsafterstroke)withbodyweightsupportversushomeexerciseprogram
Table112:Earlytreadmilltrainingexercise(2monthsafterstroke)withbodyweightsupportversushomeexerciseprogramClinicalstudy
characteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Early
body
weight
support
treadmill
Mean
(SD)
Effect
Home
exercise
program
Mean
(SD)
Mean
difference
(95%CI)
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
10metretimedwalktest(m/sec)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Duncan
201173
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(a)
0.25(0.21
)
0.23(0.20
)
0.02
(0.03,
0.07)
MD0.02
higher
(0.03
lowerto
0.07
higher)
Moderate
Serious
imprecision
(a)
0.23
(0.20)
0.25(0.22
)
0.02
(0.07,
0.03)
MD0.02
lower
(0.07
lowerto
0.03
higher)
Moderate
Serious
imprecision
(b)
81.8
(62.8)
75.9
(69.3)
5.9
(10.08,
21.88)
MD5.9
higher
(10.08
lowerto
21.88
higher)
Moderate
10metretimedwalktest(m/sec)(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Duncan
201173
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
6minutewalktest(m)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Duncan
201173
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
488
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Early
body
weight
support
treadmill
Mean
(SD)
Effect
Home
exercise
program
Mean
(SD)
Mean
difference
(95%CI)
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
6minutewalktest(m)(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Duncan
201173
RCTSingle
blinded
Noserious
limitations
Noserious
inconsistency
Noserious
indirectness
Veryserious 73.2
imprecision (69.4)
(b)
85.2
(72.9)
12.0
(29.18,
5.18)
(a)
ConfidenceintervalbothendsofdefaultMID.
MeandifferencedidnotreachtheagreedMIDof28m.
(b)
NationalClinicalGuidelineCentre,2013.
489
MD12.0
lower
(29.18
lowerto
5.18
higher)
Moderate
StrokeRehabilitation
Movement
Comparison:Treadmilltrainingexercisewithbodysupportversustreadmilltrainingexercisewithoutbodysupport
Table113:TreadmillplusbodyweightsupportversustreadmillonlyClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Body
weight
support
treadmill
Mean
(SD),
Frequenc
y(%)
Effect
Treadmill
only
Mean
(SD),
Frequenc
y(%)
Mean
difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
10metretimedwalk(m/sec)(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Hoyer2012
115
,
Visintin
1998272
RCTSingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Hoyer:
0.33
(0.24)
Visintin:
0.34
(0.26)
Hoyer:
0.32
(0.22)
Visintin:
0.25
(0.36)
0.04(
0.05,
0.13)
MD0.04
higher
(0.05
lowerto
0.13
higher)
Moderate
Hoyer:
0.4(0.27)
Visintin:
0.52
(0.32)
Hoyer:
0.36
(0.24)
Visintin:
0.30
(0.29)
0.11
(0.01,
0.21)
MD0.11
higher
(0.01to
0.21
higher)
Verylow
105
(112.2)
42.4(
8.75,
93.55)
MD42.4
higher
(8.75
lowerto
93.55
Verylow
10metretimedwalk(m/sec)(1112weeksfollowup)(Betterindicatedbyhighervalues)
2
Hoyer2012
115
,
Visintin
1998272
RCTSingle
blinded
Veryserious
limitations(a,
b)
Serious
inconsistency(
i)
Noserious
indirectness
Noserious
imprecision
Walkingendurance(m)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Visintin
1998272
RCTSingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Veryserious 147.4
imprecision( (119.35)
c)
NationalClinicalGuidelineCentre,2013.
490
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Body
weight
support
treadmill
Mean
(SD),
Frequenc
y(%)
Effect
Treadmill
only
Mean
(SD),
Frequenc
y(%)
Mean
difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
higher)
Confidence
(ineffect)
Walkingendurance(m)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Visintin
1998272
RCTSingle
blinded
Veryserious
limitations
(a,b)
Noserious
inconsistency
Noserious
indirectness
Veryserious 202.4
imprecision (122.78)
(c)
152.3
(141)
50.10(
22.82,
123.02)
MD50.1
higher
(22.82
lowerto
123.02
higher)
Verylow
13/50
(26%)
RR1.31
(0.71to
2.4)
81more
Verylow
per1000
(from75
fewerto
364more)
Proportionofparticipantsachievedovergroundwalkingspeedover0.2m/s(posttreatmenteffect)
1Barbeau15
RCTDouble
blinded
Serious
limitations(d)
Noserious
inconsistency
Noserious
indirectness
Veryserious 17/50
imprecision
(e)
Proportionofparticipantsachievedovergroundwalkingenduranceover20m(posttreatmenteffect)
15
1Barbeau
RCTDouble
blinded
Serious
limitations(d)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(f)
24/50
(48%)
19/50
(38%)
RR1.26
(0.8to
1.99)
99more
Low
per1000
(from76
fewerto
376more)
Serious
imprecision(
g)
108.39
(76.84)
98.03
(61.90)
10.36(
24.95,
45.67)
MD10.36
higher
(24.95
6metrewalktest(m)(posttreatmenteffect)(betterindicatedbyhighervalues)
1
Hoyer2012
115
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
491
Low
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Body
weight
support
treadmill
Mean
(SD),
Frequenc
y(%)
Effect
Treadmill
only
Mean
(SD),
Frequenc
y(%)
108.39
(76.84)
98.03
(61.90)
Mean
difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
lowerto
45.67
higher)
Confidence
(ineffect)
6metrewalktest(m)(11weeksfollowup)(Betterindicatedbyhighervalues)
1
Hoyer2012
115
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision(
g)
22.23(
22.91,
67.37)
MD22.23
higher
(22.91
lowerto
67.37
higher)
Low
5.35(
19.92,
9.22)
MD5.35
lower
(19.92
lowerto
9.22
higher)
Verylow
6.83(
21.35,
7.69)
MD6.83
lower
(21.35
lowerto
7.69
higher)
Verylow
FunctionalIndependenceMeasure9,shortertransfer(sec):(Postintervention)(Betterindicatedbylowervalues)
1
Hoyer2012
115
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Very
serious
imprecision(
h)
40.32
(28.98)
45.67
(28.61)
FunctionalIndependenceMeasure9,shortertransfer(sec):(11weeksfollowup)(Betterindicatedbylowervalues)
1
Hoyer2012
115
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Very
serious
imprecision(
h)
33.02
(25.07)
39.85
(31.89)
NationalClinicalGuidelineCentre,2013.
492
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Body
weight
support
treadmill
Mean
(SD),
Frequenc
y(%)
Effect
Treadmill
only
Mean
(SD),
Frequenc
y(%)
Mean
difference
/Risk
Ratio
(95%CI)
Absolute
effect/
Mean
Differenc
e(MD)
(95%CI)
Confidence
(ineffect)
FunctionalIndependenceMeasure13,StairsNinestepsupdowntransfer(sec):(Postintervention)(Betterindicatedbylowervalues)
1
Hoyer2012
115
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Very
serious
imprecision(
h)
61.31
(43.04)
80.79
(61.55)
19.48(
46.36,
7.40)
MD19.48
lower
(46.36
lowerto
7.40
higher)
Verylow
FunctionalIndependenceMeasure13,StairsNinestepsupdowntransfer(sec):(11weeksfollowup)(Betterindicatedbylowervalues)
1
Hoyer2012
115
RCTsingle
blinded
Serious
limitations(a)
Noserious
inconsistency
Noserious
indirectness
Very
serious
imprecision(
h)
48.4
(31.82)
67.03
(51.70)
(a)
Unclearallocationconcealment.
Highdropoutrate(34%)duringfollowupforVinsintin
(c)
ConfidenceintervalcrossedbothendsofdefaultMID.
(d)
Unclearallocationconcealment.28%ofthecontrolgroupdidnotcompletethestudy.
(e)
ConfidenceintervalcrossedbothendsofdefaultMID(0.75,1.25).
(f)
ConfidenceintervalcrossedtheupperlimitofMID(1.25).
(g)
ConfidenceintervalcrossesoneendofagreedMID(28m)
(h)
ConfidenceintervalcrossesbothendsofdefaultMID(0.5)
(i)
Heterogeneity:I=64%
(b)
NationalClinicalGuidelineCentre,2013.
493
18.63[
40.35,
3.09]
MD18.63
lower
(40.35
lowerto
3.09
higher)
Verylow
StrokeRehabilitation
Movement
13.8.2.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationswereidentifiedcomparingtreadmilltrainingwithusualcare,or
treadmilltrainingwithbodysupportwithtreadmilltrainingwithoutbodysupport.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
Inmoststudiesidentifiedintheclinicalreviewthemaindifferenceintermsofresourcesuse
betweentreadmilltrainingandusualcarewastheuseofthetreadmillandtherewasnosubstantial
differenceinpersonneltime.Illustrativetreadmillcostsarepresentedbelow.
AGDGmembersuppliedpricedataonaspecifictypeoftreadmillwithoutbodysupportunit.This
datawasobtainedfromthemanufacturerofthetreadmill(Cranlea&CoMedical).Thetreadmill
modelwasaWoodwaymodeldesmohip,anditscostamountedto9,421(2011prices),withan
additional500estimatedasdeliveryandinstallationcosts.TheGDGalsosupplieddataregarding
therateofutilisationofatreadmillwithoutbodyweightsupportforanNHSTrustwithaninpatient
subacutestrokerehabilitationservice(basedatGuysandStThomasNHSFoundationTrust)where
patientsareusually060dayspoststroke.Thetreadmillaloneisusedtothelevelofapproximately2
4treatmentsessionseachday,whereeachsessionlastsforabout1hour.Annuitisingthiscost
assumingausefullifetimeof5years,noresalevalueandadiscountrateof3.5%,andassuming
usageof3sessionsperday,thiswouldequatetoacostpersessionof1.94.
Theestimateforthecostofatreadmillwithbodyweightsupportwasobtainedbycontactingthe
authorofaUSstudy(Walker,2010279).Theoverallcostquotedbytheauthorwasof$20,000(of
which$2,000wasthecostofthetreadmillalone),equivalentto13,029(at2009prices).The
manufacturerofthetreadmillandoftheBSWunitwasBiodex(aUScompany).AGDGmember
supplieddataregardingtherateofutilisationofatreadmillwithbodyweightsupportforanNHS
Trustwithaninpatientsubacutestrokerehabilitationservice(basedatGuysandStThomasNHS
FoundationTrust)wherepatientsareusually060dayspoststroke.ThetreadmillwithBWSunitis
usedforapproximately1patientpermonthforapproximately46treatmentsoverall.Annuitising
thiscostassumingausefullifetimeof5years,noresalevalueandadiscountrateof3.5%,and
assumingusageof5sessionspermonth,thiswouldequatetoacostpersessionof46.47.
13.8.2.3
Evidencestatements
Clinicalevidencestatements
Twostudies74,143of83participantswithacutestrokefoundthattherewasnosignificantdifferencein
walkingcapacity(6minutewalktest)(m)betweentheparticipantswhoreceivedtreadmilltraining
withnobodyweightsupportandthosewhohadusualcareattheendofthestudy(MODERATE
CONFIDENCEINEFFECT).
Threestudies19394,131of100participantswithchronicstroke(upto24monthspoststroke)found
thattherewasnosignificantdifferenceinwalkingcapacity(6minutewalktest)(m)betweenthose
whoreceivedtreadmilltrainingwithoutbodyweightsupportandtheusualcaregroupattheendof
thestudy(LOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
494
StrokeRehabilitation
Movement
Onestudy74of49participantsfoundthattreadmilltrainingwithnobodyweightsupportwas
associatedwithastatisticallysignificantimprovementinwalkingcapacity(6minutewalktest)(m)
comparedtothosereceivingusualcareattheendofthe18weeksfollowup(MODERATE
CONFIDENCEINEFFECT).
Fourstudies74,143,147,208comprisingof148participantswithacutestrokefoundastatistically
significantimprovementingaitspeed(10metretimedwalktest)(m/sec)inthosewhoreceived
treadmilltrainingwithnobodyweightsupportcomparedtothosewhoreceivedusualcareatthe
endofthestudy(LOWCONFIDENCEINEFFECT).
Threestudies19394,131of100participantswithchronicstroke(upto24monthspoststroke)founda
statisticallysignificantlargerimprovementingaitspeed(10metretimedwalktest)(m/sec)inthose
whoreceivedtreadmilltraining,comparedtothosewhoreceivedusualcareattheendofthestudy
(MODERATECONFIDENCEINEFFECT).
Twostudies74,190comprisingof102participantsfoundnosignificantdifferenceingaitspeed(10
metretimedwalktest)(m/sec)betweenthealltreadmilltraininggroup(withandwithoutbody
weightsupport)andthosewhoreceivedusualcareattheendofthefollowup(average29weeks)
(MODERATECONFIDENCEINEFFECT).
Onestudy138comprisingof56participantsfoundnosignificantdifferencebetweenthepartialbody
weightsupporttreadmillgroupandthosewhoreceivedaggressivebracingassistedwalkingatthe
endofthestudyonthefollowingoutcomes:
walkingspeedovera2minutetestperiod(m/minute)(VERYLOWCONFIDENCEINEFFECT)
walkingendurance(m)(VERYLOWCONFIDENCEINEFFECT).
Onestudy190comprisingof60participantsfoundnosignificantdifferencebetweenthetreadmill
traininggroupwithbodyweightsupportandthosewhoreceivedusualcareat10monthsfollowup
onthefollowingoutcomes:
FIMmotoritems(LOWCONFIDENCEINEFFECT)
FIMcognitiveitems(LOWCONFIDENCEINEFFECT).
Onestudy139comprisingof30participantsfoundnosignificantdifferencebetweentheparticipants
receivedthetreadmilltraininggroupandthosewhoreceivedusualcareat6and18monthsfollow
uponthefollowingoutcomes:
walkingendurance(LOWCONFIDENCEINEFFECT)
comfortablewalkingspeed(m/sec)(LOWCONFIDENCEINEFFECT)
fastwalkingspeed(m/sec)(LOWCONFIDENCEINEFFECT).
Onestudy73comprisingof265participantsfoundthattherewassignificantdifferencebetweenthe
participantswhoreceivedearlybodyweightsupportedtreadmilltrainingandthosewhoreceived
homeexerciseprogramonthefollowingoutcomes:
10metretimedwalktest(m/sec)attheendofthestudyand6monthsfollowup(MODERATE
CONFIDENCEINEFFECT)
6minutewalktest(m)attheendofthestudyand6monthsfollowup(MODERATECONFIDENCE
INEFFECT)
Twostudies272,115comprisingof139participantsfoundnosignificantdifference(in10metretimed
walktest)betweentheparticipantswhoreceivedbodyweightsupportedtreadmilltrainingand
thosewhoreceivedonlytreadmilltrainingpostintervention(MODERATECONFIDENCEINEFFECT)
Twostudies272,115comprisingof139participantsshowedsignificantdifferencein10metretimed
walktestinfavouroftheparticipantswhoreceivedbodyweightsupportedtreadmilltraining
NationalClinicalGuidelineCentre,2013.
495
StrokeRehabilitation
Movement
comparedwiththosewhoreceivedonlytreadmilltrainingattheendoffollowup(VERYLOW
CONFIDENCEINEFFECT)
Onestudy272comprisingof100participantsfoundnosignificantdifferencebetweentheparticipants
whoreceivedbodyweightsupportedtreadmilltrainingandthosewhohadonlytreadmilltrainingon
thefollowingoutcomes:
walkingendurance(m)attheendofthestudy(VERYLOWCONFIDENCEINEFFECT)
walkingendurance(m)at3monthsfollowup(VERYLOWCONFIDENCEINEFFECT).
Onestudy15comprisingof100participantsfoundthattherewasnosignificantdifferenceonthe
proportionofparticipantsachievedovergroundwalkingspeedover0.2m/sbetweenthosewho
receivedbodyweightsupportedtreadmillandthosewhohadonlytreadmilltraining(VERYLOW
CONFIDENCEINEFFECT).
Onestudy15comprisingof100participantsfoundthattherewasnosignificantdifferenceonthe
proportionofparticipantsachievedovergroundwalkingenduranceover20mbetweenthosewho
receivedbodyweightsupportedtreadmillandthosewhohadonlytreadmilltraining(LOW
CONFIDENCEINEFFECT).
Onestudy115comprising60participantsfoundnosignificantdifference(in6metrewalktest)
betweenthegroupthatreceivedbodyweighttreadmillandthegroupthatreceivedonlytreadmill
postinterventionandattheendoffollowup(LOWCONFIDENCEINEFFECT)
Onestudy115comprising60participantsfoundnosignificantdifference(intheFunctional
Independencemeasurescales9and13)betweenthegroupthatreceivedbodyweighttreadmilland
thegroupthatreceivedonlytreadmillpostinterventionandattheendoffollowup(VERYLOW
CONFIDENCEINEFFECT)
Evidencestatementscouldnotbeproducedforthefollowingoutcome(s)asresultswerenot
presentedinawaythatenabledthesizeoftheinterventionseffecttobeestimated:
Barthelindex(treadmillwithorwithoutbodyweightsupportversususualcare)86
Rivermeadgrossfunction(treadmillwithorwithoutbodyweightsupportversususualcare)74
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.8.3
Recommendationsandlinktoevidence
100. Offerwalkingtrainingtopeopleafterstrokewhoareableto
walk,withorwithoutassistance,tohelpthembuild
enduranceandmovemorequickly.
101. Considertreadmilltraining,withorwithoutbodyweight
support,asoneoptionofwalkingtrainingforpeopleafter
strokewhoareabletowalkwithorwithoutassistance.
NationalClinicalGuidelineCentre,2013.
496
StrokeRehabilitation
Movement
Relativevaluesofdifferent
outcomes
Theoutcomesconsideredintheevidencereviewwere:walkingspeeds
(5m/10m/30m),timedwalk,walkingendurance,FIM,Bartheland
RivermeadMobilityIndex.TheGDGconsideredoutcomes
demonstratingchangesinwalkingtobeofmoresignificance.
Overall,thesixteen(16)studiesincludedinthereviewshowedtherewas
animprovementinwalkingoutcomesinboththeinterventionand
controlgroups,buttreadmillprovidednogreaterimprovementthan
otherformsofphysiotherapy.
Tradeoffbetweenclinical
benefitsandharms
Thepatientrepresentativesonthegroupfeltthattoomuchemphasis
wasbeingplacedondistanceandspeed,andthatforpatients,the
primaryconcernwasthemotivationtowalkfrompointAtoBsafelyand
feelingcomfortable.Theyfeltthatspeedofwalkingwouldnotbea
significantconcernofpatients.
Groupmembersnotedthatthetrialswereconductedinagymsetting
andthiswouldnotnecessarilytranslatetowalkingoutdoors.The
consensusviewwasthatvarietywasanimportantpartofrehabilitation
treatmentforpatients,andtreadmillisareasonabletoolforuseingait
trainingforpeoplewhoarealreadywalkingandcanincreasewalking
speedandcapacity.
Economicconsiderations
Norelevantcosteffectivenessstudieswereidentified.Themain
differenceincostofusingtreadmilltrainingoverusualcarewas
consideredtobethecostofthetreadmill.Thecapitalcostofatreadmill
ishighataround10,000,howeverwhenthiscostisspreadoverthe
lifetimeoftheequipmentandtheamountofusageitgetsthecostper
patientpersessionwasestimatedat2fortreadmillwithoutbody
weightsupport,and47fortreadmillwithbodyweightsupport.The
GDGalsonotedthatatreadmillmayalreadybeavailableinmany
hospitalsandusedforpurposesotherthanstrokerehabilitation;if
currentlynotfullyutilised,useinstrokepatientscouldbe
accommodatedwithoutincurringthefullcostsestimated.
Qualityofevidence
Allthestudiesweresmall(rangeofparticipantsintheincludedstudies
25113).Thestudiesusingboththe6minutewalktestand10metre
timedwalkasmeasurementsoverdifferentfollowupperiodspresent
mixedresultsbutoverallshowednosignificantdifferencebetweenthe
interventionarm(treadmillwithorwithoutbodyweight)andthecontrol
arminwalkingcapacityorgaitspeedoutcomes.Confidenceinthe
resultsforwalkingcapacityandspeedoutcomeswasmoderatetovery
lowduetolimitationsinstudydesignorimprecisionintheeffect
estimate.
TheGDGnotedthatallthepeoplewithinthestudieshadsomewalking
capacityexceptforonetrial86.
TheGDGquestionedwhytheresultsintheVisintinstudy(1998)showed
therewaslategainachievedinthe10metrewalkoutcomewhenthere
wasnoimmediateresponseaftertheintervention.Afterdiscussion,the
GDGconcludedthatitmaybethatwithbodysupportbettergaitis
achievedandthereforethiswouldexplainthelategain.Therewerealso
differencesingaitspeedbetweenthetwogroupswhichmayhave
allowedtheinterventionarmtoreachacertainthreshold.
TheGDGconcludedthatthepatientswithinthetrialsusedintheclinical
evidencecouldalreadywalkwithsomesupport;thereforethe
recommendationshouldstatethatthisgrouparemostlikelytobenefit
fromtreadmill.
Otherconsiderations
Treadmillfacilitiesarewidelyavailableinrehabilitationunitsandthe
NationalClinicalGuidelineCentre,2013.
497
StrokeRehabilitation
Movement
GDGagreedthattreadmillwasonemethodofdeliveringanintensive
treatment.Astreadmillwasfoundtobeasgoodasconventionaltherapy
theGDGagreeditcouldbeconsideredforpatients.Patientscurrently
wouldusuallyreceivefixedamountsoftherapywhichincludestreadmill
trainingaspartoftheirrehabilitationtherapy.Ifpatientsarestablethen
treadmilluserscanworkwithminimalsupervisiononcetheyaresetup
ontheequipment.Treadmillenablespatientstogetmuchmorepractice
walkingthantheycanbywalkingoutdoorsanditalsooffersthe
opportunitytodoforcedspeedtraining.TheGDGagreedthattreadmill
wasausefultoolandofferedsomebenefittothepatientsafterstroke
evenifnogreaterthanusualcare.TheGDGagreedthatitwas
importantforeveryonetobeofferedwalkingtraining,andthiswas
backedupbytheevidencereviewedwhichshowedbotharmsofstudies
makinganimprovementfollowingwalkingexercises.
13.9 Electromechanicalgaittraining
Anelectromechanicalgaittrainerisaroboticgaitassistivedevicethatisdesignedtoprovidephysical
supportandmechanicalwalkingactionduringgaitreeducation.Thereareseveraltypesof
electromechanicalgaittraininginterventionsthatproviderepetitivelocomotortherapy.Locomat
andRehastimarethetradenamesofthetrainersusedinthestudiesconsidered.Botharerobotic,
orservocontrolledmotorassisteddevices,andprovidevariableamountsofassistanceduring
walkingtraining,includingtimingoflegmovementswiththeoptionofbodyweightsupport(upto
40%).Theadvocatesofelectromechanicalgaittrainersclaimthatitimproveswalkingbystimulating
anormal,symmetricalgaitcycle.
Asymmetricalmuscleweakness,tonalchanges,lossofsensationincludingproprioception,andpoor
balanceandcoordinationaremajorobstaclesinthesuccessfulrehabilitationofgaitintherecovering
strokepatient.Theuseofassistivedevicessuchaselectromechanicalgaittraineraimstoassistinthe
reeducationofgaitthroughsupportedrepetitionofwalkingbehaviour.
Theuseoftheaidisassumedtobeinthecontextofaprofessionallydirectedrehabilitation
programmetoimprovewalkingability.AsuitablyqualifiedPhysiotherapistandassistantwillbe
requiredtodesigntheappropriatewalkingtraining,positionthepatientcorrectlyandtoencourage
andadvisethroughoutthedurationoftheintervention.
Thistypeofinterventionmaybeusedthroughoutanystageintherehabilitationfollowingstrokeas
longasthepatientismedicallyfitandhasnocontraindicationstoexercise.
DuetothecostandscarcityoftheequipmentthisformofinterventionisrarelyseenwithinNHS
facilities.
13.9.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
electromechanicalgaittrainingversususualcareonimprovingfunctionandreducing
disability?
ClinicalMethodologicalIntroduction
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention
Electromechanicalgaittraining
Locomattraining
Comparison
Usualcare
Outcomes
Walkingspeeds(5metres/10metres/30metres)
Anytimedwalk
Walkingendurance
NationalClinicalGuidelineCentre,2013.
498
StrokeRehabilitation
Movement
ClinicalMethodologicalIntroduction
FunctionalIndependenceMeasure(FIM)
BarthelIndex
RivermeadMobilityIndex
13.9.1.1
Clinicalevidencereview
SearcheswereconductedforsystematicreviewsandRCTscomparingtheeffectivenessof
electromechanicalgaittrainingwithusualcareasinterventionsforimprovingfunctionandreducing
disabilityforadultsandyoungpeople16orolderwhohavehadastroke.Onlystudieswitha
minimumsamplesizeof20participants(10ineacharm)andincludingatleast50%ofparticipants
withstrokewereselected.ElevenRCTswereidentified.Table114summarisesthepopulation,
intervention,comparisonandoutcomesforeachofthestudies.
Table114:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Dias,2007
Singleincident
chronicstroke
patients(>12months
ago)withlowerlimb
motordeficit.
Gaittrainer/RehaStim(40
minutespersession,5
times/weekfor5weeks):
patientswereharness
securedandpositionedon
twofootplatessupervised
byaphysiotherapist;a
servocontrolledmotor
assistedthegaitmovement
bycontrollingthegear
velocitywithamaxof30%
bodyweightreliefduring
thefirstsessions.(N=20)
Usualcare:
patientsfollowed
theclassicBobath
methodand
rehabilitation
management,
includinganinitial
20minutes
sessionforjoint
mobilisationand
muscle
strengthening
plus20minutes
balanceandgait
training.(N=20)
Rivermead
MobilityIndex
BarthelIndex
10metre
timedwalk
test(m/sec)
6minutewalk
test(m)
Hidler,
2009110
Firsttimestroke;time
sincestrokeonset<6
months;abilityto
ambulate5meters
withoutphysical
assistanceandaself
selectedwalking
speedbetween0.1to
0.6m/s
Lokomatgaitorrobotic
assistedtraining(3
days/weeksfor810weeks
toamaximumtotalof24
sessions(1.5hourseach
session)):
patientswereharness
securedwithupto40%
bodyweightsupportwith
thelevelofbodysupport
decreasedincrementallyper
session.Thetraining
intensitywasincreasedby
changingthespeedandthe
levelofbodyweight
supportanddurationof
continuouswalking(N=36).
Conventionalgait
training:
focusedon
therapistassisted
staticand
dynamicpostural
tasks,trunk
positioning,
improvinglower
andupperlimb
rangeofmotion,
overground
walking,then
higherlevel
balanceandgait
activities
includingstairs.
(N=36)
5metretimed
walktest
(selfselected
velocity)
(m/sec)
6minutewalk
test(m)
Rivermead
MobilityIndex
Hornby,
2008114
Patientswith
hemiparesisof
>6monthsduration
afterstroke;abilityto
Roboticassistedlocomotor
training(Lokomat)(atotal
of12sessionsfor30
minutes/session):patients
Therapistassisted 6minutewalk
locomotor
test(m)
training(atotalof 10metre
12sessionsfor30
timedwalk
64
NationalClinicalGuidelineCentre,2013.
499
StrokeRehabilitation
Movement
STUDY
POPULATION
walk>10mwithout
physicalassistanceat
speedsofatleast
0.8m/satself
selectedvelocity
usingassistivedevices
andbracingbelow
thekneeasneeded.
INTERVENTION
wereharnesssecuredto
provide3040%bodyweight
inthefirstsessionand
decreasedapproximately
10%incrementspersession
astoleratedwithout
substantialkneebucklingor
toedrag.(N=31)
COMPARISON
minutes/session):
patientswere
harnesssecured
andtrainedat
similarweight
supportand
speedstorobotic
assistedgroup.
(N=31)
Husemann,
2007118
Firsttimestroke
patientswithsevere
hemiparesiswith
lowerextremity
strengthgraded3or
lessonMRCscalein
>2musclegroups.
Intervalbetween
strokeandstartof
treatment28200
days.
Roboticassistedtraining(a
totalof40sessions(30
minutes/session)for4
weeks):patientswalkedon
atreadmillwiththehelpof
aLokomat.30%ofbody
weightofeachsubjectwas
supportedinitially.The
velocityofthetreadmillset
tothemaximumspeed
toleratedbythepatients,
forceofthedrivesregulated
andbodyweightsupport
wasreducedassoonas
patientscouldtolerateit.
Patientsalsoreceived30
minutesphysiotherapy
sessions.(N=17)
Therapistassisted 10metre
gait
timedwalk
rehabilitation:
test(m/sec)
exercisingtrunk
BarthelIndex
stabilityand
(German
symmetry,step
version)
initiationand
weightsupport.
Treadmilltraining
wasprovidedif
possiblewiththe
helpoftherapists.
Patientsalso
received30
minutes
physiotherapy
sessions.
(N=15)
Peurala
2005204
Chronicfirsttime
strokepatients(>6
months)whohad
slowordifficult
walking
Gaittrainer/RehaStim(20
minutes/sessionforatotal
of3weeks):supportedwith
aharnessandtheirfeet
wereonmotordriven
footplates+partialbody
weightsupportverballyor
manuallyguidedby
physiotherapists(N=15)
Walking
overground:
practised
overgroundor
overuneven
terrainwith
walkingaids.
(N=15)
Peurala
2009203
Firsttimestroke
patients(within10
daysofstrokeonset)
Gaittrainer/RehaStim)(in
totalof15sessionsfor3
weeks):patientswere
supportedwithaharness
andtheirfeetwere
supportedonmotordriven
footplates.Amountofbody
supportwaschosen
accordingtoindividual
needs.(N=22)
Controlgroup1
10metre
timedwalk
Walkingtraining:
(m/sec)
patientspractised
6minutewalk
walking
test(m)
overgroundwith
1or2
Rivermead
physiotherapists,
MobilityIndex
usingwalking
aids.
(N=21)
Controlgroup2
Conventional
treatment:
patientsreceived
oneor2
physiotherapy
sessionsdaily,but
NationalClinicalGuidelineCentre,2013.
500
OUTCOMES
test(self
selected
velocity)
(m/sec)
10metre
timedwalk
(m/sec)
6minutewalk
test(m)
Functional
Independence
Measure
(FIM)(total
score)
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
notatthesame
intensityasthe
othertwogroups.
(N=13)
Pohl,2007209
Firsttimestroke
patients(<60days);
abletosit
unsupportedwith
feetsupported,could
notwalkatall,or
requiredthehelpof
oneortwotherapists
irrespectiveofthe
useofananklefoot
orthosisorawalking
aid.
Gaittrainer/RehaStim
(everyweekdayfor4
weeks):patientsreceived20
minutesofrepetitive
locomotortherapyonthe
gaittrainer,immediately
followedby25minutesof
onetoonephysiotherapy
everyweekdayfor4weeks
(N=77)
Usualcare:
patientsreceived
45minute
physiotherapy.
(N=78)
Schwartz,
2009233
Firsttimestroke
patients(<3months)
whowere
independent
ambulatedbeforethe
stroke.
Roboticassistedgait
training(Lokomat)(30
minutes/session/workday,
3timesaweekfor6weeks):
thespeedoftreadmillsetto
maximumtoleratedbythe
patients.Approximately
50%ofbodyweight
supportedbyharness
initially,thesupportwas
graduallyreducedin
approximately10%
incrementspersessionas
toleratedwithout
substantialkneebucklingor
toedrag.Patientsalso
receivedregular
physiotherapy.(N=37)
Usualcare:
FIM
patientstreated
10metre
withadditional
timedwalk
regular
test(m/sec)
physiotherapyfor
gaittrainingfor
30minutes
(overall60
minutesof
regular
physiotherapy)3
timesaweek.In
everysessionthe
patientwalked
somestepswith
thehelpof
therapists.(N=30)
Tong,2006260
Firststrokepatients
admittedtoinpatient
rehabilitationunitin
HongKong(<6weeks
afteronsetofstroke)
withsignificantgait
deficit(FACscore<3);
abilitytostand
upright,supportedor
unsupportedfor1
min.
Electromechanicalgait
trainer/RehaStim(1
trainingsessionof
20min/weekdayovera
totalof4weeks):
gaittrainerstimulateda
normalgaitcycleina
symmetricmannerwitha
ratioof6040%betweenthe
stanceandswingphases.
Bodyweightpartially
supportedbyaharness.
Patientshadalsoregular
weekday40min
physiotherapysessionsand
1.5hourmultidisciplinary
treatments.(N=15)
Controlgait
training:
trainingbasedon
principlesof
proprioceptive
neuromuscular
facilitationand
Bobathconcepts
byaphysical
therapist.
Patientshadalso
regularweekday
40min
physiotherapy
sessionsand
1.5hourmulti
disciplinary
treatments.
(N=20)
NationalClinicalGuidelineCentre,2013.
501
BarthelIndex
10metre
timedwalk
(m/sec)
6minutewalk
test(m)
Rivermead
MobilityIndex
5metretimed
walk(m/sec)
FIM
Barthelindex
StrokeRehabilitation
Movement
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Werner
2002281
Subacute,non
ambulatory
hemipareticpatients
(412weeksafter
stroke)
Gaittrainer/RehaStim(15
20minutes/sessionfor2
weeks):harnesssecured
patientswerepositionedon
2footplates.Aservo
controlledmotorassisted
thegaitmovement;body
weightwaspartially
supportedandthesupport
wasgraduallyreduced.
(N=15)
Treadmill
training:
motordriven
treadmill,
patientsworea
modified
parachute
harnessanda
pulleyreleased
partofthebody
weight.(N=15)
10metre
timedwalk
(m/sec)
(maximum
speed)
Morone
2012179
Participantswithsub
acutestrokestratified
intogroupsaccording
tomotorimpairments
(lowmotricityand
highmotricity
groups).Theauthors
usedthemotricity
indexscorewitha
cutoffof29to
definealowmotricity
(LM)groupof
patientsversusa
groupofpatientswith
highmotricity(HM;
MI>29).
Gaittrainer/RehaStim:2
Conventionalgait
therapysessionsperday5
training(N=24)
daysperweekfor3months;
GaitTrainerforfirst4weeks
(N=24)
NationalClinicalGuidelineCentre,2013.
502
Functional
ambulation
Classification
BarthelIndex
Rivermead
MobilityIndex
StrokeRehabilitation
Movement
Comparison:Electromechanicalgaittrainingversususualcare
Table115:ElectromechanicalgaittrainingversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
Usualcare
Mean(SD)
/median
(IQR)
Mean
difference
(95%CI)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
5and10metretimedwalk(m/sec)(posttreatmenteffect)(Betterindicatedbyhighervalues)
6
Hornby2008114
Husemann
2007118
Peurala2005204
Pohl2007209
Tong2006260
Werner2002281
RCTs2
Veryserious
single
limitations
blinded,3 (a)
unblinded,
1cross
overtrial
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
Changefrom
baseline
Hornby:
0.52(0.21)
Husemann:
0.2(0.12)
Pohl:
0.44(0.47)
Tong:
0.47(0.21)
Werner:
0.42(0.21)
Finalvalues
Peurala:
35.9(29.9)
Change
0.22(
from
0.00,0.43)
baseline
Hornby:
0.56(0.28)
Husemann
:0.2(0.18)
Pohl:
0.32(0.36)
Tong:
0.24(0.30)
Werner:
0.37(0.23)
Final
values
Peurala:
NationalClinicalGuidelineCentre,2013.
503
SMD
0.22
higher
(0.00
lowerto
0.43
higher)
Low
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
Usualcare
Mean(SD) Mean
/median
difference
(IQR)
(95%CI)
32.1(15.9)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
5metretimedwalk(m/sec)(posttreatmenteffect)*(Betterindicatedbyhighervalues)
1
Hidler2009110
RCT
unblinded
Veryserious
limitations
(b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
0.12(0.03)
0.25(0.03) 0.13(
MD0.13
0.14,0.12) lower
(0.14to
0.12
lower)
Low
5metretimedwalk(m/sec)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Hidler2009110
RCT
unblinded
Veryserious
limitations
(b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
0.15(0.04)
0.30
(0.03)
0.15(
0.17,
0.13)
MD0.15
lower
(0.17to
0.13
lower)
Low
0.52
(0.25)
0.02(
0.13,0.09)
MD
0.02
lower
(0.13
lowerto
0.09
higher)
Low
10metretimedwalk(m/sec)(6monthsfollowup)(>6monthsstrokeonset)^(Betterindicatedbyhighervalues)
1
Hornby2008114
RCTs
single
blinded
Veryserious
limitations
(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(d)
0.50(0.21)
NationalClinicalGuidelineCentre,2013.
504
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
Usualcare
Mean(SD)
/median
(IQR)
Mean
difference
(95%CI)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
10metretimedwalk(m/sec)(6monthfollowup)(<2monthsstrokeonset)^(Betterindicatedbyhighervalues)
1
Pohl2007209
RCTs
unclear
blinding
Serious
limitations
(e)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(d)
0.53(0.31)
0.36(0.42) 0.17(
0.05,0.29)
MD0.17
higher
(0.05
lowerto
0.29
higher)
Low
Serious
imprecision
(q)
Changefrom
baseline
Dias:
18.92
(26.33)
Finalvalues
Hornby:
186(88)
Pohl:
134.4
(125.5)
Peurala:
151.7
Change
from
baseline
Dias:
23.28
(2.16)
Final
values
Hornby:
204(96)
Pohl:
92.5
(104.9)
Peurala:
SMD
0.20
higher
(0.03
lowerto
0.44
higher)
Verylow
6minutewalktest(m)(posttreatmenteffect)(Betterindicatedbyhighervalues)
4
Hornby2008114
Peurala2005204
Pohl2007209
Dias200764
RCTs2
Veryserious
single
limitations
blinded,1 (f)
unblinded,
1trialwith
unclear
blinding
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
505
0.20(
0.03,0.44)
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
(97.4)
Usualcare
Mean(SD)
/median
(IQR)
135.1
(67.9)
Mean
difference
(95%CI)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
6minutewalktest(m)(posttreatmenteffect)(changeinscoresfromthebaseline*)(Betterindicatedbyhighervalues)
1
Hidler2009110
RCT
unblinded
Veryserious
limitations
(b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
50.17(9.906)
83.515
(10.79)
33.34
(38.13,
28.56)
MD
33.34
lower
(38.13
to28.56
lower)
Low
62.21
(14.87)
101.96
(15.18)
39.75(
46.69,
32.81)
MD
Low
39.75
lower
(46.69to
32.81
lower)
203(104)
10.00(
59.35,
39.35)
MD10
lower
(59.35
lowerto
6minutewalktest(m)(selfselected)(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Hidler2009110
RCT
unblinded
Veryserious
limitation(b)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
6minutewalktest(m)(6monthsfollowup)(>6monthstrokeonset^)(Betterindicatedbyhighervalues)
1
Hornby2008114
RCTs
single
blinded
Veryserious
limitations
(c)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(q)
193(94)
NationalClinicalGuidelineCentre,2013.
506
Verylow
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
Usualcare
Mean(SD)
/median
(IQR)
Mean
difference
(95%CI)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
39.35
higher)
6minutewalktest(m)(6monthfollowup)(<60daysstrokeonset^)(Betterindicatedbyhighervalues)
1
Pohl2007209
RCTs
unclear
blinding
Serious
limitations
(g)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(d)
165.5(152.5)
112.1
(127.7)
53.40
(9.09,
97.71)
MD
53.40
higher
(9.09to
97.71
higher)
Low
Very
serious
imprecision
(i)
100.9(12.3)
102.3
(10.9)
1.40(
9.72,6.92)
MD1.40
lower
(9.72
lowerto
6.92
higher)
Verylow
(k)
91(17)
89.5(26.5) (l)
(l)
Moderate
(k)
Serious
66.9(15.6)
60.3(14.8) 6.60(
MD6.6
Low
FIM(totalscore)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Peurala2005204
RCT
unblinded
Veryserious
limitations
(h)
Noserious
inconsistency
Serious
indirectness
FIM(totalscore)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Tong2006260
RCT
unblinded
Serious
Noserious
limitations(j) inconsistency
Noserious
indirectness
FIM(motoritems)(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
RCT
Serious
Noserious
Noserious
NationalClinicalGuidelineCentre,2013.
507
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Schwartz
2009233
Design
unblinded
Effect
Limitations
limitations
(m)
Inconsistency
inconsistency
Indirectness
indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
imprecision
(n)
Usualcare
Mean(SD)
/median
(IQR)
Mean
difference
(95%CI)
0.70,
13.90)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
higher
(0.70
lowerto
13.9
higher)
RivermeadMobilityIndex(posttreatmenteffect)(Betterindicatedbyhighervalues)
3
Dias200764
Pohl2007209
Hidler2009110
RCTs
2single
blind,1
unblinded
Veryserious
limitations
(o)
Noserious
inconsistency
Noserious
indirectness
Noserious Changefrom
imprecision baseline
Dias:0.35
(0.75)
Hidler:2.0(0.3)
Finalvalues
Pohl:
8.5(3.9)
Change
from
baseline
Dias:
1.26
(1.82)
Hidler:
1.6(0.3)
Pohl:
6.3(3.7)
0.83(0.35,
1.31)
SMD
0.83
higher
(0.35
lowerto
1.31
higher)
Low
2.2(0.5)
0.40(0.19,
0.61)
MD0.4
higher
(0.19to
0.61
higher)
Low
RivermeadMobilityIndex(3monthsfollowup)(Betterindicatedbyhighervalues)
1
Hidler2009110
RCT
unblinded
Veryserious
limitations
(b)
Noserious
inconsistency
Noserious
indirectness
Noserious 2.6(0.4)
imprecision
NationalClinicalGuidelineCentre,2013.
508
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
Design
Effect
Limitations
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
Usualcare
Mean(SD)
/median
(IQR)
Mean
difference
(95%CI)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
RivermeadMobilityIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Pohl2007209
RCT
single
blinded
Serious
limitations
(e)
Noserious
inconsistency
Noserious
indirectness
Noserious 10.0(4.1)
imprecision
7.8(4.8)
Noserious
imprecision
72.3(21)
(k)
Noserious
imprecision
2.20(0.80,
3.60)
MD2.20
higher
(0.80to
3.60
higher
Moderate
58.7(21.6) 13.60
(6.89,
20.31)
MD13.6
higher
(6.89to
20.31
higher)
Moderate
Tong:
84.0(19)
Husmann:
50.0(25)
Tong:
(l)
73.0(32.5)
Husmann:
50(10)
(l)
Moderate(k
)
77.5(23.1)
65.1(28.0) 12.40
(4.32,
MD12.4
higher
Moderate
BarthelIndex(posttreatmenteffect)(Betterindicatedbyhighervalues)
1
Pohl2007209
RCT
single
blinded
Serious
limitations
(e)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(posttreatmenteffect)(Betterindicatedbyhighervalues)
2
Tong2006260
Husemann
2007118
RCTs1
single
blinded,1
unblinded
Serious
limitations
(jp)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
1
Pohl2007209
RCT
single
Serious
limitations
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
509
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noofstudies
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
(l)
(m)
(n)
(o)
(p)
(q)
Design
blinded
Effect
Limitations
(a)
Inconsistency
Indirectness
Electro
mechanicalgait
training
Mean(SD)/
Imprecision median(IQR)
Usualcare
Mean(SD)
/median
(IQR)
Mean
difference
(95%CI)
20.48)
Standard
ised
Mean
Differen
ce(SMD)
/Mean
Differen
ce(MD) Confidence
(95%CI) (ineffect)
(4.32to
20.48
higher)
Unclearrandomisation(Hornby2008;Peurala2005),inadequateallocationconcealment(Hornby2008;Pohl2007;Peurala2005;Tong2006),unclearallocationconcealment(Hidler2009)unclearblinding
(Hornby2008),unblindedstudies(Peurala2005;Tong2006;Hidler2009)highdropoutrateincontrolgroup(30%)(Hornby2008).
Unblindedstudywithunclearallocationconcealment.No ITTanalysis.Therewasasignificantdifferenceinagebetweengroups,reportedbyauthor.RivermeadMobilityIndexatbaselinewasstatistically
significantlyhigherincontrolgroupcomparedtotheintervention(Hidler2009).
Studywithunclearblindingandrandomisationhighdropoutrateincontrolgroup(30%)(Hornby2008).
ConfidenceintervalcrossedoneendofdefaultMID.
Singleblindedstudywithinadequateallocationconcealment(Pohl2007).
1unblinded(Peurala2007),1withunclearblinding(Hornby2008),2studieshadunclearrandomization(Hornby2008;Peurala2005)andthemajorityhadinadequateallocationconcealment(Hornby2008;
Pohl2007;Peurala2005)andunclearallocationconcealment(Dias2007).
Inadequateallocationconcealment(Pohl2007).
Unblindedstudywithunclearrandomizationandinadequateallocationconcealment(Peurala2007).
MeandifferencedidnotreachtheagreedMIDof22points.
Studywasunblindedwithunclearallocationconcealment(Tong2006).
Imprecisioncouldnotbeassessedbecauseresultswerereportedasmedianandinterquartilerange.
Relativeandabsoluteeffectscouldnotbecalculatedbecauseresultswerereportedasmedianandinterquartilerange.
Studywasnotblindedandunclearallocationconcealment;Dropoutrategreaterininterventiongroupcomparedtocontrols(Schwartz2009).
MeandifferencedidnotreachtheagreedMIDof17points.
Unblinded(Hidler2009),withonestudywithinadequateallocationconcealment(Pohl2007)andtwostudieswithunclearallocationconcealment(Dias2007,Hidler2009).
GermanversionofBarthelIndexwasusedinthestudy,Husemannetal(2007).
MeandifferencedidnotreachtheagreedMIDof28m.
2
*Walkingvelocity:5and10metretimedwalktest(posttreatmenteffect):wecouldnotmetaanalyseHidleretal.withotherstudiesastherewassubstantialheterogeneity(I :99%),thereforetheresultsofthis
studywerepresentedseparately.
^Walkingvelocity:5and10metretimedwalktest(6monthsfollowup):wecouldnotmetaanalyseHornbyetal.andPohletal.togetherduetothelargedifferenceintimebetweenstrokeonsetandstudyentry
amongtheparticipants.
NationalClinicalGuidelineCentre,2013.
510
StrokeRehabilitation
Movement
OneRCT(Morone2012179)stratifiedparticipantswithstrokeaccordingtotheirinitialmotorimpairmentlevelsintoseparategroupsfromtheoutset.
ResultsofthistrialarepresentedinaseparateGRADEtablesinceoverallvalueswerenotprovided.
Comparison:Electromechanicalgaittrainingversusconventionalgaittraining(ingroupsstratifiedbylevelofmotorimpairments)
Table116:GRADEcharacteristicsandclinicalsummaryoffindings(Note.LM=lowmotricitygreaterlevelofimpairments;HM=highmotricitylower
levelofimpairment)
SummaryofFindings
Qualityassessment
Noof
studies Design
Effect
Riskof
bias
Inconsistency
Indirectness
Imprecision
Roboticgait
trainingMeans
(SD)
Mean
difference
(MD)(95% Confidence
CI)
(ineffect)
Conventional
gaittraining
Mean(SD)
Mean
difference
(95%CI)
2.1(1.2)
1.9(1.05,
2.75)
MD1.9
higher
(1.05to
2.75
higher)
Moderate
1.6(0.81,
2.39)
MD1.6
higher
(0.81to
2.39
higher)
Moderate
0.1(0.74,
0.94)
MD0.1
higher
(0.74
lowerto
0.94
higher)
VeryLow
FunctionalambulationclassificationDischargeLM(rangeofscores:05;Betterindicatedbyhighervalues)
Moron randomised Serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
4(0.9)
FunctionalambulationclassificationFollowupat2yearsLM(rangeofscores:05;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
4.7(0.5)
3.1(1.3)
FunctionalambulationclassificationDischargeHM(rangeofscores:05;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
very
serious(b)
3.8(1.1)
3.7(1.0)
NationalClinicalGuidelineCentre,2013.
511
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies Design
Effect
Riskof
bias
Inconsistency
Indirectness
Imprecision
Roboticgait
trainingMeans
(SD)
Conventional
gaittraining
Mean(SD)
Mean
difference
(95%CI)
Mean
difference
(MD)(95% Confidence
CI)
(ineffect)
FunctionalambulationclassificationFollowupat2yearsHM(rangeofscores:05;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
very
serious(b)
4.3(0.9)
4.0(1.0)
0.3(0.46,
1.06)
MD0.3
higher
(0.46
lowerto
1.06
higher)
VeryLow
52.1(14.1)
17.5(5.81,
29.19)
MD17.5
higher
(5.81to
29.19
higher)
Moderate
64.7(14.0)
17.5(5.81,
29.19)
MD12.2
higher
(1.95to
22.45
higher)
Moderate
74.2(14.1)
10(
26.61,
6.61)
MD10
lower
(26.61
lowerto
6.61
higher)
VeryLow
BarthelIndexDischargeLM(rangeofscores:0100;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
69.6(15.1)
BarthelIndexFollowupat2yearsLM(rangeofscores:0100;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
76.9(11.5)
BarthelIndexDischargeHM(rangeofscores:0100;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
very
serious(c)
64.2(21.2)
NationalClinicalGuidelineCentre,2013.
512
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies Design
Effect
Riskof
bias
Inconsistency
Indirectness
Imprecision
Roboticgait
trainingMeans
(SD)
Mean
difference
(MD)(95% Confidence
CI)
(ineffect)
Conventional
gaittraining
Mean(SD)
Mean
difference
(95%CI)
77.6(20.4)
3.3(
18.96,
12.36)
MD3.3
lower
(18.96
lowerto
12.36
higher)
4.9(2.0)
4.5(2.6,
6.4)
MD4.5
higher(2.6 Moderate
to6.4
higher)
BarthelIndexFollowupat2yearsHM(rangeofscores:0100;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
very
serious(c)
74.3(18.7)
VeryLow
RivermeadMobilityIndexDischargeLM(rangeofscores:015;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
noserious
imprecision
9.4(2.7)
RivermeadMobilityIndexFollowupat2yearsLM(rangeofscores:015;Betterindicatedbyhighervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
serious(d)
11.8(3.5)
7(3.6)
4.8(1.96,
7.64)
MD4.8
higher
(1.96to
7.64
higher)
Low
7.4(4.1)
10.1(4.0)
2.7(5.94, MD2.7
0.54)
lower
(5.94
lowerto
0.54
higher)
Low
RivermeadMobilityIndexDischargeHM(Betterindicatedbylowervalues)
Moron randomised serious
e
trials
(a)
2012179 (assessor
blinded)
noserious
inconsistency
noserious
indirectness
serious(e)
RivermeadMobilityIndexFollowupat2yearsHM(rangeofscores:015;Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
513
StrokeRehabilitation
Movement
SummaryofFindings
Qualityassessment
Noof
studies Design
Effect
Riskof
bias
Inconsistency
Indirectness
Imprecision
noserious
inconsistency
noserious
indirectness
very
serious(b)
Roboticgait
trainingMeans
(SD)
Conventional
gaittraining
Mean(SD)
Mean
difference
(95%CI)
10.4(3.6)
10.6(3.9)
0.2(3.2,
2.8)
Mean
difference
(MD)(95% Confidence
CI)
(ineffect)
MD0.2
lower(3.2 VeryLow
lowerto
2.8higher)
(a)
Unclear allocation concealment, a large proportion (up to 50%) of participants in all study arms did not finish the 4 week treatment.
The confidence interval crosses both default MIDs (0.5 of standard mean difference) ranging from benefit associated with the control treatment to benefit associated with robotic gait training.
The confidence interval crosses both agreed MIDs for the Barthel Index (1.85), i.e. robotic gait training could have positive or negative effects
(d)
The confidence interval crosses one default MID (0.5 standard mean difference) from appreciable benefit to no effect associated with robotic gait training
(e)
The confidence interval crosses one default MID (0.5 standard mean difference) indicating possible benefit associated with the conventional gait training to no effect.
(b)
(c)
Narrativesummary
ThefollowingstudyissummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Onestudy203randomised56patientstogaittrainerexercise,walkingtrainingandconventionaltreatment.Attheendof3weekstraining,meanwalking
velocity(10metretimedwalktest)andwalkingdistance(6minutewalktest)werenotdifferentbetweenthegaittrainerexerciseandwalkinggroups(10
metretimedwalktest,p=0.452;6minutewalktest,p=0.547).TheRivermeadMobilityIndeximprovedinallgroups(frombaselinetoendoftreatment)
butpvalueforgroupdifferencewasnotstatisticallysignificant(p=0.703).Analysiswasbasedonthenumberofpatientswhowereabletowalk20minutes
(differentlevelofpatientsparticipationindifferentmeasurementsatdifferenttimepoints)andreconstructeddatafor10metretimedwalktestand6
minutewalktestwasused;thereforethisstudywasnotincludedinthemetaanalysis.
NationalClinicalGuidelineCentre,2013.
514
StrokeRehabilitation
Movement
13.9.1.2
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingelectromechanicalgaittrainingwithusualcarewere
identified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
ThemanufacturersofLokomatandofRehaStimelectromechanicalgaittrainerswerecontactedand
theyeachsuppliedcostsfortheirproducts.TheLokomatelectromechanicalgaittrainercostsranged
between~173,000to~264,000(costsprovidedbyHocomabyemail,20thJune2011;VATis
excluded).TheRehaStimelectromechanicalgaittrainercostwasprovidedbutisnotreportedhere
asitwasdeemedcommercialinconfidence.Assumingadiscountrateof3.5%,alifeexpectancyfor
themachineof10years,autilizationrateofthemachineof208daysperyearandof4hourseach
day,foraninterventionconsistingof6hoursofuseoftheelectromechanicalgaittraining,the
attributablecostfortheinterventionusingaLokomattrainerwouldbebetween~145and~221.
Tothesecostsitmaybenecessarytoaddpersonnelcostswhenthepatientneedstobeaidedin
usingtheelectromechanicalgaittrainer.
13.9.1.3
Evidencestatements
Clinicalevidencestatements
Sixstudies114,118,204,209,260,281of344participantsfoundnosignificantdifferencein5and10metre
timedwalktest(m/sec)betweentheelectromechanicalgaittraininggroupandtheusualcaregroup
attheendoftheintervention(LOWCONFIDENCEINEFFECT).
Onestudy110of72participantsfoundthatthosewhoreceivedusualcarewasassociatedwitha
statisticallysignificantimprovementin5metretimedwalktest(m/sec)comparedwiththe
electromechanicalgaittraininggroupattheendoftheintervention(LOWCONFIDENCEINEFFECT).
Onestudy110of72participantsfoundthatthosewhoreceivedusualcarewasassociatedwitha
statisticallysignificantimprovementin5and10metretimedwalktest(m/sec)thanthe
electromechanicalgaittraininggroup,at3monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy114of62participants(>6monthspoststroke)foundnosignificantdifferencein5and10
metretimedwalktest(m/sec)betweentheelectromechanicalgaittraininggroupandtheusualcare
groupattheendof6monthsfollowup(LOWCONFIDENCEINEFFECT).
Onestudy209of155participants(<2monthspoststroke)foundthattheelectromechanicalgait
traininggroupwasassociatedwithastatisticallysignificantimprovementin10metretimedwalktest
(selfselected)(m/sec)comparedwiththeusualcaregroupattheendof6monthsfollowup(LOW
CONFIDENCEINEFFECT).
Fourstudies64,114,204,209of287participantsfoundnosignificantdifferencein6minutewalktest(m)
betweentheelectromechanicalgaittrainingandtheusualcaregroupattheendofintervention
(LOWCONFIDENCEINEFFECT).
NationalClinicalGuidelineCentre,2013.
515
StrokeRehabilitation
Movement
Onestudy110of72participantsfoundthatthosewhoreceivedusualcarewasassociatedwitha
statisticallysignificantgreaterimprovementin6minutewalktest(m)comparedwiththe
electromechanicalgaittraininggroupattheendofintervention(LOWCONFIDENCEINEFFECT).
Onestudy110of72participantsfoundnosignificantdifferencein6minutewalktest(m)(self
selected)betweentheelectromechanicalgaittrainingandtheusualcaregroupat3monthsfollow
up(LOWCONFIDENCEINEFFECT).
Onestudy114of62participants(>6monthspoststroke)foundnosignificantdifferencein6minute
walktest(m)betweentheelectromechanicalgaittrainingandtheusualcaregroupat6months
followup(VERYLOWCONFIDENCEINEFFECT).
Onestudy209of155participants(<60dayspoststroke)foundthatthosewhoreceived
electromechanicalgaittrainingwasassociatedwithastatisticallysignificantgreaterimprovementin
6minutewalktest(m)comparedwiththosewhoreceivedusualcareat6monthsfollowup(LOW
CONFIDENCEINEFFECT).
Onestudy204of30participantsfoundnosignificantdifferenceintotalFunctionalIndependence
Measurebetweentheelectromechanicalgaittrainingandtheusualcaregroupattheendof
intervention(VERYLOWCONFIDENCEINEFFECT).
Onestudy233of67participantsfoundnosignificantdifferenceinFunctionalIndependenceMeasure
(Motoritem)betweenelectromechanicalgaittrainingandusualcareattheendofintervention
(LOWCONFIDENCEINEFFECT).
Threestudies64,110,209of267participantsfoundthatelectromechanicalgaittrainingwasassociated
withastatisticallysignificantimprovementontheRivermeadMobilityIndexcomparedwiththe
usualcaregroupattheendofintervention(LOWCONFIDENCEINEFFECT).
Onestudy110of72participantsfoundthatelectromechanicalgaittrainingwasassociatedwitha
statisticallysignificantimprovementontheRivermeadMobilityIndexcomparedwiththosewho
receivedusualcareat3months(LOWCONFIDENCEINEFFECT)
Onestudy209of155participantsfoundthatelectromechanicalgaittrainingwasassociatedwitha
statisticallysignificantimprovementontheRivermeadMobilityIndexcomparedwiththosewho
receivedusualcareat6monthsfollowup(MODERATECONFIDENCEINEFFECT).
Onestudy209of155participantsfoundthatelectromechanicalgaittrainingwasassociatedwitha
statisticallysignificantimprovementonBarthelIndexcomparedwiththosewhoreceivedusualcare
attheendofintervention(MODERATECONFIDENCEINEFFECT)and6monthsfollowup(MODERATE
CONFIDENCEINEFFECT).
Electromechanicalgaittrainingversusconventionalgaittrainingingroupsdividedbyinitialmotor
impairmentlevel
Onestudy179of48participantsfoundthatroboticgaittrainingsignificantlyimprovedfunctional
ambulatoryabilities(asmeasuredbytheFunctionalAmbulationClassificationscale)inthosewith
higherlevelsofmotorimpairments(atbaseline).Theseimprovementswereobservedbothat
dischargeandat2yearfollowupcomparedtoconventionalgaittraining.However,these
improvementswerenotobservedinparticipantswithfewerimpairments(atbaseline).
Onestudy179of48participantsfoundthatroboticgaittrainingsignificantlyimprovedperformancein
activitiesofdailyliving(asmeasuredbytheBarthelIndex)inthosewithhigherlevelsofmotor
impairments(atbaseline).Theseimprovementswereobservedbothatdischargeandat2year
followupcomparedtoconventionalgaittraining.However,theseimprovementswerenotobserved
inparticipantswithfewerimpairments(atbaseline).
NationalClinicalGuidelineCentre,2013.
516
StrokeRehabilitation
Movement
Onestudy179of48participantsfoundthatroboticgaittrainingsignificantlyimprovedmobility(as
measuredbytheRivermeadMobilityIndex)inthosewithhigherlevelsofmotorimpairments(at
baseline).Theseimprovementswereobservedbothatdischargeandat2yearfollowupcompared
toconventionalgaittraining.However,theseimprovementswerenotobservedinparticipantswith
fewerimpairments(atbaseline).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.9.2
Recommendationsandlinktoevidence
102. Offerelectromechanicalgaittrainingtopeopleafterstrokeonlyin
thecontextofaresearchstudy.
Relativevaluesof
differentoutcomes
Theoutcomesofinterestincludedinthereviewwerewalkingspeedand
endurance,theRivermeadMobilityIndexandtwomeasuresofdependencethe
FunctionalIndependenceMeasure,andtheBarthelIndex.TheGDGconsidered
theresultsofthewalkingoutcomeswereofmorerelevancetotheintervention.
Tradeoffbetween
clinicalbenefitsand
harms
Notapplicable.Theavailabilityandusageofthisequipmentiscurrently
extremelylimitedwithintheNHS.
Economicconsiderations Nocosteffectivenessstudieswereidentifiedforthisquestion.TheGDGnoted
thatthemaincostcomponentfortheseinterventionsconsistsofthecostof
acquiringandmaintainingthemachine,aswellasthepersonnelcosts(for
examplephysiotherapisttime)thatmayberequiredtoaidthepatientinusing
theelectromechanicalgaittraining.Inaddition,theGDGnotedthatthereisvery
limiteduseofelectromechanicalgaittrainingdevicescurrentlyintheUKNHS.
Consideringthehighinitialoutlaycostforelectromechanicalgaittrainersand
thelimitedevidencefortheirpotentialhealthbenefits,theGDGconcludedthat
therewasinsufficientevidencetoconcludethatelectromechanicalgaintraining
representsacosteffectiveuseofNHSresources.
Qualityofevidence
TheGDGnotedthatmanyofthestudiespresentedhadseveretoveryserious
limitationsintermsofsamplesizeandstudydesignandtherewasinsufficient
evidencetosupporttheuseofelectromechanicalgaittraining.
Manystudiesdidnotshowasignificantdifferenceforthewalkingspeedor
capacityoutcomes114,118,204,233,260,281.InonewelldesignedstudybyPohl209there
isevidencethatinpatientsearlyafterstroke(upto2monthspoststroke)with
verypoormobility,whousedtheelectromechanicalgaittrainershowedan
improvementinwalkingspeedoverusualpracticewhichwasmaintainedafter6
months.Inaddition,patientsintheelectromechanicalgaittrainergroupshowed
aclinicallysignificantimprovementinRivermeadMobilityIndex(posttreatment
andat3,6monthsfollowup)andtheBarthelIndex(6monthsfollowup).The
recentpublicationofastudybyMoroneandcolleagues(2012)179witha2year
followupshowedthatroboticgaittrainingimprovedperformanceinthe
RivermeadMobilityandtheBarthelindexaswellasintheFunctional
Ambulationcategories.However,thiseffectwasrestrictedtothosewithmore
severelyimpairedmotorfunctioningattheoutset.Itwasasmallstudywith12
participantsineacharmandtheconfidenceineffectswasveryvariableranging
frommoderatetoverylow.
Otherconsiderations
Thestudiesexaminedtwodifferentelectromechanicalgaittrainerswhichvaryin
designandmayfeedintodifferentphysiologicalmechanisms.
TheLokomatisadrivengaitorthosiswithelectricaldrivesinkneeandhipjoints
with4forcetransducerswith4amplifiersthatautomateslocomotiontherapy
NationalClinicalGuidelineCentre,2013.
517
StrokeRehabilitation
Movement
onatreadmill. Theorthosisisadaptabletosubjectsfemurlength.
RehaStimisanelectromechanicalgaittrainerwith2footplateswhose
movementssimulatedstanceandswingphases.Steplengthandwalkingspeed
arecontinuouslyadjustablebyaservomotor.
Bothuseanelementofbodyweightsupport.Futurestudiesshouldaddressthe
underlyingmechanismsofaction.Thistypeofinterventionisusedinsomeunits
butnotcommonlywithintheUK.
TherewasagreementbytheGDGthattheevidence,basedontwostudies(Pohl
etal.,2007andMoroneetal.,2012)209)179infavouroftheRehaStimtrainer
wasnotstrongenoughtomakearecommendationforusewithintheNHS,but
thattheinterventionshowedpromiseandanonlyinresearchrecommendation
shouldbemade.
13.10 Anklefootorthoses
AnAnkleFootOrthosis(AFO)isanappliancedesignedtosupportthefootandankle.Afterstroke,it
istypicallyprescribedforwalkingproblemswherethefootneedstobehelduptopreventdragging
(footdrop)and/ortogivesupporttotheankletopreventthelegfromcollapsingoverthefootand
ankleinstance.TherearemanydifferentAFOs,buttwocommontypesarethosewhicharerigid
whichoffergreaterstabilityandthosethatarehingedwhichofferhelpwithdorsiflexionbutless
stabilityatthesubtalarjoint.AFOsmaybecustommadeorofftheshelfandcanbemadefrom
widerangeofdifferentmaterials.AssessmentforuseofanAFOshouldbecarriedoutbyan
appropriatelytrainedprofessional.AnAFOisanadjuncttotherapyandthusshouldbeconsideredin
thecontextofacomprehensiverehabilitationprogramwithinputfromamultidisciplinaryteam.
13.10.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
AnkleFootorthosesofalltypestoimprovewalkingfunctionversususualcare?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention:
Comparison:
Usualcare
Outcomes:
Alltypesoforthosesincluding:
SoftandScotchcasts
Splint
Brace
Lowtemperaturesplints
AnkleFootOrthosis(AFO)
GroundReactionAnkleFootOrthosis(GRAFO)
DynamicAnkleFootOrthosis(DAFO)
Walkingspeed:6minutewalktest,10metretimedwalk
LowerlimbMAS(stairs)
Walkingendurance
FunctionalIndependenceMeasure(FIM)/BarthelIndex
RivermeadMobilityIndex
Cadence
Gaitsymmetry(stancetime,steplength)
QualityofLifeoutcomes
NationalClinicalGuidelineCentre,2013.
518
StrokeRehabilitation
Movement
13.10.1.1
Clinicalevidence
Searcheswereconductedforsystematicreviews(ofrandomizedcontrolledtrials(RCTs)andcohort
studies)andRCTsthatcomparedtheeffectivenessofalltypesofanklefootorthoseswithusualcare
toimprovewalkingfunctionforadultsandyoungpeople16orolderwhohavehadastroke.Only
studieswithaminimumsamplesizeof20participants(10ineacharm)andincludingatleast50%of
participantswithstrokewereselected.5RCTs(2paralleland3crossoverRCTs)wereidentified.
Table117:
Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Beckerman
199621
Strokepatients
aged1875years
withwalking
problemscaused
byaspastic
equinesor
equinovarus
positionofthe
foot;atleast4
monthspost
stroke
AnkleFootOrthosis
(andplacebo
thermocoagulation):
AFO:Polypropylene
AFOwascustom
madeforeach
patient.AFOin5
dorsiflexion,
correctedforshoe
heelheight,designed
topreventanequinus
orequinovarus
positionofthefoot
duringwalkingand
inhibitthesynergistic
extensionpattern.
Placebo
thermocoagulation:
needleplacedinto
thetibialnerve,
localizedand
anesthetizedwithno
radiofrequency
output.
Studyduration:3
months.(N=16)
PlaceboAnkle
FootOrthosis
(andplacebo
thermocoagulatio
n)
PlaceboAFO:
Polypropylene
hingedAFOthat
allowsnormal
rangeofmotion
ofdorsiflexion
andplantar
flexion.
Placebo
thermocoagulatio
n:needleplaced
intothetibial
nerve,localized
andanesthetized
withno
radiofrequency
output.(N=14)
Walkingability
measuredwiththe
SicknessImpact
Profile(SIP)
Walkingspeed
(m/sec)
deWit200457
Strokepatients
aged4075years,
atleast6months
poststroke.
Walkingwithplastic, Walkingwithout
nonarticulatedAnkle AFO(N=10)
FootOrthosis(AFO)
(3types:AFOwitha
smallorlarge
posteriorsteelorwith
twocrossedposterior
heelreinforcements
andanopenheel)
Studyduration:at
least6months.
(N=10)
Walkingspeed
(cm/sec)
TimedUpandGo
(TUG)test(sec)
Stairstest
Erel,201177
Poststroke
patientsofatleast
6monthsduration
(chronichemi
pareticpatients);
atacognitivelevel
tounderstandthe
DynamicAnkleFoot
Orthoses(DAFO)
fabricatedbya
physiotherapist.
Fabricationtimewas
23daysonaverage.
DAFOwereworn
TimedUpandGo
Test(sec)
TimedUpStairs
TimedDownStairs
Walkingspeed
(m/sec)
NationalClinicalGuidelineCentre,2013.
519
Controlgroup
woreonlytennis
shoes.(N=16)
StrokeRehabilitation
Movement
STUDY
POPULATION
aimofthestudy;
wereatlevel35
accordingto
Functional
Ambulation
Classification;had
amaximum
spasticitylevelof3
accordingtothe
ModifiedAshworth
Scale;hadarange
ofpassive
dorsiflexionupto
atleast90
degrees;were
above18years
old.
INTERVENTION
insidetennisshoe.
Studyduration3
months.(N=16)
COMPARISON
OUTCOMES
Tyson2001266 Strokepatients
agedover18years
withhemiplegia
(severe
impairments)
HingedAFOmadefor
eachpatientbyan
orthotistusing4mm
polypropylenewitha
metalanklejointand
adjustableplantar
flexionstopwhich
wassettoprevent
plantarflexionbut
allowedfull
dorsiflexion.
Studyduration:1
month.(N=12)
Noorthosis.
(N=13)
stridelengthofWeak
andsoundleg(cm)
steplengthofWeak
andsoundleg(cm)
Stepsymmetry
Walkingspeed
(m/sec)
Cadence(step/min))
Tyson2009265
AnkleFootOrthosis
(AFO)individually
fittedforeachpatient
(N=20)
Nodevice:
Walkingwithout
theorthosis.
(N=20)
Walkingspeed
(m/sec)
Steplengthofthe
weakleg(m)
Strokepatients
withsevere
walkingproblems
for2weekspost
stroke.
NationalClinicalGuidelineCentre,2013.
520
StrokeRehabilitation
Movement
Comparison:AnkleFootOrthosis(AFO)ofalltypesversususualcare
Table118:
AnkleFootOrthosisversususualcareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Limitations
Inconsistency
Indirectness
Dynamic/An
kleFoot
Orthosis
Mean
(SD)/median
Imprecision (range)
Effect
Usualcare
Mean(SD)/
median
(range)
Mean
difference
(95%CI)
Mean
Difference
(MD)(95%CI)
Confidence
(ineffect)
Walkingspeed(10mwalkway)(cm/sec)(patientsworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbyhighervalues)
deWit,
200457
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(c)
49.6(24.3)
44.9(24.0)
4.8(0.76,
8.84)
MD4.8higher
(0.76to8.84
higher)
Low
3.6(5.59,
1.61)
MD3.6lower
(5.59to1.61
lower)
Low
Timedupandgo(sec)(patientsworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbylowervalues)
deWit,
200457
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecisio
n(d)
25.6(11.7)
29.2(12.9)
Stairstest(sec)(patientsworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbylowervalues)
deWit,
200457
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecisio
n(e)
73.0(37.8)
81.6(44.4)
8.6(
14.46,
2.74)
MD8.6lower
(14.46to2.74
lower)
Low
39.3(13.7)
43.8(14.0)
4.5(0.91,
8.09)
MD4.5higher
(0.91to8.09
higher)
Moderate
39.4(14.3)
44.3(14.1)
4.9(1.48,
8.32)
MD4.9higher
(1.48to8.32
higher)
Moderate
SoundStridelength(cm)(1monthfollowup)(Betterindicatedbyhighervalues)
Tyson,
2001266
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecisio
n
WeakStrideLength(cm)(1monthfollowup)(Betterindicatedbyhighervalues)
Tyson,
2001266
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecisio
n
Soundsteplength(cm)(1monthfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
521
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Tyson,
2001266
Design
Limitations
Inconsistency
Indirectness
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Dynamic/An
kleFoot
Orthosis
Mean
(SD)/median
Imprecision (range)
Very
serious
imprecisio
n(h)
Effect
Usualcare
Mean(SD)/
median
(range)
Mean
difference
(95%CI)
Mean
Difference
(MD)(95%CI)
Confidence
(ineffect)
19.4(9.9)
20.8(9.6)
1.40(
1.44,4.24)
MD1.4higher
(1.44lowerto
4.24higher)
Verylow
Very
serious
imprecisio
n(h)
21.7(9.5)
23.7(11.7)
2.0(
1.81,5.81)
MD2higher
(1.81lowerto
5.81higher)
Verylow
Very
serious
imprecisio
n(h)
2.6(4.9)
3.0(7.8)
0.4(
1.34,2.14)
MD0.4higher
(1.34lowerto
2.14higher)
Verylow
Noserious
imprecisio
n
53.1(16.8)
62.5(17.2)
9.4(3.44,
15.36)
MD9.4higher
(3.44to15.36
higher)
Moderate
0.18(0.1)
0.25(0.1)
0.07(0.04,
0.10)
MD0.07higher
(0.04to0.10
higher)
Low
Weaksteplength(cm)(1monthfollowup)(Betterindicatedbyhighervalues)
Tyson,
2001266
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
StepSymmetry(1monthfollowup)(Betterindicatedbyhighervalues)
Tyson,
2001266
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Cadence(1monthfollowup)(Betterindicatedbyhighervalues)
Tyson,
2001266
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Walkingspeed(m/sec)(1monthfollowup)(Betterindicatedbyhighervalues)
Tyson,
2001266
RCT
crossov
ertrial
Serious
limitations
(a)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecisio
n(c)
ImprovementinSicknessImpactProfile(SIP)ambulation(3monthsfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
522
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Beckerman,
199621
Dynamic/An
kleFoot
Orthosis
Mean
(SD)/median
Imprecision (range)
Design
Limitations
Inconsistency
Indirectness
RCT
single
blind
Serious
limitations
(b)
Noserious
inconsistency
Noserious
Very
indirectness serious
imprecisio
n(h)
Effect
Usualcare
Mean(SD)/
median
(range)
Mean
difference
(95%CI)
Mean
Difference
(MD)(95%CI)
0.00(9.0,
15.8)
2.26(
3.37,7.89)
(f)
MD2.26higher
(3.37lowerto
7.89higher)
3.05(1.5,10)
0.84(11.2,
8.6)
0.8(3.3,
4.90)(f)
MD0.8higher
Verylow
(3.3lowerto4.9
higher)
2.52(1.9,
10.2)
1.02(9.9,
8.1)
2.27(
1.85,6.39)
(f)
MD2.27higher
(1.85lowerto
6.39higher)
Verylow
0.01(0.07,
0.19)
0.01(
0.05,0.07)
(f)
MD0.01higher
(0.05lowerto
0.07higher)
Low
0.02(0.21,
0.87)
0.06(
0.02,0.14)
(f)
MD0.06higher
(0.02lowerto
0.14higher)
Moderate
MD4.28lower
(11.2lowerto
Low
3.23(12.8,
13.2)
Confidence
(ineffect)
Verylow
ImprovementinSIPphysicaldimension(3monthsfollowup)(Betterindicatedbyhighervalues)
Beckerman,
199621
RCT
single
blind
Serious
limitations(b
)
Noserious
inconsistency
Noserious
indirect
ness
Very
serious
imprecisio
n(h)
ImprovementinSIPtotalscore(3monthsfollowup)(Betterindicatedbyhighervalues)
Beckerman,
199621
RCT
single
blind
Serious
limitations(b
)
Noserious
inconsistency
Noserious
Very
indirectness serious
imprecisio
n(h)
Walkingspeedcomfortablewithshoes(m/sec)(3monthsfollowup)(Betterindicatedbyhighervalues)
Beckerman,
199621
RCT
single
blind
Serious
limitations(b
)
Noserious
inconsistency
Noserious
Serious
indirectness imprecisio
n(c)
0.05(0.15,
0.17)
Walkingspeedmaximalsafewithshoes(m/sec)(3monthsfollowup)(Betterindicatedbyhighervalues)
Beckerman,
199621
RCT
single
blind
Serious
limitations(b
)
Noserious
inconsistency
Noserious
Noserious
indirectness imprecisio
n
0.04(0.33,
0.18)
TimedUpandGoTest(sec)(patientsneverworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbylowervalues)
Erel,201177
RCT
not
Serious
limitations
Noserious
inconsistency
Noserious
indirectness
Serious
imprecisio
14.79(10.36)
19.07(8.19)
NationalClinicalGuidelineCentre,2013.
523
4.28(
11.20to
StrokeRehabilitation
Movement
Summaryoffindings
Qualityassessment
Noof
studies
Design Limitations
blinded (g)
Inconsistency
Indirectness
Dynamic/An
kleFoot
Orthosis
Mean
(SD)/median
Imprecision (range)
n(d)
Effect
Usualcare
Mean(SD)/
median
(range)
Mean
difference
(95%CI)
2.64)
Mean
Difference
(MD)(95%CI)
2.64higher)
Confidence
(ineffect)
TimedDownStairs(sec)(patientsneverworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbylowervalues)
Erel,201177
RCT
Serious
not
limitations
blinded (g)
Noserious
inconsistency
Noserious
indirectness
Very
serious
imprecisio
n(h)
13.29(11.21)
15.36(8.37)
2.07(
9.40to
5.26)
MD2.07lower
(9.4lowerto
5.26higher)
Verylow
TimedUpStairs(sec)(patientsneverworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbylowervalues)
Erel,201177
RCT
Serious
not
limitations
blinded (g)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecisio
n(i)
12(10.21)
15(7.29)
3(9.57to MD3lower
3.57)
(9.57lowerto
3.57higher)
Low
Walkingspeed(m/sec)(patientsneverworeAFObeforetreatment)(posttreatmenteffect)(Betterindicatedbyhighervalues)
Erel,201177
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
RCT
Serious
not
limitations
blinded (g)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecisio
n
0.99(0.45)
0.72(0.20)
0.27(0.01
to0.53)
MD0.27higher
(0.01to0.53
higher)
Moderate
Randomizationnotclear.
Inadequateallocationconcealment.
MeandifferencedidnotreachtheagreedMIDof20cm/sec
MeandifferencedidnotreachtheagreedMIDof10sec
MeandifferencedidnotreachtheagreedMIDof15sec
Resultswereadjustedforbaselinedifferenceswithrespecttoage,periodpoststroke,andquadricepsstrength.
Unblinded,randomisationandallocationconcealmentnotclear
ConfidenceintervalcrossedbothendsofdefaultMID.
ConfidenceintervalcrossedoneendofdefaultMID.
Narrativesummary
ThefollowingstudyissummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
NationalClinicalGuidelineCentre,2013.
524
StrokeRehabilitation
Movement
Tyson,2009265comparedAnkleFootOrthosis(AFO)uswithnoAFOuseinacrossovertrial.Outcomesreportedwerefunctionalmobility(measuredwith
theFunctionalAmbulatoryCategory[FAC]scores)andwalkingimpairments(walkingspeedandsteplength).Thestudydesignhadseriouslimitationsas
therewasnoclearrandomization[NB:Therandomisationwastheorderofthe5differenttrialconditions]andtheoutcomeassessorswerenotblinded.
AuthorsreportedthatfunctionalmobilityimprovedsignificantlywithAFOuse(P=.0001),whilethewalkingimpairmentswereunchanged(mean
difference=0;P[speed(m/s)]=0.935,P[weaksteplength(m)]=0.998.Thestudyincludedseverelyimpairedacutestrokepatientswhowerenotwalking
outsideofPhysiotherapytreatments.
NationalClinicalGuidelineCentre,2013.
525
StrokeRehabilitation
Movement
13.10.1.2
Economicevidence
Literaturereview
NorelevanteconomicevaluationscomparingAnkleFootorthoseswithusualcarewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
AnexpertadvisortotheGDGprovidedcostsforAFOssimilartotheonesinDeWit,2004study57
includedintheclinicalreview(thesewereprefabricated,thatisnotcustommade):
AFOwithsmallposteriorstrut,30.90+VAT
AFOwithbigposteriorstrut,35.54+VAT
AFOwithtwocrossedposteriorstruts,51.05
CustommadeAFOswouldbemadebyamemberofspecialistmultidisciplinaryorthoticsteamand
wouldincurhighercosts.Inaddition,therewouldbepersonnelcostsrelatedtothetimerequiredto
fit,trialandadjusttheAFOtotakeintoaccountthespecificpatientsneeds.TheGDGhassuggested
that,inmostcases,anorthotistwouldbeperformingthistask.Adjustmentsmaybemadebyeither
orthotistsandexperiencedphysiotherapistsoroccupationaltherapists(band6or7),dependingon
therequirements(forexampleorthotiststendtomakepermanentandmorecomplexadjustments).
Theestimatedcostsrangefrom45to59perhourofclientcontacts.
Evidencestatements
13.10.1.3
Clinicalevidencestatements
Onestudy57of20participantsfoundthattherewasastatisticallysignificantimprovementinthe
groupwithAnkleFootOrthosiscomparedwiththeusualcaregroup(posttreatmenteffect)inthe
followingoutcomes:
Walkingspeed(cm/sec)(LOWCONFIDENCEINEFFECT)
TimedUpandGo(TUG)test(sec)(LOWCONFIDENCEINEFFECT)
stairstest(LOWCONFIDENCEINEFFECT)
Onestudy266of25participantsfoundthattheAnkleFootOrthosisgroupwasassociatedwitha
statisticallysignificantimprovementcomparedwiththeusualcaregroupatonemonthinthe
followingoutcomes:
soundandweakstridelength(MODERATECONFIDENCEINEFFECT)
cadence(stepfrequency)(MODERATECONFIDENCEINEFFECT)
s EstimatedbasedondataandmethodsfromthePersonalSocialServicesResearchUnitUnitcostsofhealthandsocial
carereportandAgendaforChangesalarybands6and751(typicalsalarybandidentifiedbyclinicalGDGmembers).
Assumedthatanorthotistiscostedsimilartoaphysiotherapist.
NationalClinicalGuidelineCentre,2013.
526
StrokeRehabilitation
Movement
walkingspeed(m/sec)(LOWCONFIDENCEINEFFECT)
Onestudy266of25participantsfoundthattherewasnosignificantdifferencebetweenthegroup
withAnkleFootOrthosisandthegroupwithoutAnkleFootOrthosisatonemonthinthefollowing
outcomes:
soundandweaksteplength(VERYLOWCONFIDENCEINEFFECT)
stepsymmetry(VERYLOWCONFIDENCEINEFFECT)
Onestudy21of30participantsshowednosignificantdifferencebetweenthegroupwithAnkleFoot
Orthosisandtheusualcaregroupat12weeksinthefollowingoutcomes:
walkingabilityusingSicknessImpactProfilescores:totalscore(VERYLOWCONFIDENCEIN
EFFECT),ambulation(VERYLOWCONFIDENCEINEFFECT)andphysicaldimension(VERYLOW
CONFIDENCEINEFFECT)
walkingspeed:comfortablewithshoes(LOWCONFIDENCEINEFFECT)andmaximalsafe,with
shoes(m/sec)(MODERATECONFIDENCEINEFFECT)
Onestudy77of32participantsfoundnosignificantdifferencebetweenthegroupwithAnkle/Foot
Orthosesandtheusualcaregroup(posttreatmenteffect)inthefollowingoutcomes:
TimedUpandGo(TUG)test(sec)(LOWCONFIDENCEINEFFECT)
TimedDownStairs(sec)(VERYLOWCONFIDENCEINEFFECT)
TimedUpStairs(sec)(LOWCONFIDENCEINEFFECT)
Onestudy77of32participantsshowedastatisticalsignificantimprovementintheAnklefoot
orthosesgroupcomparedtothegroupthatreceivedusualcareinwalkingspeed(m/sec)(post
treatmenteffect)(MODERATECONFIDENCEINEFFECT)
13.10.1.4
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
13.10.2
Recommendationsandlinktoevidence
103. Consideranklefootorthosesforpeoplewhohavedifficulty
withswingphasefootclearanceafterstroke(forexample,
trippingandfalling)and/orstancephasecontrol(for
example,kneeandanklecollapseorkneehyperextensions)
thataffectswalking.
104. Assesstheabilityofthepersonwithstroketoputonthe
anklefootorthosisorensuretheyhavethesupportneeded
todoso.
105. Assesstheeffectivenessoftheanklefootorthosisforthe
personwithstroke,intermsofcomfort,speedandeaseof
walking.
106. Assessmentforandtreatmentwithanklefootorthoses
shouldonlybecarriedoutaspartofastrokerehabilitation
programmeandperformedbyqualifiedprofessionals.
NationalClinicalGuidelineCentre,2013.
527
StrokeRehabilitation
Movement
107. Forguidanceonfunctionalelectricalstimulationforthe
lowerlimbseeFunctionalelectricalstimulationfordropfoot
ofcentralneurologicalorigin(NICEinterventionalprocedure
guidance278).
Relativevaluesofdifferent
outcomes
EffectiveAFOsshouldleadtoimprovementsinwalkingspeedand
endurance.Anumberoffactorsareimportantindeterminingthelong
termeffectivenessofanAFO,includingcomfortandtheabilitytoputon
theAFOeasily.SomeofthestudiesconsideredbytheGDGmaybe
regardedasefficacytrials(Tyson,2001266)inthattheyexamined
immediatebenefitsandnotlongtermoutcomes.Attentionneedstobe
paidtolongtermfunctionaloutcomeswithinthehomeandcommunity.
Tradeoffbetweenclinical
benefitsandharms
Noharmswerereportedinthestudiesreviewed.TheGDGagreedankle
footorthoses(AFOs)shouldhaveabiomechanicalrationale(toimprove
function),shouldbecomfortableandwellfittedtopreventpainand
pressuresores.
Economicconsiderations
Nocosteffectivenessstudieswereidentifiedforthisquestion.The
typicalcostofAFOswasestimatedtobebetween30and51
dependingonthetype.CustommadeAFOSwouldcostmore.Inaddition
thereissomepersonneltimerequiredtomakeadjustmentsforthe
patient.TheGDGconsideredthatinselectedpatientstheadditionalcost
ofAFOs,bothpremadeandcustommade,hadthepotentialtobeoffset
bybenefitstothepatientintermsofimprovedfunction,andtherefore
improvedqualityoflife.TheGDGwereawarethatlimiteduseismadeof
manyprescribedorthoseswithsignificantcostimplications.
Qualityofevidence
ThreesmallstudiesdemonstratedthattheuseofanAFOresultedina
statisticallysignificanteffectonvelocityatposttreatment(deWitt,
2004,Erel,2011,Tyson,200177.57,266).ThestudybyErel(2011)
demonstratedaclinicallysignificantimprovementinwalkingspeedin
theanklefootorthosesgroup.
Confidenceintheeffectsshownforthisoutcomerangedfromverylow
tomoderateduetolimitationsinstudydesignandthemeandifference
notreachingtheminimalimportantdifferenceintwoofthestudies(de
Witt2004andTyson200157,266).ItwasnotedbytheGDGthatthe
effectsshownintheDeWitstudy57maybeunderestimatedbecausea
flexibleAFOwasused.InclinicalpracticearigidAFOwouldnormallybe
usedifthepatientwasveryimmobile.Themobilityofthepatients
withinthisstudywaspoorthereforeforthepatienttheresultsmaybe
consideredhighlyclinicallysignificant.
Inonestudy(byBeckerman199621)someparticipantshadalreadyused
anklefootorthoses(AFOs)whichmayhaveintroducedabiasbutitis
unclearwhatdirectionthebiaswouldeffect.Thepopulationwere
stableandwalkedindependentlyatamedianvelocityof.32.45m/s,
however,therewasalargerangeofwalkingspeedsineachgroupand
noothertreatmentinterventions.Therefore,thoughthestudyshowed
noeffect,itisunclearhowmuchdifferencewouldbeseenwithtraining.
TheGDGconsideredtheTyson2009study265tobeoneofftests,and
althoughitdemonstratedastatisticallysignificantdifferenceinfavourof
AFOitisunclearhowthesewilltranslateintohomeandcommunity
settings.TheGDGconcludedthatfurtherresearchneedstobe
undertakentoevaluatetheuseofAFOsinthecommunitysetting.
Otherconsiderations
TheGDGconsideredthatAFOsareusedtosupportswingphasefoot
clearancetopreventtrippingorfallingandstancephasecontrolto
NationalClinicalGuidelineCentre,2013.
528
StrokeRehabilitation
Movement
preventtheankleofkneecollapsingandthereforeAFOsshouldbe
consideredforpatientswhohavethesedifficulties.Theviewofthe
groupwasthatAFOsimprovewalkingspeedinselectedpatientsandthe
studiesrevieweddemonstratethis.Thepersonwouldneedtobeableto
putontheAFOthemselvesorhaveafamilymember/carerabletodo
thisforthem.
TheGDGagreedthatallstrokeunitsshouldhaveaccesstoanorthotics
service.AFOsshouldonlybeprovidedafterassessment,fittingandtrial
byanappropriatelytrainedandskilledmultidisciplinaryteam.Patients
shouldbeofferedregularreviewandfollowuptoensurecomfort,the
appropriatenessoftheprescriptiontotheindividualsdaytoday
requirementsandtoensureregularuse.
NationalClinicalGuidelineCentre,2013.
529
StrokeRehabilitation
Selfcare
14 Selfcare
14.1 Intensityofoccupationaltherapyforpersonalactivitiesofdaily
living
PersonalActivitiesofDailyLiving(PADLs)arethosetaskswhichallofusundertakeeverydayofour
lives in order to maintain our level of care (Hopson, 1981) for example, eating, washing, brushing
teeth,anddressing.
A core aspect of Occupational Therapy is the skilled analysis of performance and the impact of
physical, sensory, psychological and emotional domains on function. Specific therapeutic goals are
thenset,andtreatmentdeliveredwhichtargetsfunctionalperformanceforexample,dressinginthe
contextofthephysical,sensoryorcognitiveimpairments.Gradingofactivitiesisoftenafeatureof
the intervention so that activities increase in complexity as patients develop necessary skills. The
theoretical perspective of occupational therapy is twofold, using restorative and compensatory
approachestointervention.
14.1.1
14.1.1.1
Evidencereview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
intensiveoccupationaltherapyfocusedspecificallyonpersonalactivitiesofdailyliving
(dressing/others)versususualcare?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehadastroke
Intervention:
Intensiveoccupationaltherapy(OT)dressing,grooming,bathing,
feeding/eating,washing,toileting
Comparison:
Usualcare(OTonceaweek)/nocare
Outcomes:
NottinghamExtendedActivitiesofDailyLiving(NEADL)
ExtendedActivitiesofDailyLiving(EADL)
FunctionalIndependenceMeasure(FIM)
BarthelIndex
NottinghamStrokeDressingAssessment
NorthwickParkNursingDependencyScale
RivermeadMobilityIndex
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparingtheclinicalandcost
effectivenessofintensiveoccupationaltherapyfocusedonpersonalactivitiesofdailylivingwith
usualcareornocareinadultsoryoungpeopleof16yearsoldorolderafterstroke.Onlystudieswith
aminimumsamplesizeof20participants(10ineacharm)wereselected.Weincludedseven(7)
RCTs.
Table119summarisesthepopulation,intervention,comparisonandoutcomesforeachofthe
studies.
Table119:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
43
Chiu,2004
POPULATION
INTERVENTION
COMPARISON
OUTCOME
Patientswitha
strokewhoareable
tofollow
Anadditionalhome
basedtraining
programmeon
Predischargehome
visit(mean1.39
visits)butno
Functional
Independence
Measure(FIM)
NationalClinicalGuidelineCentre,2013.
530
StrokeRehabilitation
Selfcare
STUDY
POPULATION
instructionsand
verbally
communicateand
whoarelivingat
homewithfamily
support.
INTERVENTION
COMPARISON
bathingdevices
treatmentpost
includingatleast2
discharge.(N=23)
visitsbutnotmore
than3(mean2.74)
1stvisit:
explanationof
deviceandsafety;
2ndvisit:
opportunityto
discussproblems
usingdevicesand
devicescheckedfor
fitandsafety;
3rdvisit:optional,
dependingon
patient'sproficiency
usingdevice.(N=30)
OUTCOME
Corr,199548
Firstandsecond
strokepatients.
(Mediannumberof
dayssincestroke=
50).
Teachingnewskills,
facilitating
independencein
activitiesofdaily
living,facilitating
returnoffunction,
enablingpatientsto
useequipment,
givinginformation
topatient/carer;
referringtoor
liaisingwithother
agencies.Home
visitsbyanOTafter
dischargeand
offeredfurther
rehabilitationand
reviewedat2,8,16
and24weeks.
(N=55)
Usualcare.(N=55)
BarthelIndex
Nottingham
Extended
ActivitiesofDaily
Living
Gilbertson,
200091
Patientswitha
clinicaldiagnosisof
stroke(excluding
subarachnoid
haemorrhage)
withoutsevere
cognitiveor
communication
problems.
Domiciliary
occupational
therapy(OT):6
weekprogramme
(around10sessions
lasting3045
minutes)tailoredto
recoverygoalsset
bypatient(for
exampleregaining
selfcareor
domesticorleisure
activities).
(N=67)
Inpatient
multidisciplinary
rehabilitation,pre
dischargehome
visitforselected
patients,support
servicesand
equipment,regular
reviewin
multidisciplinary
strokeclinicand
selectedpatients
referredtoday
hospital.(N=71)
BarthelIndex
Nottingham
Extended
ActivitiesDaily
Living
Logan,1997158
Firsttimestroke
patientsdischarged
fromhospitaland
referredto
Enhanced
occupational
therapy(OT):equal
accesstoaidsand
Usualcare;85mean Extended
minutespertherapy
ActivitiesofDaily
forameannumber
Living(EADL)
of2.5visits(N=58). BarthelIndex
NationalClinicalGuidelineCentre,2013.
531
StrokeRehabilitation
Selfcare
STUDY
POPULATION
occupational
therapy
department.
INTERVENTION
COMPARISON
budgetsfor
adaptations.
Patientsseenand
treatedbyasingle
researchOT(sooner
thanpossiblewith
routineservice)for
adurationof240
meanminutesper
therapyfor
meannumberof6
visits.(N=53)
OUTCOME
Parker,2001199
Patientswithstroke
notmorethan6
monthswithout
severeillnessand
nodocumented
historyofdementia
Treatmentgoals:
Nocare(N=157).
improving
independencein
selfcaretaskssuch
aswashing,dressing
orbathing.Home
occupational
therapy(total10
sessions)lasting
notlessthan30
minuteseach
sessionforupto6
months(N=156).
Nottingham
ExtendedADL
BarthelIndex
Sackley,2006226 Participantsin
residentialhomes
withmoderateto
severestroke
relateddisability(BI
score4to15)were
included.
Aimedatimproving
independencein
personalactivities
ofdailyliving,such
asfeeding,dressing,
toileting,bathing,
transferringand
mobilizing.
Frequencyand
durationof
therapieswas
dependenton
residentand
therapistsagreed
goals(over3
months).Median
visitsperresident
permonth:2.7
(range125);time
spentwiththerapist
perresidentper
month4.5hours
(range110);most
sessionslasted
around30minutes.
(N=63)
Usualcare
(occupational
therapistnot
involved(N=55).
BarthelIndex
Rivermead
MobilityIndex
Walker,1999278 Patientswithstroke
lessthan1month;
notbeenadmitted
tothehospital;not
Aimedatimproving
independencein
personaland
instrumental
Nocare(N=91).
Extended
ActivitiesofDaily
Living(EADL)
BarthelIndex
NationalClinicalGuidelineCentre,2013.
532
StrokeRehabilitation
Selfcare
STUDY
POPULATION
livinginanursingor
residentialhome.
INTERVENTION
COMPARISON
activitiesofdaily
living.Visitsfrom
theoccupational
therapistforupto5
months(frequency
ofvisitswasagreed
betweenthe
therapist,patient,
andcarer).Mean
numberofvisits:5.8
(SD3.3,range115).
Meanlengthof
eachvisitwas52
min(11.8,2490).
(N=94)
NationalClinicalGuidelineCentre,2013.
533
OUTCOME
StrokeRehabilitation
Selfcare
14.1.1.2
Comparison:Intensiveoccupationaltherapyfocussedonpersonalactivitiesofdailylivingversususualcare/nocare.
Table120:Intensiveoccupationaltherapyversususualcare/nocareClinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Authors
Design
Limitations
Inconsistency
Indirectness
Usual
care/no
care
Median
(IQR)/Mean
(SD)/
Frequency
(%)
Effect
Intensive
occupational
therapy
Median
(IQR)/Mean
(SD)/Frequency
(%)
Veryserious
imprecision
(b)
78.6(7.4)
73.5(10.9)
5.1(0.08to
10.28)
MD5.1
higher(0.08
lowerto
10.28
higher)
Serious
imprecision
(k
d)
1(02)
0(31)
1(0.42to
1.58)
MD1higher Low
(0.42to1.58
higher)
Noserious
imprecision
10.8(5.5)
8.2(5.2)
2.6(0.54to
4.66)
MD2.6
higher(0.54
to4.66
higher)
Moderate
Serious
imprecision
Gilbertson:17
(1519)
Gilbertson:
17(1318)
0.59(0.55,
1.73)
MD0.59
higher(0.55
Low
Imprecision
Relative
Risk/Mean
difference
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
FunctionalIndependenceMeasure(3monthsfollowup)(Betterindicatedbyhighervalues)
Chiu,
200443
RCTSingle
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Verylow
BarthelIndex(2monthsfollowup)(Betterindicatedbyhighervalues)
Gilbertson
,200091
RCTsingle
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(3monthsfollowup)(Betterindicatedbyhighervalues)
Sackley,
2006226
ClusterRCT
Singleblinded
Serious
limitations(e
)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
3
Gilbertson
RCTs3single
blinded,1
Serious
limitations(c
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
534
StrokeRehabilitation
Selfcare
Summaryoffindings
Qualityassessment
Authors
,200091;
Walker,
1999278
Sackley,
2006226
Design
ClusterRCT
Limitations
)
Inconsistency
Indirectness
Usual
care/no
Intensive
care
occupational
therapy
Median
(IQR)/Mean
Median
(SD)/
(IQR)/Mean
(SD)/Frequency Frequency
(%)
(%)
Walker:20(18 Walker:18
20)
(1620)
Sackley
(changes):0.3 Sackley
(4.2)
(changes):
2.1(3.7)
Effect
Relative
Risk/Mean
difference
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
lowerto
1.73higher)
(f)
16(120)
16(220)
(g)
(g)
Moderate
(f)
(f)
18(1520)
17(1520)
(g)
(g)
Moderate
(f)
Veryserious
imprecision
(m)
22/55(40%)
22/55(40%)
1(0.63to
1.58)
0fewerper Verylow
1000(from
148fewerto
232more)
(f)
17(1419)
17(1420)
(g)
(g)
Imprecision
(d)
Confidence
(ineffect)
BarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
Logan,
1997158
RCTSingle
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
Parker,
2001199
RCTSingle
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(score<12)(1yearfollowup)(Betterindicatedbyhighervalues)
Corr,1995 RCTSingle
48
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
BarthelIndex(1yearfollowup)(Betterindicatedbyhighervalues)
Parker,
2001199
RCTSingle
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
NationalClinicalGuidelineCentre,2013.
535
Moderate
(f)
StrokeRehabilitation
Selfcare
Summaryoffindings
Qualityassessment
Authors
Design
Limitations
Inconsistency
Indirectness
Imprecision
Usual
care/no
Intensive
care
occupational
therapy
Median
(IQR)/Mean
Median
(SD)/
(IQR)/Mean
(SD)/Frequency Frequency
(%)
(%)
Effect
Relative
Risk/Mean
difference
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
NottinghamExtendedActivitiesofDailyLiving(total)(2monthsfollowup)(Betterindicatedbyhighervalues)
Gilbertson
,200091
RCT
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(i)
27(1943)
23(1135)
4(0.43to
8.43)
MD4higher
(0.44lower
to8.44
higher)
Low
Parker:21
(1438)
Gilbertson:
33.1(18.9)
3.41(0.00,
6.82)
MD7higher
(12.37
lowerto
26.37
higher)
Low
33.3(19.5)
0.8(4.32to
5.92)
MD0.8
higher(4.32
lowerto
5.92higher)
Low
0.58(0.3to
1.1)
145fewer
per1000
(from242
fewerto35
more)
Low
NottinghamExtendedActivitiesofDailyLiving(total)(6monthsfollowup)(Betterindicatedbyhighervalues)
2Parker,
2001199
Gilbertson
,200091
RCTsSingle
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(i)
Parker:28(15
38)
Gilbertson:34.7
(18.4)
NottinghamExtendedActivitiesofDailyLiving(total)(1yearfollowup)(Betterindicatedbyhighervalues)
Parker,
2001199
RCTSingle
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(i)
34.1(19.1)
NottinghamExtendedActivitiesofDailyLiving(notabletofeed)(1yearfollowup)(Betterindicatedbyhighervalues)
Corr,1995 RCTSingle
48
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
11/55(20%)
19/55
(34.5%)
NottinghamExtendedActivitiesofDailyLiving(Notabletousethetelephone)(1yearfollowup)(Betterindicatedbyhighervalues)
NationalClinicalGuidelineCentre,2013.
536
StrokeRehabilitation
Selfcare
Summaryoffindings
Qualityassessment
Authors
Design
Corr,1995 RCTSingle
48
blinded
Limitations
Inconsistency
Indirectness
Imprecision
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
Serious
imprecision
(j)
Usual
care/no
Intensive
care
occupational
therapy
Median
(IQR)/Mean
Median
(SD)/
(IQR)/Mean
(SD)/Frequency Frequency
(%)
(%)
Effect
16/55(29.1%)
29/55
(52.7%)
0.55(0.34to 237fewer
Low
0.89)
per1000
(from58
fewerto348
fewer)
2(021)
(g)
(g)
Low
(f)
8(019)
3(018)
(g)
P<0.01(g)
(h)
Moderate
(f)
5.2(3.7)
3.5(3.1)
1.7(0.40to
3.00)
MD1.7
higher(0.4
to3higher)
Moderate
4.5(3.5)
3.4(2.7)
1.1(0.20to
2.4)
MD1.1
higher(0.2
lowerto2.4
higher)
Verylow
Relative
Risk/Mean
difference
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
NottinghamExtendedActivitiesofDailyLiving(total)(1yearfollowup)(Betterindicatedbyhighervalues)
Corr,1995 RCTSingle
48
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
(f)
3(020)
ExtendedActivitiesofDailyLiving(total)(3monthsfollowup)(Betterindicatedbyhighervalues)
Logan,
1997158
RCTSingle
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
(f)
RivermeadMobilityIndex(3monthsfollowup)(Betterindicatedbyhighervalues)
Sackley,
2006226
Cluster
Randomised
TrialSingle
blinded
Serious
limitations(e
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
RivermeadMobilityIndex(6monthsfollowup)(Betterindicatedbyhighervalues)
Sackley,
2006226
Cluster
Randomised
TrialSingle
blinded
Serious
limitations(e
)
Noserious
inconsistency
Noserious
indirectness
Veryserious
imprecision
(l)
NationalClinicalGuidelineCentre,2013.
537
StrokeRehabilitation
Selfcare
Summaryoffindings
Qualityassessment
Authors
Design
Limitations
Inconsistency
Indirectness
Imprecision
Usual
care/no
Intensive
care
occupational
therapy
Median
(IQR)/Mean
Median
(SD)/
(IQR)/Mean
(SD)/Frequency Frequency
(%)
(%)
Effect
16(1118.75)
12(617)
3(0.78to
5.22)
MD3higher Moderate
(0.78to5.22
higher)
6(018)
(g)
(g)(k)
Relative
Risk/Mean
difference
(95%CI)
Absolute
effect/
Mean
Difference
(MD)(95%
CI)orP
value
Confidence
(ineffect)
ExtendedActivitiesofDailyliving(6monthsfollowup)(Betterindicatedbyhighervalues)
Walker,
1999278
RCTSingle
blinded
Serious
limitations(c
)
Noserious
inconsistency
Noserious
indirectness
Noserious
imprecision
ExtendedActivitiesofDailyLiving(total)(6monthsfollowup)(Betterindicatedbyhighervalues)
Logan,
1997158
RCTSingle
blinded
Serious
limitations(a
)
Noserious
inconsistency
Noserious
indirectness
(f)
8(021)
(a)
Unclearrandomisationandallocationconcealment.
MeandifferencedidnotreachtheagreedMIDof17points.
(c)
Unclearblinding
(d)
MeandifferencedidnotreachtheagreedMIDof1.85points.
(e)
Unclearallocationconcealment
(f)
Imprecisioncouldnotbeassessedbecauseonlymedianandinterquartilerangesofdatareported
(g)
Relative/Absoluteeffectcouldnotbeestimatedbecauseonlymedianandinterquartilerangesofdatareported
(h)
P valueasreportedbytheauthors.
(i)
ConfidenceintervalcrossesthelowerlimitofspecifiedMID(0.9)
(j)
ConfidenceintervalcrossesoneendofthedefaultMID(0.75).
(k)
Authorsreportednosignificantdifferencebetweentheintensiveoccupationaltherapygroupandusualcaregroup.
(l)
ConfidenceintervalcrossedbothendsofdefaultMID.
(m)
ConfidenceintervalcrossesbothendsofdefaultMID(0.75to1.25)
(b)
NationalClinicalGuidelineCentre,2013.
538
Moderate
(f)
StrokeRehabilitation
Selfcare
14.1.1.3
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingintensiveoccupationaltherapywithusualcarewere
identified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
TheGDGnotedthatthemaindifferenceintermsofresourcesbetweenintensivetherapyandusual
carewasthetimeoccupationaltherapistswouldspendwithpatients.
Theestimatedcostperhourofclientcontacttforaband6occupationaltherapistis45(hospital
based)or48(communitybased).TheGDGalsonotedthattothesecostsitmaybenecessarytoadd
thecostofadditionalspecificaids(suchasbarsusedtofacilitatetheuseofbathrooms)thatcanbe
usedintheseinterventions.
14.1.1.4
Evidencestatements
Clinicalevidencestatements
Onestudy43comprising53participantsfoundnosignificantdifferenceinFunctionalIndependence
Measureat3monthsafterstrokebetweenthegroupthatreceivedintensiveoccupationaltherapy
andtheusualcaregroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy91comprising138participantsfoundasignificantdifferenceintheBarthelIndexat2
monthsinfavourofthegroupthatreceivedintensiveoccupationaltherapycomparedtotheusual
caregroup,althoughthisdifferencewasnotofclinicalimportance(LOWCONFIDENCEINEFFECT).
Onestudy226comprising118participantsfoundasignificantdifferenceintheBarthelscoresat3
monthsafterstrokeinfavourofthegroupthatreceivedintensiveoccupationaltherapycomparedto
theusualcaregroup(MODERATECONFIDENCEINEFFECT).
Threestudies91,226,278comprising441participantsfoundnosignificantdifferenceintheBarthelIndex
at6monthsfollowupbythegroupreceivingintensiveoccupationaltherapycomparedtotheusual
caregroup(LOWCONFIDENCEINEFFECT).
Onestudy48comprising110participantsfoundnosignificantdifferenceintheproportionof
participantsachievinglessthan12inBarthelscoresat1yearafterstrokebetweenthegroupthat
receivedintensiveoccupationaltherapyandtheusualcaregroup(VERYLOWCONFIDENCEIN
EFFECT).
Onestudy91comprising138participantsfoundnosignificantdifferenceintheNottinghamExtended
ADLscoresat2monthsbetweenthegroupthatreceivedintensiveoccupationaltherapyandthe
usualcaregroup(LOWCONFIDENCEINEFFECT).
t EstimatedbasedondataandmethodsfromthePersonalSocialServicesResearchUnitUnitcostsofhealthandsocial
51
carereportandAgendaforChangesalaryband6 (typicalsalarybandidentifiedbyclinicalGDGmembers).
NationalClinicalGuidelineCentre,2013.
539
StrokeRehabilitation
Selfcare
Twostudies91,199comprising451participantsfoundnosignificantdifferenceintheNottingham
ExtendedADLscoresat6monthsafterstrokebetweenthegroupthatreceivedintensive
occupationaltherapyandtheusualcaregroup(LOWCONFIDENCEINEFFECT).
Onestudy199comprising313participantsfoundnosignificantdifferenceintheNottinghamExtended
ADLscoresat12monthsafterstrokebetweenthegroupthatreceivedintensiveoccupational
therapyandtheusualcaregroup(LOWCONFIDENCEINEFFECT).
Onestudy48comprising110participantsfoundnosignificantdifferenceintheproportionof
participantsabletofeedthemselvesasmeasuredbytheNottinghamExtendedActivitiesofDaily
Livingscaleat1yearafterstrokeintheintensiveoccupationaltherapygroupcomparedtotheusual
caregroup(LOWCONFIDENCEINEFFECT).
Onestudy48comprising110participantsshowedthatasignificantlyhigherproportionofparticipants
intheintensiveoccupationaltherapygroupwereabletousethetelephoneasmeasuredbythe
NottinghamExtendedActivitiesofDailyLivingscaleat1yearafterstrokecomparedtotheusualcare
group(LOWCONFIDENCEINEFFECT).
Onestudy226comprising118participantsfoundasignificantdifferenceintheRivermeadmobility
scoresat3monthsfollowupinfavourofthegroupthatreceivedintensiveoccupationaltherapy
comparedtotheusualcaregroup(MODERATECONFIDENCEINEFFECT).
Onestudy226comprising118participantsfoundnosignificantdifferenceintheRivermeadmobility
scoresat6monthsfollowupbetweenthegroupthatreceivedintensiveoccupationaltherapyand
theusualcaregroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy278comprising185participantsfoundasignificantdifferenceintheExtendedActivitiesof
DailyLivingscoresat6monthsfollowupinfavourofthegroupthatreceivedintensiveoccupational
therapycomparedtotheusualcaregroup(MODERATECONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
14.1.2
RecommendationsandLinktoEvidence
108. Provideoccupationaltherapyforpeopleafterstrokewho
arelikelytobenefit,toaddressdifficultieswithpersonal
activitiesofdailyliving.Therapymayconsistofrestorativeor
compensatorystrategies.
Restorativestrategiesmayinclude:
- encouragingpeoplewithneglecttoattendtothe
neglectedside
- encouragingpeoplewitharmweaknesstoincorporate
botharms
- establishingadressingroutineforpeoplewithdifficulties
suchaspoorconcentration,neglectordyspraxiawhich
makedressingproblematic.
Compensatorystrategiesmayinclude:
- teachingpeopletodressonehanded
- teachingpeopletousedevicessuchasbathingand
NationalClinicalGuidelineCentre,2013.
540
StrokeRehabilitation
Selfcare
dressingaids.
109. Peoplewhohavedifficultiesinactivitiesofdailylivingafter
strokeshouldhaveregularmonitoringandtreatmentby
occupationaltherapistswithcoreskillsandtraininginthe
analysisandmanagementofactivitiesofdailyliving.
Treatmentshouldcontinueuntilthepersonisstableorable
toprogressindependently.
110. Assesspeopleafterstrokefortheirequipmentneedsand
whethertheirfamilyorcarersneedtrainingtousethe
equipment.Thisassessmentshouldbecarriedoutbyan
appropriatelyqualifiedprofessional.Equipmentmayinclude
hoists,chairraisersandsmallaidssuchaslonghandled
sponges.
111. Ensurethatappropriateequipmentisprovidedand
availableforusebypeopleafterstrokewhentheyare
transferredfromhospital,whateverthesetting(including
carehomes).
Relativevaluesofdifferent
outcomes
Theoutcomesincludedinthereviewwere:FunctionalIndependence
Measure,BarthelIndex,NottinghamExtendedADLIndex,Extended
ActivitiesofDailyLivingScaleandRivermeadMobilityIndex,
TheGDGconsideredthestudiesthatreportedFIM(Motorsubscale)and
Bartheloutcomestobethemostusefulforassessingfunctional
outcomes.
Tradeoffbetweenclinical
benefitsandharms
Providedtheinterventionisdeliveredbyanappropriatelytrained
OccupationalTherapisttheGDGdidnotconsidertheretobeany
significantharmsassociatedwiththistypeofinterventionandthatthe
benefitsgainedbybeingabletoparticipateinactivitiesofdailyliving
weresignificantintermsofpatientsqualityoflife.
TheGDGagreedthattherapiesshouldincludebotharestorative
approach(aimingtoregainfunction)andacompensatory(useofaids
andequipment)approachtohelpanindividualcompensateforresidual
impairments.Appropriateequipmentneedstobeprovidedtostroke
patientsoncedischargedfromhospital,whateverthesettingtheyare
dischargedto,includingnursinghomes.Thisprovisionwouldideallybe
followingassessmentbyanOccupationalTherapistandmayinclude
practiceandtrainingwithequipment.
Economicconsiderations
Nocosteffectivenessstudieswerefoundforthisquestion.Occupational
therapyiscurrentlyroutinelyprovidedtostrokepatients.Delivering
moreintensiveinterventionwouldrequirehigherpersonnelinput,and
possiblymoreequipment,hencemoreresourceswouldbeneeded.
However,thesemaybeoffsetbyareductioninsocialandhealthfunded
carepackagesandimprovementsinpatientsqualityoflife.
Qualityofevidence
TheGDGrecognisedthatmostofthestudieswerecommunitybasedand
thereforehaveapplicabilitytoearlysupporteddischargeandtothelong
termmanagementofstroke.
ItwasnotedthatthepatientpopulationintheSackleystudy,2006was
differentfromtheotherstudiesastheywereolderandinresidential
nursinghomes(thesepatientswereseentomaintainperformancein
comparisonwithcontrolgroup,whodeteriorated).TheGDGalso
NationalClinicalGuidelineCentre,2013.
541
StrokeRehabilitation
Selfcare
consideredthatmanagementofpatientshadchangedsincethe
publicationoftheWalkerstudy278asthesepatientshadnotbeen
admittedtohospital,howeveritwasusefulanditmayreflecta
populationwhowouldnowreceiveearlysupporteddischargeasthey
scoredhigherontheBarthelindexatbaseline.
TheGDGconsideredthestudiesincludedinthereviewtobefeasibility
studies.Confidenceintheeffectshowninmostoftheoutcomeswere
lowtomoderateduetolimitationsinstudydesign(unclear
randomisationandallocationconcealment)andimprecisionaroundthe
effectestimate.Althoughitwasfoundthattherewasaclinically
significanteffectofintensiveOTinBarthelIndexat3monthscompared
tousualcaregroup(Sackley2006226)andtheconfidenceonthiseffect
wasmoderate,thiseffectwasnotpreservedatsixmonthsfollowup.
TheParkerstudy199wastheonlylargemulticentred,RCT,howeverthis
producedequivocalresults.Theinclusionofthisdatainametaanalysis
151
(whichincludedstudieswithsmallernumbersthanincludedinthis
clinicalreview)hasshownthattherewassignificantbenefitshowninthe
intensivearmofoccupationaltherapy.
OveralltheGDGagreedstudiesshowedshorttermfunctionalgainat3
monthsbutnotoveralongerterm.Somelimitedevidenceshowedthat
functionalgainsaremaintainedat6monthsand1yearintheintensive
OTgroups(Gilbertson2000,Sackley2006,Walker199991,226,278)
TheGDGagreedthatfromtheevidenceavailableforthosepatientswith
strokewhoaremanagedwithinthecommunity,occupationaltherapy
providessomebenefit,butthereiscurrentlynoevidenceforthose
patientswithmoderatestrokewhoaremanagedintheacute(hospital)
settingandfurtherresearchisrequired.
TheGDGconsideredthatthepatientsincludedinthestudiestendedto
bethosewithmoderatestrokeandphysicallyfitter,thereforetreatment
withinthecommunityratherthaninhospitalwouldbeappropriatefor
thisparticularpopulation,butwouldnotbeapplicableforallstroke
patients.
Otherconsiderations
TheGDGrecognisedthatdefiningintensityischallengingandcan
bedefinedintermsoffrequencyoftreatment,totalamountof
treatment,durationoftreatment,ormodeofdelivery.
Theamountofoccupationaltherapymentionedinthereviewed
studiesvariedbutwastypicallylessthanthecurrent(5sessionx
45minutesperweek)recommendedinthecurrentNICEQuality
standards.Thestudiesidentifiedgavelittleindicationhowmuch
occupationaltherapyisneededbutdidindicatethatoccupational
therapyiseffective.
TheGDGnotedthatthedescriptionofoccupationaltherapy
interventionswaslimitedwithinthestudiesreviewed,butthey
didemployarangeofrestorativeandcompensatorystrategies.
Consensusrecommendationsweremadetoreflectthisand
examplesofthetypesofinterventionsdeliveredwereindicated.
NationalClinicalGuidelineCentre,2013.
542
StrokeRehabilitation
Communityparticipationandlongtermrecovery
15 Communityparticipationandlongtermrecovery
Atsomepointaftertheonsetofstroke,nofurtherchangescanreasonablybeexpectedatthe
impairmentlevel,howeverchangescanstillbemadeintermsofreintegrationintoanindividuals
family,socialandcommunitylifeRehabilitationservicesshouldaimtowithdrawonlywhenpeople
withstrokehavetheknowledge,skillsandifnecessarythesupporttheyneedtomanagethis
process.Thischapterfocussesonselfmanagement,longtermhealthandsocialsupportandreturn
totheworkplace.
Asearchforsystematicreviewevidenceforthetopicoflongtermhealthandsocialsupportofthe
personafterstroke.Therewasnoevidenceidentifiedandthereforerecommendationswerebased
onmodifiedDelphiconsensusstatementsthatweredrawnupbasedonrecommendationsin
publishednationalandinternationalguidelinesseeAppendixFfordetails.
15.1 Returntowork
Workcontributestoadultidentity,confersfinancialbenefitsandstatus,andcanimprovequalityof
lifeandreducesillhealth.TheStrokeStrategy61highlightedtheneedforpeoplewhohavehada
strokeandtheircarerstobeenabledtoparticipateinpaid,supportedandvoluntaryemployment.
TheNationalServiceFrameworkforpeoplewithlongterm(neurological)conditions60identifiedthe
needindividualsmayhaveforvocationalrehabilitationofferedbylocalorspecialistrehabilitation
servicesto:entertrainingorworkopportunities;remaininorreturntotheirexistingjob;identify
andprepareforsuitablealternativeworkoptions;planwithdrawalfromworkatanappropriatetime
(conservingpensionandotherrights);andaccessappropriatealternativeoccupationaland
educationalopportunities.
Vocationalrehabilitationisoftendeliveredbyhealthprofessionals,linkedtocommunity
rehabilitationservicesandaimstoreviewandoptimisetheskillsrequiredtoengageinmeaningful
occupation,whichmightbepaidorvoluntaryandmayormaynotbetheroletheywereemployedin
priortotheirstroke.Ideallyspecialistvocationalrehabilitationservicesarebothmultidisciplinary
andmultiagency(withrehabilitationservicesworkingalongsideJobCentrePlus),butinpractice
mostvocationalrehabilitationisdeliveredbyoccupationaltherapistsandpsychologistsbasedin
communityrehabilitationteams.
Interventionsaremosteffectivewhentheyaretailoredtotheindividualsimpairmentsinthe
contextofthedemandsoftheworkplace.Theymayincludetheuseofmemorystrategies,
computeruse,confidencebuilding,planningandpacing,aswellasliaisingwithemployersregarding
educationonstrokespecificissues,reasonableaccommodationsandgradedreturntowork
activities.
15.1.1
EvidenceReview:Inpeopleafterstrokewhatistheclinicalandcosteffectivenessof
interventionstoaidreturntoworkversususualcare?
ClinicalMethodologicalIntroduction
Population:
Adultsandyoungpeople16orolderwhohavehada
stroke
Intervention:
Jobretention
Returntowork(tailoredtotheimpairmentofthe
patientrecognisingthedemandsofthejob)
Comparison:
Usualcare(nothing)
Outcomes:
Samejobsameemployer
Samejobdifferentemployer
NationalClinicalGuidelineCentre,2013.
543
StrokeRehabilitation
Communityparticipationandlongtermrecovery
ClinicalMethodologicalIntroduction
15.1.2
Differentjobsameemployer
Differentjobdifferentemployer
Unemployment
Retiredduetoillhealth
Voluntarywork
Benefitclaims
Clinicalevidence
SearcheswereconductedforsystematicreviewsandRCTscomparinginterventionstoaidreturnto
workwithusualcareforadultsoryoungpeopleof16yearsoldafterstroke.Onlystudieswitha
minimumsamplesizeof20participants(10ineacharm)andincludingatleast50%ofparticipants
withstrokewereselected.OneRCTwasidentified.Thisstudywasconductedinamixedpopulation
ofparticipantswithacquiredbraininjury,ofwhom59.1%hadexperiencedstroke.
Table121:Summaryofstudiesincludedintheclinicalevidencereview.Forfulldetailsofthe
extractionpleaseseeAppendixH.
STUDY
POPULATION
INTERVENTION
COMPARISON
OUTCOMES
Trexler
2010262
Patientswith
acquiredbrain
injury(traumatic
braininjury
(31.8%),
intracranial
haemorrhage
(31.8%),stroke
(27.3%),other
(13.6%))aged
between1860
years.
Participantshad
beenemployed
and/orattended
schoolfor2
yearspriorto
theinjuryand
hadagoalto
returntoworkor
toschool.
Resourcefacilitation
interventiontoreturn
towork(duration6
months)including
assessmentof
patientscurrent
status,needsand
resources,planand
documentmutually
agreeduponneeds,
identificationof
communityresources
forservicesand
supports,facilitation
ofaccesstoresources
througheducationand
advocacy,proactively
monitoringofthe
statusoftheplan
throughtelephone,
internetandpersonal
contactswithpatient
orcaregiverata
minimumofevery2
weeksandprovisionof
education
(informationonthe
injury,personal
advocacyand
partnership
development)withthe
patientorcaregiver.
Formeremployerwas
involvedthrough
education,titrating
Usualcare:patients
receivedonly
recommended
servicesbytheir
healthcareproviders
(outpatient
rehabilitation
therapies,
neuropsychological
services,medical
followup).No
contactduringthe6
monthsfollowupby
aresourcefacilitator
wasmade.(N=11)
Fulltime
employment
Parttime
employment
Employment
Unemployment
NationalClinicalGuidelineCentre,2013.
544
StrokeRehabilitation
Communityparticipationandlongtermrecovery
STUDY
POPULATION
INTERVENTION
COMPARISON
returntowork
schedulesand
facilitatingutilization
ofjobsupports.(N=12)
NationalClinicalGuidelineCentre,2013.
545
OUTCOMES
StrokeRehabilitation
Communityparticipationandlongtermrecovery
Comparisonofresourcefacilitationinterventionforreturntoworkversususualcare
Table122:Resourcefacilitationinterventionforreturntoworkversususualcareclinicalstudycharacteristicsandclinicalsummaryoffindings
Summaryoffindings
Qualityassessment
Noof
studies
Design
Usualcare
Frequencies
(%)
Effect
Limitations
Inconsistency Indirectness
Imprecision
Resource
facilitation
Frequencies
(%)
Very
serious
limitations
(a)
Noserious
inconsistency
Serious
indirectness
(b)
Veryserious
imprecision
(c)
60moreper
1000(from
177fewerto
895more)
Verylow
Very
serious
limitations
(a)
Noserious
inconsistency
Serious
indirectness
(b)
Veryserious
imprecision
(c)
3/12(25%)
159moreper
1000(from61
fewerto1972
more)
Verylow
Very
serious
limitations
(a)
Noserious
inconsistency
Serious
indirectness
(b)
Veryserious
imprecision
(c)
7218more
per1000
(from131
fewerto1095
more)
Verylow
Very
serious
limitations
(a)
Noserious
inconsistency
Serious
indirectness
(b)
Veryserious
imprecision
(c)
223fewerper
1000(from
452fewerto
293more)
Verylow
Relative
Risk
(95%ci)
Absolute
effect(95%
CI)
Confidence
(ineffect)
Fulltimeemployment
1Trexler
2010262
RCT
unblinded
Parttimeemployment
1Trexler
2010262
RCT
unblinded
1/11(9.1%)
2.75
(0.33to
22.69)
Anyemployment
1Trexler
2010262
RCT
unblinded
Unemployment
1Trexler
2010262
RCT
unblinded
(a) Unblindedstudy.Nodetailsonrandomizationandunclearallocationconcealment.
(b) Patientshadexperiencedtraumaticbraininjury(TBI)withnomentiononthephaseoftheirillness.59.1%participantshadstroke.
(c) ConfidenceintervalcrossedbothendsofdefaultMID(0.75,1.25)
NationalClinicalGuidelineCentre,2013.
546
StrokeRehabilitation
Communityparticipationandlongtermrecovery
Narrativesummary
ThefollowingstudyissummarisedasanarrativebecausetheresultswerenotpresentedinnumericaldatathatcouldbeincludedintheGRADEtable:
Theauthorsoftheonceincludedstudy262alsoreportedthatthedistributionofthethreecategories(fulltime,parttimeandunemployed)ofthe
employmentitemoftheMayoPortlandAdaptabilityInventory4ParticipationIndex(M2PI)weresignificantlydifferentbetweentheresourcefacilitation
interventiongroupandtheusualcare(P<0.0001).
NationalClinicalGuidelineCentre,2013.
547
15.1.2.1
Economicevidence
Literaturereview
Norelevanteconomicevaluationscomparingoccupationaltherapytoaidreturntoworkwithusual
carewereidentified.
Interventioncosts
Intheabsenceofcosteffectivenessanalysisforthisreviewquestion,theGDGconsideredthe
expecteddifferencesinresourceusebetweenthecomparatorsandrelevantUKNHSunitcosts.
Considerationofthisalongsidetheclinicalreviewofeffectivenessevidencewasusedtoinformtheir
qualitativejudgementaboutcosteffectiveness.
TheGDGconsideredthattypicallyaband7communityoccupationaltherapistwoulddeliverthis
service.Theestimatedcostperhourofclientcontactis59u.Typicalresourceuseperpatientwas
estimatedtobeintherangeof9to15hours.
15.1.2.2
Evidencestatements
Clinicalevidencestatements
Onestudy262comprisingof23participantsshowedthattherewasnosignificantdifferenceinthe
proportionofparticipantsreturningtofulltimeemploymentbetweenthosewhoreceivedresource
facilitatorinterventionandtheusualcaregroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy262comprisingof23participantsshowedthattherewasnosignificantdifferenceinthe
proportionofparticipantsreturningtoparttimeemploymentbetweenthosewhoreceivedresource
facilitatorinterventionandtheusualcaregroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy262comprisingof23participantsshowedthattherewasnosignificantdifferenceinthe
proportionofparticipantsreturningtoemploymentbetweenthosewhoreceivedresourcefacilitator
interventionandtheusualcaregroup(VERYLOWCONFIDENCEINEFFECT).
Onestudy262comprisingof23participantsshowedthattherewasnosignificantdifferenceinthe
proportionofunemployedparticipantsbetweenthosewhoreceivedresourcefacilitatorintervention
toaidtoreturnworkandtotheusualcaregroup(VERYLOWCONFIDENCEINEFFECT).
Economicevidencestatements
Nocosteffectivenessevidencewasidentified.
15.1.3
Recommendationsandlinktoevidence
112. Returntoworkissuesshouldbeidentifiedassoonaspossible
afterthepersonsstroke,reviewedregularlyandmanaged
actively.Activemanagementshouldinclude:
identifyingthephysical,cognitive,communicationand
psychologicaldemandsofthejob(forexample,multitasking
byansweringemailsandtelephonecallsinabusyoffice)
u EstimatedbasedondataandmethodsfromPersonalSocialServicesResearchUnitUnitcostsofhealthandsocialcare
reportandAgendaforChangesalaryband751(typicalsalarybandidentifiedbyclinicalGDGmembers).
NationalClinicalGuidelineCentre,2013.
548
identifyinganyimpairmentsonworkperformance(for
example,physicallimitations,anxiety,fatiguepreventing
attendanceforafulldayatwork,cognitiveimpairments
preventingmultitasking,andcommunicationdeficits)
tailoringanintervention(forexample,teachingstrategiesto
supportmultitaskingormemorydifficulties,teachingtheuse
ofvoiceactivatedsoftwareforpeoplewithdifficultytyping,
anddeliveryofworksimulations)
educatingabouttheEqualityAct2010vandsupportavailable
(forexample,anaccesstoworkscheme)
workplacevisitsandliaisonwithemployerstoestablish
reasonableaccommodations,suchasprovisionofequipment
andgradedreturntowork.
113. Managereturntoworkorlongtermabsencefromworkfor
peopleafterstrokeinlinewithrecommendationsinManaging
longtermsicknessandincapacityforwork(NICEpublichealth
guidance19).
Relativevaluesofdifferent
outcomes
Returntoworkwastheonlyoutcomeconsideredinthisreview.TheGDG
notedtheNationalStrokeStrategywhichstatesthatpeopleshouldbe
encouragedintheirparticipationroles61andcommentedthatvocational
rehabilitationprogrammesmayenableanindividualtoimprovethestructure
totheirdayandengageinmeaningfuloccupationwhilenotreturningto
work.Thismaymakeadifferenceintermsofimprovingthequalityoflifefor
thepatientandtheircarers.However,ifpeopledonotreturntoworkthe
therapycanberegardedasafailuredespiteresultingingreaterparticipation
insocialroles.
TheGDGacknowledgedthatotherparticipationrolesareasimportantas
returntoworkandsuggestedthatfuturestudiesshouldrecordHrQoLand
participationoutcomes.TheGDGagreedthatworkisoneofarangeof
participationroleswhichneedstobeconsidered,inacomprehensive
rehabilitationprogramme.
Tradeoffbetweenclinical
benefitsandharms
Manyofthedisabilitiesthatpreventreturntoworkarenotobvioussuchas:
lowselfesteem,lowconfidence,cognitiveimpairments,fatigue,lowmood
anddepression,andtreatmentofthesedifficultiescanhaveasignificant
impactnotonlyontheabilitytoreturntoworkbutmoregenerallyonquality
oflife32,33.
Economicconsiderations
Nopublishedstudieswereidentifiedassessingcostorcosteffectiveness.
Theestimatedcostofoccupationaltherapytoaidreturntoworkwas
estimatedtobeintherangeof531to885perpatients.
TheGDGconsideredthattheadditionalcostoftheinterventionmaybe
offsetbythepotentialincreaseinpatientandcarerqualityoflife.
Qualityofevidence
ThiswasasmallunderpoweredRCTwithamixedpopulation(acutebrain
injury)with59.1%oftheparticipantshadhadastroke.Althoughthestudy
describedamixedpopulationofacutebraininjuryincludingischaemicstroke
andintracerebralhaemorrhage,theGDGagreedthatinprinciplethe
NationalClinicalGuidelineCentre,2013.
549
interventionmaybetransferableto astrokepopulationprovidedthe
impairmentsinthedifferentpathologicalgroupsweresimilar.However,the
rangeofimpairmentswasnotclearlydescribed.Thegroupalsonotedthat
fromthelimiteddescriptiongivenoftheinterventionitwouldbedifficultto
reproduceinotherstudies.Confidenceintheeffectforthereturntowork
outcomes(fulltime,parttimeemployment,anyemployment,
unemployment)wasverylow.TheGDGwasnotconfidentintheresults
presentedbecauseofthestudylimitations(unblindedwithunclear
randomizationandallocationconcealment).HowevertheGDGnotedthere
areseriousproblemsonplanning(includingagreedblinding),recruiting,
conductingandanalysingRCTsofthissort.
TheGDGagreedthatbecauseofthesedifficultiestherewouldbeveryfew
studieslookingatinterventionsforreturntowork.Incontrast,therearea
largenumberofpaperssynthesisingconsensusnarrativesofwhatneedsto
bedonebutlittleRCTorhealtheconomicevidencethatwouldpromotethe
commissioningofthistypeofintervention.Thegrouphadwishedto
ascertainwhetherRCToreconomicevidencewasavailableandthisreview
hasansweredthatquestion.
Otherconsiderations
TheconsensusviewheldbytheGDGwasthatreturningtoworkafterstroke
wasclearlythebestoutcomeforanindividualofworkingageandthisstudy
providesoneexampleofaninterventiontoaidreturntowork.TheGDG
consideredthatidentifyingneedsandanyobstaclestoreturningtowork
shouldbeexploredassoonaspossibleinordertoplanreintegrationback
intotheworkplace.Becauseofthelackofevidencethereisnosystematic
wayofassessingpeopleontheircapacityforwork,whichcomponentsofa
vocationalinterventionworkorwhatthedurationofaninterventionshould
be.Futurestudiesneedtobeconductedtoascertainwhatdoesanddoesnt
workinthisfield.
TheGDGrecognisedtheimportanceoftheEqualOpportunitiesActin
creatinganenvironmentinwhichpatientsaresupportedtoreturntowork.
TheGDGrecognisedthenationalconsensusthatexitsaroundthekey
elementsofvocationalrehabilitation,namelyanalysisoftheimpairment,the
demandsofthejob,educationofthepatient,tailoredinterventions,work
placevisitsandestablishingworkplaceaccommodations.Althoughidentified
asanimportantinterventionwithintheStrokeStrategyandtheNational
ServiceFrameworkforlongterm(neurological)conditions,theGDGagreed
thatthereislimitedresourcetodelivervocationalrehabilitation
interventionswithintheNHS.TheGDGformulatedconsensus
recommendationsbasedondiscussionofcurrentpracticeofinterventionsin
theirownrehabilitationservicesandknowledgeofothernationalguidancein
thisarea.Thegroupwereinagreementthatthiswasanareawhere
provisionneededtoimproveandthereforedirectiverecommendationswere
drafted.
TheGDGagreedthatservicesneedthecapacitytodelivervocational
rehabilitationtopeoplewithneurologicalimpairmentssuchastroke.The
grouprecognisedthatmanyworkingagepatientshavevocationalneeds
whichareunmet,oftenleadingtojobloss.Thegroupnotedthatitisthe
resilienceofthepatientandwillingnessoftheemployertosupportthe
individualwhichwilldeterminewhetherapersonreturnstoworkornot.
Thegroupagreedthatmoredataneedstobecollectedonthenumberof
peoplewhoreturntoworkafterstroke.
Randomisedcontrolledtrialsofthetrainingneedsofhealthprofessionals
deliveringvocationalrehabilitationwouldbeanareawherefurtherresearch
isneeded.TheGDGacknowledgedthathealthprofessionalsreceivevery
littletraininginthisarea.
NationalClinicalGuidelineCentre,2013.
550
ItwasnotedthatotherNICEguidanceavailableonreturntoworkis
available:Managinglongtermsicknessabsenceandincapacityforwork
(NICEpublichealthguidance19).
Researchneedstobeundertakentoestablishthestructuralprocessesthat
alloweffectiveinteragencywork.
15.2 Longtermhealthandsocialsupport
Therewasalackofdirectevidenceforthistopic.Thereforerecommendationsinthissectionwere
basedonthemodifiedDelphiconsensusstatementsthatweredraftedfromrecommendationsof
publishednationalandinternationalguidelines.Weprovidetablesofstatementsthatreached
consensusandstatementsthatdidnotreachconsensusandgiveasummaryofhowtheywereused
todrawuptherecommendations.Fordetailsontheprocessandmethodologyusedforthemodified
DelphisurveyseeAppendixF.
15.2.1
Whatongoinghealthandsocialsupportdothepersonafterstrokeandtheircarer(s)
requiretomaximisesocialparticipationandlongtermrecovery?
Population
Adultsandyoungpeople16orolderwhohavehadastroke
Components
Continuedmonitoringandreaccessintorehab
Longtermsupport/careathome
Socialparticipationactivities
Carer/familysupport&education
Outcomes
Patientandcarersatisfaction
Qualityoflife
optimisedstrategiestominimiseimpairmentandmaximiseactivity/participation
15.2.2
Delphistatementswhereconsensuswasachieved
Table123:Tableofconsensusstatements,resultsandcomments(percentageintheresultscolumn
indicatestheoverallrateofresponderswhostronglyagreedwithastatementand
amountofcommentsinthefinalcolumnreferstorateofresponderswhousedthe
openendedcommentsboxes,i.e.No.peoplecommented/No.peoplewhoresponded
tothestatement)
Number
Statement
Ifthereisanewidentifiedneedfor
furtherstrokerehabilitationservices,
thepersonwhohashadastroke
shouldbeabletoselfreferwiththe
supportofaGPorspecialist
communityservices.
Results
%
66.7
NationalClinicalGuidelineCentre,2013.
551
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Inround223/99(23%)panel
memberscommented;11/81(14%)
inround3:
Oneissuethatwashighlightedisthe
demandthismaycreate(Directself
referralcouldleadtodemand
outstrippingresourcesofthestroke
rehabilitationservice.Theredoes
needtobeanassessment.).
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
Otherpanelmembersthoughtthat
thephrasewiththesupportofwas
unclearsincethiswouldnotmean
selfreferralanymore.
weoperateselfreferralforanyone
previouslyknowntothestoke
service.
thisisunclear,howcanyouself
referwiththesupportofaGP,are
theystillbeinggatekeepersthen?
Number
Statement
Focusonlifeafterstrokemayinclude:
Informationanddiscussionabout
communityaccess
Participationincommunityactivities
Socialroles
Informationaboutdriving
Opportunitiestodiscussissuesaround
sexualfunction
75.2
72.9
70.2
76.4
68.2
Inround210/98(10%)panel
memberscommented;37/85(44%)
inround3(directpromptgivenin
round3):
Afewotherareasoffocuswere
suggested.
Returntowork/training
Relationships,childcareissues
Secondarypreventiondiet,
exercise
Psychological/emotionaladaptation
Strokegroupscommunication
supportactivities
Supportforcarers
Accesstowelfarebenefitsand
allowances,equipment
Whilethepersonwithstrokeisin
hospitallocalprocessesshouldensure
thatreferralismadetoadultsocial
careforanassessmentofneed(ifthe
personhasaneedforsocialcare).
67
Inround219/99(19%)panel
memberscommented;21/84(25%)
inround3:
Itwashighlightedthatasocial
workershouldbepartoftheMDT.
atanappropriatetimetoallowthe
socialworkertoworkalongsidethe
MDTtofullyappreciatethepatients
difficultiesandgettoknowthemand
theirfamily.Thisshouldntbe
startedrightattheendofthe
inpatientstay,butworkedup
duringtheinpatientstay.
Somepeoplecommentedthatthis
shouldbeajoinedupprocessand
happeninatimelymanner.
yes,priortodischargesothereis
notalonggapbetweenservices
NationalClinicalGuidelineCentre,2013.
552
Number
15.2.3
Statement
Results
%
Amount(No.panelmemberswho
commented/No.panelmembers
whoresponded)andcontentof
panelcommentsorthemes
endingandothersbeginning.
Acoupleofpeoplecommentedthat
thestatementwasnotveryclear(for
examplelocalprocesseswasnot
definedandalso,whowouldbe
makingtheassessmentisunclear)
Delphistatementwhereconsensuswasnotreached
Table124:Tableofnonconsensusstatementswithqualitativethemesofpanelcomments
Number
Statement
Results
Amountandcontentofpanel
commentsorthemes
1.
Reviewintervalsneedtobespecified
andagreedwiththepersonwhohas
hadastrokeinregardstotheirlong
termrehabilitationneeds.
65.0
Inround213/100(13%)panel
memberscommented;9/80(11%)in
round3:
Opinionsweredivided.Some
memberssuggestedthatthisshould
beneedsbasedandflexiblewhereas
otherssaidthatthe6and12month
followupwassufficient.
2.
Areviewofhealthandsocialcare
needsofthepersonwhohashada
strokethatisformallyreportedand/
orcoordinatedorconductedwiththe
GPservicesshouldtakeplaceatleast
(options:6months,12months,
unspecified)
44.6
Inround240/98(41%)panel
memberscommented;20/83(24%)
inround3:
Themajorityofcommentsstated,6
weeks,6monthsandthenannually.
(inaccordancewiththeNational
StrokeStrategy@6/52,6/12then
annually).
Thereweresomecomments
recommendinganeedbasedsystem
thatwouldallowmorefrequent
intervalsifnecessary.
Itwasalsocommentedthatthis
woulddependonthetimepost
discharge.
Therewasaconcernthatifitwere
tobeaneedsbasedapproach
peoplewouldnotbegivenan
opportunityforameetingunless
theyhaveaneed
Iflefttoindividualneedsittendsto
resultincrisesmanagement
NationalClinicalGuidelineCentre,2013.
553
Number
Statement
Results
Amountandcontentofpanel
commentsorthemes
meetings.Thereshouldbesomesort
ofstructureandprocesstoensure
thatreviewsarefrequentenoughto
monitorthepatientlongerterm
safelyandreasonablybutnottoo
frequenttobeunnecessaryand
possiblydevaluingthemerit
3.
Wherethepersonswhohavehada
strokearestillmakingprogresslikely
toleadtofunctionalchange,they
shouldbeofferedagoalfocused
enablingcarepackage.
56.9
Inround321/84(25%)panel
memberscommented;15/72(21%)
inround4:
Somepeoplecommentedthatthis
statementwasnotveryclearand
thatthetermenablingcarepackage
wasnotuniversallyunderstood.
Extract:
Isuspecttherewillbesomeissues
astohowyoumeasurefunctional
changeandthewordlikelyshould
therebeatimescaleputonthisas
thiscaveatwouldsuggestthatmost
patients/clientswouldfallintothis
categoryandserviceswillfindthis
verydifficulttodeliver
Somecommentsweremadeabout
thetermfunctionalchangeand
thatthestatementwasunclear
aboutwhatitmaybereferringto.
4.
Whenapersonwithstrokeleaves
56.9
hospital,thereshouldbeareviewof
thedischargeprocesswiththeperson
whohashadastroketogetherwith
theirfamilyandcarersbyamemberof
thecommunitystrokerehabilitation
team.Theaimofthisreviewisto
ensurethatthedischargeplanwas
followedandcarriedout,thattheir
currentstatusandgoalsarereviewed,
andacontinuingrehabilitationplanis
devised.
Inround222/99(22%)panel
memberscommented;16/85(19%)
inround3and11/72(15%)inround
4:
Thereweresomecommentsabout
theamountofreviewsthatwere
suggested.
Thisshouldbedoneaccordingto
needsincesomepeoplemaybe
dischargedanddonotwishorneed
apostdischargemeeting.
Willthisapplytoeverystroke
patientoronlythosedischargedwith
adisabilityIagreeitshouldbe
everystrokepatientbutthatwould
createahugeworkloadforthe
communitystroketeamIfeelthat
thisshouldbereconsideredand
reflectthevariedpoststrokeneeds
ofpatientsandtheircarers.
NationalClinicalGuidelineCentre,2013.
554
Number
5.
15.2.4
Statement
Results
Selfmanagementandtrainingneeds
formpartoflongtermhealth
educationforthepersonafterstroke.
61.1
Amountandcontentofpanel
commentsorthemes
Inround212/98(12%)panel
memberscommented;13/84(11%)
inround3and9/72(15%)inround
4:
Itwasstatedthattherewas
insufficientevidencetoconclude
thatthisworks.
Thereisalsotheissuethatit
dependsonthelevelofpoststroke
ability.
inordertosupportsecondary
preventionandmoreindependence,
educationisimportant.
selfmanagementisntjustabout
education,apersonmayneedother
interventionstofacilitatebehaviour
change.
RecommendationsandlinkstoDelphiconsensussurvey
Statements
34.Ifthereisanewidentifiedneedforfurtherstrokerehabilitation
services,thepersonwhohashadastrokeshouldbeabletoselfrefer
withthesupportofaGPorspecialistcommunityservices.
35.Focusonlifeafterstrokemayinclude:
Informationanddiscussionaboutcommunityaccess
Participationincommunityactivities
Socialroles
Informationaboutdriving
Opportunitiestodiscussissuesaroundsexualfunction
36.Whilethepersonwithstrokeisinhospitallocalprocessesshould
ensurethatreferralismadetoadultsocialcareforanassessmentof
need(ifthepersonhasaneedforsocialcare).
Forhealthandsocialcareinterfacerecommendationssee5.3.2.
114. Informpeopleafterstrokethattheycanselfrefer,usuallywiththe
supportofaGPornamedcontact,iftheyneedfurtherstroke
rehabilitationservices.
115. Provideinformationsothatpeopleafterstrokeareableto
recognisethedevelopmentofcomplicationsofstroke,including
frequentfalls,spasticity,shoulderpainandincontinence.
116. Encouragepeopletofocusonlifeafterstrokeandhelpthemto
achievetheirgoals.Thismayinclude:
facilitatingtheirparticipationincommunityactivities,suchas
NationalClinicalGuidelineCentre,2013.
555
shopping,civicengagement,sportsandleisurepursuits,visiting
theirplaceofworshipandstrokesupportgroups
supportingtheirsocialroles,forexample,work,education,
volunteering,leisure,familyandsexualrelationships
providinginformationabouttransportanddriving(including
DVLArequirements;seewww.dft.gov.uk/dvla/medical/aag).
117. Manageincontinenceafterstrokeinlinewithrecommendationsin
Urinaryincontinenceinneurologicaldisease(NICEclinicalguideline
148)andFaecalincontinence(NICEclinicalguideline49).
118. Reviewthehealthandsocialcareneedsofpeopleafterstrokeand
theneedsoftheircarersat6monthsandannuallythereafter.These
reviewsshouldcoverparticipationandcommunityrolestoensure
thatpeoplesgoalsareaddressed.
119. Forguidanceonsecondarypreventionofstroke,follow
recommendationsinLipidmodification(NICEclinicalguideline67),
Hypertension(NICEclinicalguideline127),Type2diabetes(NICE
clinicalguideline87)andAtrialfibrillation(NICEclinicalguideline
36).
120. Provideadviceonprescribedmedicationsinlinewith
recommendationsinMedicinesadherence(NICEclinicalguideline
76).
Economic
considerations
Otherconsiderations
Therearesomecostsassociatedwiththereferralandwiththereviewof
healthandsocialcareneedsintermsofstafftime;howeverthese
interventionswillalsoimprovethequalityoflifeofthepersonwith
stroke;anannualfrequencyofreview(afterthefirstreviewat6
months)wasconsideredbalancedintermsofcostsandeffectiveness.
TheGDGnotedthedifferentviewsexpressedintheDelphisurveywith
regardsbeingabletoreaccessrehabilitationservicesatalatertime
point.Somepeoplehadcommentedthatthiswouldcreatetoogreata
demandwhichcouldnotbemet,whileothersthoughtthestatement
unclearasitimpliedaGPwasrequiredtomakearecommendationfor
furtherservices,andthereforethiswasnotselfreferral.Itwasthought
thatreferralbackintorehabilitationwouldoftenbeafteraconversation
withaGPornamedcontactbuttheGDGagreedthatitshouldbe
possibleforapersontoselfreferthemselves.Informationandsupportto
enablethepersonandtheirfamiliestoreadjustafterwhatisoftenalife
changingexperiencewasrecognisedasextremelyimportant.TheGDG
agreedwiththecommentssuggestingadditionalareassuchas
informationaboutvoluntaryorganisationsandpatientgroups,
recognisingtheneedsofcarersetc.Thevarietyofotherexamplesgiven
withinthecommentshighlightedtotheGDGthatitwasnotpossibleto
capturethevastrangeofdifferentinformationneedsinthewordingof
recommendations.Itwasagreedinformationandsupportisindividual
andprovidingalistofareastoinclude(ortheprofessionalwhowouldbe
responsibletodothis)wasnothelpful.Thestatementonreferralto
NationalClinicalGuidelineCentre,2013.
556
socialcarewasfelttobeunclear,andfollowingconsultationtheGDG
agreedtoremovethisasitwasfelttobecoveredwithinthesocialcare
interfaceandtransferofcaresectionsoftheguideline.
TheGDGdiscussedthestatementaboutregularlyreviewinghealthand
socialcareneedswhichdidnotachieveconsensus.Itwasfeltthatthere
wasdisagreementaboutthefrequencyofthereviewratherthanabout
whetherareviewshouldtakeplace.TheGDGagreedthatopinion
seemeddividedonadoptinganeedsbasedapproachorhavinga
structuredprocessdeterminingwhenareviewshouldbeundertaken.
Thegroupfeltitwasextremelyimportanttoemphasisethatreviews
needtobedone,inordertopickupanyproblemsordifficultiesandto
highlighttobothpatientsandtheircarersthatthisiswhattheyshould
expecttohappen.TheGDGnotedthattheNationalStrokeStrategy
recommendedareviewat6monthsandthenannually,andthatmanyof
thecommentshadstatedthesameapproachshouldbeadopted.
NationalClinicalGuidelineCentre,2013.
557
16 Acronymsandabbreviations
AAT
AachenerAphasiaTest
ABMT
AbbreviatedMentalTestScore
ADL
ActivitiesofDailyLiving
AFO
AnkleFootOrthosis
AGREEII
AdvancingGuidelineDevelopment,ReportingandEvaluationinHealthCare
ANELTA
AmsterdamNijmeganEverydayLanguageTest,scaleA
ANELTA
AmsterdamNijmeganEverydayLanguageTest
APT
AttentionProcessTraining
ARAT
ActionResearchArmTest
ASCOT
AngloScandinavianCardiacOutcomesTrial
ASHAFACS
FunctionalAssessmentofCommunicationSkillsforAdults
BADL
BasicActivitiesofDailyLiving
BBS
BergBalanceScale
BIT
BehaviouralInattentionTest
BNT
BostonNamingTest
CBSWRAT
CriterionTestofBasicSkillsWideRangeAchievementTest
CESD
CenterforEpidemiologicStudiesDepressionScale
CETI
CommunicationEffectivenessIndex
CETI
CommunicationEffectivenessIndex
CFQ
CognitiveFailuresQuestionnaire
CIMT
ConstraintInducedMovementTherapy
CMSA
ChedokeMcMasterStrokeAssessment
CMSMR
ChedokeMcMasterStagesofMotorRecovery
CNS
CentralNervousSystem
COAST
CommunicationOutcomesAfterStrokescale
CPT
ContinuousPerformanceTest
CRPS
ComplexRegionalPainSyndromes
DAFO
DynamicAnkleFootOrthosis
DWP
DepartmentofWorkandPensions
EMG
Electromyography
EQVAS
EuroQolVisualAnalogueScale
ESD
EarlySupportedDischarge
ESUS
ExtendedStrokeUnitService
FAI
FrenchayActivityIndex
FAST
FrenchayAphasiaScreeningTest
FCP
FunctionalCommunicationProgramme
FCTP
FunctionalCommunicationTherapyPlanner
FCTP
FunctionalCommunicationTherapyPlanner
FES
FunctionalElectricalStimulation
FIM
FunctionalIndependenceMeasure
FPA
FunctionalCommunicationProfile
GDS
GeriatricDepressionScale
NationalClinicalGuidelineCentre,2013.
558
GHQ
GeneralHealthQuestionnaire
GMW
GeneralMedicalWard
GRAFO
GroundReactionAnkleFootOrthosis
HADS
HospitalAnxietyandDepressionScale
IADL
InstrumentalActivitiesofDailyLiving
ICF
InternationalClassificationofFunctioning,DisabilityandHealth
ICF
InternationalClassificationofFunctioning,DisabilityandHealth
LLFMA
LowerlimbsectionoftheFuglMeyerAssessment
MAS
MotorassessmentScale
MID
MinimalImportantDifference
MMSE
MiniMentalStateExamination
MOCA
MontrealCognitiveAssessment
MPCA
MeasureofParticipationinConversationforAdultswithAphasia
MRC
MedicalResearchCouncil
MRW
MixedRehabilitationWard
MSCA
MeasureofSkillinProvidingSupportedConversationforAdultswithAphasia
MUST
MalnutritionUniversalScreeningTool
NAO
NationalAuditOffice
NIHSS
NationalInstitutesofHealthStrokeScale
OKN
OptokineticNystagmus
OSUS
OrdinaryStrokeUnitService
PASAT
PacedAuditorySerialAdditionTest
PEG
PercutaneousEndoscopicGastrostomy
PHQ
PatientHealthQuestionnaire
PICA
PorchIndexofCommunicativeAbility
PROM
PassiveRangeofMobility
PRT
ProgressiveResistanceTraining
PSSRU
PersonalSocialServicesResearchUnit
PTH
PlaceboThermocoagulation
RPAB
RivermeadPerceptualAssessmentBattery
SADQ
StrokeAphasicDepressionQuestionnaire
SIAS
StrokeImpairmentAssessmentSet
SSS
ScandinavianStrokeScale
TOM
TherapyOutcomeMeasure
UEFT
UpperExtremityFunctionTest
USN
UnilateralSpatialNeglect
WAB
WesternAphasiaBattery
WABAQ
WesternAphasiaBatteryAphasiaQuotient
WHO
WorldHealthOrganisation
WMFT
WolfMotorFunctionTest
YSQ
YaleSingleQuestion
NationalClinicalGuidelineCentre,2013.
559
17 Glossary
AachenerAphasieTest
(AAT)
CommonlyusedmajorcomprehensivelanguagetestinGermanspeaking
countries.
Abbreviatedmentaltest
score(ABMT)
Testtoassessforconfusionandothercognitiveimpairments.
Abstract
Summaryofastudy,whichmaybepublishedaloneorasanintroductiontoa
fullscientificpaper.
ActionResearchArmTest
(ARAT)
Observationaltestusedtodetermineupperlimbfunction.
ActivitiesofDailyLiving
(ADL)
Termusedinhealthcaretorefertodailyselfcareactivitieswithinan
individual'splaceofresidence,outdoorenvironments,orboth.
AddenbrooksCognitive
Examination
Examwhichincorporatesfivesubdomainscoresforcognition:
orientation/attention,memory,verbalfluency,languageandvisuospatial.
Adjustedanalysis
Usuallyreferstoattemptstocontrol(adjust)forbaselineimbalances
betweengroupsinimportantpatientcharacteristics.Sometimesusedto
refertoadjustmentsofPvaluetotakeaccountofmultipletesting.
Allocationconcealment
Theprocessusedtopreventadvanceknowledgeofgroupassignmentina
RCT.Theallocationprocessshouldbeimpervioustoanyinfluencebythe
individualmakingtheallocation,bybeingadministeredbysomeonewhois
notresponsibleforrecruitingparticipants.
Anosognosia
Alackofawarenessofimpairment,notknowingthatadeficitorillness
exists,inmemoryorotherfunction
Aphasia
Lossorimpairmentoftheabilitytouseandcomprehendlanguageusually
resultingfrombraindamage
Applicability
Thedegreetowhichtheresultsofanobservation,studyorreviewarelikely
toholdtrueinaparticularclinicalpracticesetting.
Apraxia(ofspeech)
Difficultyininitiatingandexecutingthevoluntarymovementneededto
producespeechwhenthereisnoweaknessofspeechmuscles.Itmaycause
difficultyproducingthecorrectspeechorchangesintherhythmorrateof
speaking.
Arm(ofaclinicalstudy)
Subsectionofindividualswithinastudywhoreceiveoneparticular
intervention,forexampleplaceboarm
Assessment
Adetailedprocesswhichaimstodefinethenatureandimpactofan
impairment,anddeviseatreatmentplan.
Association
Statisticalrelationshipbetweentwoormoreevents,characteristicsorother
variables.Therelationshipmayormaynotbecausal.
BarthelIndex
TheBarthelIndexconsistsof10itemsthatmeasureaperson'sdaily
functioning,specificallytheactivitiesofdailylivingandmobility.Theitems
includefeeding,movingfromwheelchairtobedandreturn,grooming,
transferringtoandfromatoilet,bathing,walkingonlevelsurface,goingup
anddownstairs,dressing,continenceofbowelsandbladder.
Theinitialsetofmeasurementsatthebeginningofastudy(afterrunin
periodwhereapplicable),withwhichsubsequentresultsarecompared.
Baseline
BasicactivitiesofDaily
Living(BADL)
Listofbasicactivitiesthatneedtobeperformedindependentlyinorderfor
anindividualtotakecareofhimself/herself.
BeckDepressionInventory
Amultiplechoiceselfreportinventory,usedformeasuringtheseverityof
depression.
Beforeandafterstudy
Astudythatinvestigatestheeffectsofaninterventionbymeasuring
particularcharacteristicsofapopulationbothbeforeandaftertakingthe
intervention,andassessinganychangethatoccurs.
NationalClinicalGuidelineCentre,2013.
560
Bergbalancescale
Awidelyusedclinicaltestofaperson'sfunctionalbalance.
Bias
Systematic(asopposedtorandom)deviationoftheresultsofastudyfrom
thetrueresultsthatiscausedbythewaythestudyisdesignedor
conducted.
Blinding
Keepingsomeorallstudyparticipants,caregivers,researchersoroutcome
assessorsunawareabouttheinterventionstowhichtheparticipantshave
beenallocatedinastudy.Seesingle,doubleandtripleblindingand
allocationconcealment.
BostonDiagnosticAphasia
Examination
Testusedtoevaluateadultssuspectedofhavingaphasia
BostonNamingTest(BNT)
Aconfrontationnamingtestusedtomeasurewordretrievalperformancein
aphasicpatients
BoxandBlocktest
Testusedtoevaluatethegrossmanualdexterityofindividualswitha
physicalimpairment.
Brunnstromapproach
Physicaltherapythatemphasisesthesynergicpatternofmovementwhich
developsduringrecoveryfromhemiplegia.Thisapproachencourages
developmentofflexorandextensorsynergiesduringearlyrecovery,withthe
intentionthatsynergicactivationofmuscleswill,withtraining,transitioninto
voluntaryactivationofmovements.
C&Ecancellation
Testusedtodetectthepresenceofunilateralspatialneglectinthenear
extrapersonalspaceinpatientswithstroke.
Caregiverburdenscale
Questionnaireusedtomeasurethesubjectiveburdenofcaregiversinfive
domains:generalstrain,isolation,disappointment,emotionalinvolvement,
andenvironment.
Carer(caregiver)
Someoneotherthanahealthprofessionalwhoisinvolvedincaringfora
personwithamedicalcondition.
Carersstrainindexrankin
Isabrief,easilyadministeredinstrumentwhichcanidentifystrainininformal
careproviders.Itisdividedintofivecategories.Itisa13questiontoolthat
measuresstrainrelatedtocareprovision.
Cerebrovasculardisease
Diseaseofthebloodvesselssupplyingthebrain,whichmayresultinbrain
dysfunctionincludingStroke.
ChedokeMcMasterstages
ofMotionRecovery
Testusedtoassessthefunctionalstateoftheaffectedupperextremity.
Clinicaleffectiveness
Theextenttowhichaninterventionproducesanoverallhealthbenefitin
routineclinicalpractice.
Clustertrial
Atypeofrandomizedcontrolledtrial(RCT),inwhichgroupsorclustersof
individualsratherthanindividualsthemselvesarerandomized.
CochraneReview
TheCochraneLibraryconsistsofaregularlyupdatedcollectionofevidence
basedmedicinedatabasesincludingtheCochraneDatabaseofSystematic
Reviews(reviewsofrandomisedcontrolledtrialspreparedbytheCochrane
Collaboration).
Cohortstudy
Aretrospectiveorprospectivefollowupstudy.Groupsofindividualstobe
followeduparedefinedonthebasisofpresenceorabsenceofexposuretoa
suspectedriskfactororintervention.Acohortstudycanbecomparative,in
whichcasetwoormoregroupsareselectedonthebasisofdifferencesin
theirexposuretotheinterventionofinterest.
Comorbidity
Coexistenceofmorethanonediseaseoranadditionaldisease(otherthan
thatbeingstudiedortreated)inanindividual.
Comparability
Similarityofthegroupsincharacteristicslikelytoaffectthestudyresults
(suchashealthstatusorage).
Confidenceinterval(CI)
Arangeofplausiblevaluesforthepopulationmean(oranotherpopulation
parametersuchasanestimationofriskincrease/decrease),calculatedfrom
NationalClinicalGuidelineCentre,2013.
561
data.Aconfidenceintervalwith(conventionally)a95%confidencelevelhas
a95%chanceofcapturingthepopulationmean.Thismeansthat,ifthe
experimentwererepeatedmanytimes,95%oftheconfidenceintervals
wouldcontainthetruepopulationmean.
Confounding
Inastudy,confoundingoccurswhentheeffectofaninterventiononan
outcomeisdistortedasaresultofanassociationbetweenthepopulationor
interventionoroutcomeandanotherfactor(theconfoundingvariable)that
caninfluencetheoutcomeindependentlyoftheinterventionunderstudy.
Consensusmethods
Techniquesthataimtoreachanagreementonaparticularissue.Consensus
methodsmaybeusedwhenthereisalackofstrongevidenceonaparticular
topic.
Continuousdata
Datawithapotentiallyinfinitenumberofpossiblevaluesalongacontinuum.
Height,weightandbloodpressureareexamplesofcontinuousvariables.
Contralateral
Onorrelatingtotheoppositesideofthebody.
Contralesional
Describingthehalfofapatient'sbrainorbodyawayfromthesiteofalesion.
Controlgroup
Agroupofpatientsrecruitedintoastudythatreceivesnotreatment,a
treatmentofknowneffect,oraplacebo(dummytreatment)inorderto
provideacomparisonforagroupreceivinganexperimentaltreatment,such
asanewdrug.
Costbenefitanalysis
Atypeofeconomicevaluationwherebothcostsandbenefitsofhealthcare
treatmentaremeasuredinthesamemonetaryunits.Ifbenefitsexceed
costs,theevaluationwouldrecommendprovidingthetreatment.
Costconsequencesanalysis
(CCA)
Atypeofeconomicevaluationwherevarioushealthoutcomesarereported
inadditiontocostforeachintervention,butthereisnooverallmeasureof
healthgain.
Costeffectivenessanalysis
(CEA)
Aneconomicstudydesigninwhichconsequencesofdifferentinterventions
aremeasuredusingasingleoutcome,usuallyinnaturalunits(Forexample,
lifeyearsgained,deathsavoided,heartattacksavoided,casesdetected).
Alternativeinterventionsarethencomparedintermsofcostperunitof
effectiveness.
Costeffectivenessmodel
Anexplicitmathematicalframework,whichisusedtorepresentclinical
decisionproblemsandincorporateevidencefromavarietyofsourcesin
ordertoestimatethecostsandhealthoutcomes.
Costutilityanalysis(CUA)
Aformofcosteffectivenessanalysisinwhichtheunitsofeffectivenessare
qualityadjustedlifeyears(QALYs).
CredibleInterval
TheBayesianequivalentofaconfidenceinterval.
Crossovertrial
Atypeofclinicaltrialcomparingtwoormoreinterventionsinwhichthe
participants,uponcompletionofthecourseofonetreatmentareswitched
toanother.Forexample,foracomparisonoftreatmentsAandB,halfthe
participantsarerandomlyallocatedtoreceivethemintheorder,A,Band
halftoreceivethemintheorderB,A.Aproblemwiththisdesignisthatthe
effectsofthefirsttreatmentmaycarryoverintotheperiodwhenthesecond
isgiven.
DerSimonianandLaird
Amethodofrandomeffectsmetaanalysis(seebelow).
Diplopia
Doublevision.
Discounting
Costsandperhapsbenefitsincurredtodayhaveahighervaluethancostsand
benefitsoccurringinthefuture.Discountinghealthbenefitsreflects
individualpreferenceforbenefitstobeexperiencedinthepresentrather
thanthefuture.Discountingcostsreflectsindividualpreferenceforcoststo
beexperiencedinthefutureratherthanthepresent.
Dominance
Aninterventionissaidtobedominatedifthereisanalternativeintervention
thatisbothlesscostlyandmoreeffective.
NationalClinicalGuidelineCentre,2013.
562
EuroQolVisualAnalogue
Scale(EQVAS)
Qualityoflifemeasure.
Dorsiflexion
Movementwhichdecreasestheanglebetweenthedorsum(superior
surface)ofthefootandtheleg,sothatthetoesarebroughtclosertothe
shin.
Doubleblind
Also,Doublemasked.Neithertheparticipantsinatrialnortheinvestigators
(outcomeassessors)areawareofwhichinterventiontheparticipantsare
given.Thepurposeofblindingtheparticipants(recipientsandprovidersof
care)istopreventperformancebias.Thepurposeofblindingthe
investigators(outcomeassessors,whomightalsobethecareproviders)isto
protectagainstdetectionbias.Seealsoblinding,singleblind,tripleblind,
andallocationconcealment.
Dropout
Aparticipantwhowithdrawsfromatrialbeforetheend.
Dysarthria
Difficultyinarticulatingwords.
Dysarthrophonia
Anacquiredneurologicalspeechimpairmentthataffectsrespiration,
productionofspeechsounds,articulationandintonationofspeech.
Dysphagia
Difficultyinswallowing.
Dyspraxia
Difficultyinplanningandexecutingmovement
Earlysupporteddischarge
Aserviceforpeopleafterstrokewhichallowstransferofcarefroman
inpatientenvironmenttoaprimarycaresettingtocontinuerehabilitation,at
thesamelevelofintensityandexpertisethattheywouldhavereceivedin
theinpatientsetting.
Economicevaluation
Comparativeanalysisofalternativehealthstrategies(interventionsor
programmes)intermsofboththeircostsandconsequences.
EdmansActivitiesofDaily
LivingIndex
Agradedtestwhichassessesfunctionalabilitiesinstrokepatients,including
theactivitiesnecessarytoenableapersontoliveindependentlyathome.
Effectsize
1.Agenerictermfortheestimateofeffectforastudy.
2.Adimensionlessmeasureofeffectthatistypicallyusedforcontinuous
datawhendifferentscales(forexampleformeasuringpain)areusedto
measureanoutcomeandisusuallydefinedasthedifferenceinmeans
betweentheinterventionandcontrolgroupsdividedbythestandard
deviationofthecontrolorbothgroups.Seestandardisedmeandifference.)
Effectiveness
SeeClinicaleffectiveness.
Efficacy
SeeClinicalefficacy.
EQ5D(EuroQol5D)
Astandardisedinstrumentusedtomeasureahealthoutcome.Itprovidesa
singleindexvalueforhealthstatus.
Equinovarus
Adevelopmentaldisorderofthefootinwhichwalkingisdoneonthetoes
andoutersideofthesole
Errorlesslearning
Procedurewhichallowsdiscriminationlearningtooccurwithfeworno
responsestothenegativestimulus.
Evidence
Informationonwhichadecisionorguidanceisbased.Evidenceisobtained
fromarangeofsourcesincludingrandomisedcontrolledtrials,observational
studies,expertopinion(ofclinicalprofessionalsand/orpatients).
Exclusioncriteria(clinical
study)
Criteriathatdefinewhoisnoteligibletoparticipateinaclinicalstudy.
Exclusioncriteria(literature
review)
Explicitstandardsusedtodecidewhichstudiesshouldbeexcludedfrom
considerationaspotentialsourcesofevidence.
Extendeddominance
IfOptionAisbothmoreclinicallyeffectivethanOptionBandhasalower
costperunitofeffect,whenbotharecomparedwithadonothing
alternativethenOptionAissaidtohaveextendeddominanceoverOptionB.
NationalClinicalGuidelineCentre,2013.
563
OptionAisthereforemoreefficientandshouldbepreferred,otherthings
remainingequal.
Extrapolation
Indataanalysis,predictingthevalueofaparameteroutsidetherangeof
observedvalues.
Followup
Fixedeffectsmetaanalysis
Theascertainmentofoutcomesofaninterventionatoneormorestated
timesaftertheinterventionhasended.
Afixedeffectmodelofmetaanalysisisbasedonamathematicalassumption
thateverystudyisevaluatingacommontreatmenteffect.Thatmeansthe
effectoftreatment,allowingfortheplayofchance,wasthesameinall
studies.
FrenchayActivitiesIndex
(FAI)
Measureofinstrumentalactivitiesofdailylivingforusewithpatients
recoveringfromstroke.Thisindexcoversabroadrangeofactivities
associatedwitheverydaylife.
FrenchayArmTest
Testusedtoassessproximalcontrolanddexterity.
FuglMeyerAssessment
Strokespecific,performancebased,impairmentindex;designedtoassess
motorfunctioning,balance,sensationandjointfunctioninginhemiplegic,
poststrokepatients.
Functionalambulation
category
Afunctionalwalkingtestthatevaluatesambulationability.This6pointscale
assessesambulationstatusbydetermininghowmuchhumansupportthe
patientrequireswhenwalking,regardlessofwhetherornottheyusea
personalassistivedevice
Functionalambulation
classification
Assessesfunctionalmobilityandgaitinpatientsundergoingphysicaltherapy.
FunctionalAssessmentof
CommunicationSkillsfor
Adults(ASHAFACS),
Thisassessmentassistswithmeasuringandrecordingthefunctional
communicationofadultswithspeech,language,andcognitive
communicationdisorders.Itassessesfunctionalcommunicationinfour
areas:socialcommunication;communicationofbasicneeds;reading,
writing,andnumberconcepts;anddailyplanning.
FunctionalIndependence
Measure(FIM)
Scaleusedtomeasurethefunctionalabilitiesofpatientsundergoing
rehabilitation.
Generalisability
Theextenttowhichtheresultsofastudybasedonmeasurementina
particularpatientpopulationand/oraspecificcontextholdtrueforanother
populationand/orinadifferentcontext.Inthisinstance,thisisthedegreeto
whichtheguidelinerecommendationisapplicableacrossbothgeographical
andcontextualsettings.Forinstance,guidelinesthatsuggestsubstituting
oneformoflabourforanothershouldacknowledgethatthesecostsmight
varyacrossthecountry.
Geriatricdepressionscale
Thisissuitableasascreeningtestfordepressivesymptomsintheelderly;
idealforevaluatingtheclinicalseverityofdepression.
GlobalNottinghamHealth
Profile1&2
Patientcompletedtwopartquestionnairedesignedtodetermineand
quantifyperceivedhealthproblems.Partonecovers6areas(sleep,mobility,
energy,pain,emotionalreactions,socialisolation);andparttwocovers
specificaspectsofdailylife(employment,householdchores,sociallife,
relationships,sexlife,hobbies,holidays).
GRADE/GRADEprofile
AsystemdevelopedbytheGRADEWorkingGrouptoaddressthe
shortcomingsofpresentgradingsystemsinhealthcare.TheGRADEsystem
usesacommon,sensibleandtransparentapproachtogradingthequalityof
evidence.TheresultsofapplyingtheGRADEsystemtoclinicaltrialdataare
displayedinatableknownasaGRADEprofile.
HabitualGaitVelocity
Alsoknownascomfortablegaitspeed,itisdefinedasapersonsusualor
comfortable,selfselectedpace.
Handgripforce
Thestrengthappliedbythehandtopullonorsuspendfromobjectsandisa
specificpartofhandstrength.
NationalClinicalGuidelineCentre,2013.
564
Harms
Adverseeffectsofanintervention.
Healtheconomics
Thestudyoftheallocationofscarceresourcesamongalternativehealthcare
treatments.Healtheconomistsareconcernedwithbothincreasingthe
averagelevelofhealthinthepopulationandimprovingthedistributionof
health.
Healthrelatedqualityoflife
(HRQoL)
Acombinationofanindividualsphysical,mentalandsocialwellbeing;not
merelytheabsenceofdisease.
Hemiagnosia
Inabilityofapersontoprocessandperceivestimuliononesideofthebody
orenvironmentthatisnotduetoalackofsensationadeficitinattention
toandawarenessofonesideofspaceisobserved.
Hemianopia
Blindnessinonehalfofthevisualfieldofoneorbotheyes.
Hemineglect
SeeHemiagnosia
Hemiparesis
Weaknessononesideofthebody.
Hemiparetic
Pertainingtohemiparesisorapatientaffectedwithhemiparesis.
hemiplegia
Totalparalysisofthearm,leg,andtrunkononesideofthebody.
Hemispatialneglect
SeeHemiagnosia
Heterogeneity Orlackof
homogeneity.
Thetermisusedinmetaanalysesandsystematicreviewswhentheresults
orestimatesofeffectsoftreatmentfromseparatestudiesseemtobevery
differentintermsofthesizeoftreatmenteffectsoreventotheextentthat
someindicatebeneficialandotherssuggestadversetreatmenteffects.Such
resultsmayoccurasaresultofdifferencesbetweenstudiesintermsofthe
patientpopulations,outcomemeasures,definitionofvariablesordurationof
followup.
Imprecision
Resultsareimprecisewhenstudiesincluderelativelyfewpatientsandfew
eventsandthushavewideconfidenceintervalsaroundtheestimateof
effect.
Inclusioncriteria(literature
review)
Explicitcriteriausedtodecidewhichstudiesshouldbeconsideredas
potentialsourcesofevidence.
Incrementalanalysis
Theanalysisofadditionalcostsandadditionalclinicaloutcomeswith
differentinterventions.
Incrementalcost
Themeancostperpatientassociatedwithaninterventionminusthemean
costperpatientassociatedwithacomparatorintervention.
Incrementalcost
effectivenessratio(ICER)
Thedifferenceinthemeancostsinthepopulationofinterestdividedbythe
differencesinthemeanoutcomesinthepopulationofinterestforone
treatmentcomparedwithanother.
Indirectness
Theavailableevidenceisdifferenttothereviewquestionbeingaddressed,in
termsofPICO(population,intervention,comparisonandoutcome).
InstrumentalActivitiesof
DailyLiving(IADL)
Listofactivitieswhichallowsanindividualtoliveindependentlyina
community:housework,mealpreparation,takingmedications,managing
money,shoppingforgroceriesorclothing,telephoneuseandifapplicable,
theuseoftechnology.
Intentiontotreatanalysis
(ITT)
Anintentiontotreatanalysisisoneinwhichalltheparticipantsinatrialare
analysedaccordingtotheinterventiontowhichtheywereallocated,
whethertheyreceiveditornot.Intentiontotreatanalysesarefavouredin
assessmentsofeffectivenessastheymirrorthenoncomplianceand
treatmentchangesthatarelikelytooccurwhentheinterventionisusedin
practiceandbecauseoftheriskofattritionbiaswhenparticipantsare
excludedfromtheanalysis.
IntermediateOutcome
Ameasureofresultsthatindicatesprogresstowarddesiredendresultsbutis
notitselfafinaloutcome.
Interphalangeal
Betweenthephalanges.
NationalClinicalGuidelineCentre,2013.
565
Intervention
Healthcareactionintendedtobenefitthepatient,forexample,drug
treatment,surgicalprocedure,psychologicaltherapy.
Inversevariance
Amethodofaggregatingtwoormorerandomvariablestominimizethe
varianceofthesum.Eachrandomvariableinthesumisweightedininverse
proportiontoitsvariance.Wecouldassumethatvarianceisinversely
proportionaltoimportance,i.e.thelessvarianceinthestudy,themore
weightitshouldcontribute.TheInverseVariancemethodinRevMan(see
below),calculatesstudyweightsdirectlybasedonthisassumption.
Ipsilateral
Onorrelatingtothesameside(ofthebody)
Isometricelbowextension
force
Theforceusedwhenattemptingtoextendtheelbowagainstresistance.
Isometricimpliesnoactualmovementismade.
Isometricelbowflexion
force
Theforceusedwhenattemptingtobendthearmattheelbowagainst
resistance.Isometricimpliesnoactualmovementismade.
IVAContinuous
PerformanceTestFull
ScaleAttentionQuotient
Testusedtomeasureauditoryandvisualreactiontimeandstability,
simultaneously,notseparately.
JebsenTaylorHandfunction Testdesignedtoprovideashort,objectivetestofhandfunctionscommonly
test
usedinactivitiesofdailyliving(ADLs).Thetargetpatientpopulationincludes
adultswithneurologicalconditionsinvolvinghanddisabilities.Thetestwas
developedtobeusedbyhealthprofessionalsworkinginrestorationofhand
function.
Kneeextensionpeaktorque
Measurementofthepersonsabilitytoflexthequadricepsmuscleswhich
straightentheleg.
Kneeflexionpeaktorque
Sitandreachtestusedtotestflexibility.
Lengthofstay
Thetotalnumberofdaysaparticipantstaysinhospital.
Lifeyearsgained
Meanaverageyearsoflifegainedperpersonasaresultoftheintervention
comparedwithanalternativeintervention.
Likelihoodratio
Thelikelihoodratiocombinesinformationaboutthesensitivityand
specificity.Ittellsyouhowmuchapositiveornegativeresultchangesthe
likelihoodthatapatientwouldhavethedisease.Thelikelihoodratioofa
positivetestresult(LR+)issensitivitydividedby1specificity.
LineBisection
Quickmeasuretodetectthepresenceofunilateralspatialneglect.
Linebisectiontask
Standardassessmentofunilateralvisualneglect.
LocusofControlScale
Referstotheextenttowhichindividualsbelievethattheycancontrolevents
thataffectsthem.
Longtermcare
Residentialcareinahomethatmayincludeskillednursingcareandhelp
witheverydayactivities.Thisincludesnursinghomesandresidentialhomes.
Losstofollowup
Lossofcontactwithsomeparticipants,sothatresearcherscannotcomplete
datacollectionasplanned.Losstofollowupisacommoncauseofmissing
data,especiallyinlongtermstudiesandcancausebiaswhensubjectslost
fromacohorthavedifferenthealthresponsedistributionsfromsubjectswho
remaininfollowup
MantelHaenszelapproach
Markovmodel
Amethodtoanalyseoddsratiosthathasbeenextendedtoanalyserisk
ratiosandriskdifferences.Itassumesafixedeffectandcombinesstudies
usingamethodsimilartoinversevarianceapproachestodeterminethe
weightgiventoeachstudy.
Amethodforestimatinglongtermcostsandeffectsforrecurrentorchronic
conditions,basedonhealthstatesandtheprobabilityoftransitionbetween
themwithinagiventimeperiod(cycle).
MaximalGaitVelocity
Alsoknownasfastgaitspeed,itisdefinedasthespeedapersonsfastas
safelypossible,selfselectedpace.
McKennaGradedNaming
Testusedtoassessobjectnamingability,butisinadditiongradedin
NationalClinicalGuidelineCentre,2013.
566
Test
difficultytoallowforindividualdifferences.Thismeansthatitmaybeableto
detectanywordfindingdifficultyeveninthosewithanextensivenaming
vocabulary.
Measurement
theuseofpsychometricallyrobusttoolstorecordtheextentofaproblem,
whetheritisimpairment,activityorparticipationbasedandcanbegeneric
ordiseasespecific.
Mean
Theaveragevalue,calculatedbyaddingalltheobservationsanddividingby
thenumberofobservations.
Median
Thenumericalvalueseparatingthehigherhalfofasample,apopulation,ora
probabilitydistribution,fromthelowerhalf.Themedianofafinitelistof
numberscanbefoundbyarrangingalltheobservationsfromlowestvalueto
highestvalueandpickingthemiddleone.Ifthereisanevennumberof
observations,thenthereisnosinglemiddlevalue;themedianisthenusually
definedtobethemeanofthetwomiddlevalues.
MedicalResearchcouncil
Scale(MRCScale)for
Musclestrength
Scaleforassessingmuscleweakness/strength.
MesulamVerbal
CancellationTest
TestusedtoevaluatehemispatialdominanceinStrokePatients.
Metaanalysis
Astatisticaltechniqueforcombining(pooling)theresultsofanumberof
studiesthataddressthesamequestionandreportonthesameoutcomesto
produceasummaryresult.Theaimistoderivemorepreciseandclear
informationfromalargedatapool.Itisgenerallymorereliablylikelyto
confirmorrefuteahypothesisthantheindividualtrials.
MinimalImportant
Differences(MID)
Forcontinuousoutcomes,theMIDisdefinedasthesmallestdifferencein
scoreintheoutcomeofinterestthatinformedpatientsorinformedproxies
perceiveasimportant,eitherbeneficialorharmful,andthatwouldleadthe
patientorcliniciantoconsiderachangeinthemanagement(refs).Aneffect
estimatelargerthantheMIDisconsideredtobeclinicallyimportant.For
dichotomousoutcomestheMIDisthesmallestdecreaseorincreaseisthe
incidenceofanoutcomethatwouldbeconsideredtoshowaclear
appreciablebenefitorharmfromaninterventionthiscanbeconsideredin
relativeterms(usingtheriskratio)butpreferablyshouldbebasedon
absoluteriskdifferences.
Mnemonicstrategies
Systematicstrategiesforstrengtheninglongtermretentionandretrievalof
information.
ModifiedAshworthScale
Thisscalemeasuresresistanceduringpassivesofttissuestretching.
Modifiedrankin
Acommonlyusedscaleformeasuringthedegreeofdisabilityordependence
inthedailyactivitiesofpeoplewhohavesufferedastroke.
MotricityIndex
Staffcompletedindexoflimbmovementaimingtomeasuregeneralmotor
impairment.Threemovementsforeachlimbareassessedbasedonthe
MedicalResearchCouncilstrengthgradesandweighted,zeroforno
movement,nineforpalpablemovement,fourteenformovementseen,
nineteenforfullrangeagainstgravity,twentyfiveformovementagainst
resistanceandtwentytwofornormalpower.
Neglect
Inabilitytoorienttowardsandattendtostimuli,includingbodyparts,onthe
sideofthebodyaffectedbythestroke.
Neuroplasticity
Structuralandfunctionalchangestothebrainandnervoussystemasaresult
ofinputfromtheenvironment.
Neuropsychological
Relatedtothestructureandfunctionofthebrainspecifictopsychological
processesandbehaviours.
Nonparetic
Usuallyreferstotheunaffectedlimb
NationalClinicalGuidelineCentre,2013.
567
NottinghamExtended
ActivitiesofDailyLiving
Selfreportscaledesignedprimarilyforuseinthestrokepopulationfor
functionalassessment.
Observationalstudy
Retrospectiveorprospectivestudyinwhichtheinvestigatorobservesthe
naturalcourseofeventswithorwithoutcontrolgroups;forexample,cohort
studiesandcasecontrolstudies.
Opportunitycost
Thelossofotherhealthcareprogrammesdisplacedbyinvestmentinor
introductionofanotherintervention.Thismaybebestmeasuredbythe
healthbenefitsthatcouldhavebeenachievedhadthemoneybeenspenton
thenextbestalternativehealthcareintervention.
Orthosis
Adevicethatsupportsorcorrectsthefunctionofalimborthetorso.
Orthotics
Specialtywithinthemedicalfieldconcernedwiththedesign,manufacture
andapplicationoforthoses.Anorthosis(plural:orthoses)isanorthopaedic
devicethatsupportsorcorrectsthefunctionofalimborthetorso.
Outcome
Measureofthepossibleresultsthatmaystemfromexposuretoapreventive
ortherapeuticintervention.Outcomemeasuresmaybeintermediate
endpointsortheycanbefinalendpoints.SeeIntermediateoutcome.
Palmargriptorque
Palmargriptorqueisameasurementoftheabilitytoholdlargerandheavier
objectssuchascansandbottlesbetweenthepalmofthehandandthefour
fingers.
Perimetry
Awaytosystematicallytest,usedtomapandquantifythevisualfield,
especiallyattheextremeperipheryofthevisualfield.Thenamecomesfrom
themethodoftestingtheperimeterofthevisualfield.
Pinchgripforce
Thestrengthappliedbythehandtopinchanobjectsothatthefingersareon
onesideoftheobject,andthethumbisontheother.Typically,anobject
liftedinapinchgripdoesnottouchthepalm.
Placebo
Aninactiveandphysicallyidenticalmedicationorprocedureusedasa
comparatorincontrolledclinicaltrials.
Power(statistical)
Theabilitytodemonstrateanassociationwhenoneexists.Powerisrelated
tosamplesize;thelargerthesamplesize,thegreaterthepowerandthe
lowertheriskthatapossibleassociationcouldbemissed.
Pretestprobability
Fordiagnostictests.Theproportionofpeoplewiththetargetdisorderinthe
populationatriskataspecifictimepointortimeinterval.Prevalencemay
dependonhowadisorderisdiagnosed.
Primarycare
Healthcaredeliveredtopatientsoutsidehospitals.Primarycarecoversa
rangeofservicesprovidedbygeneralpractitioners,nurses,dentists,
pharmacists,opticiansandotherhealthcareprofessionals.
Primaryoutcome
Theoutcomeofgreatestimportance,usuallytheoneinastudythatthe
powercalculationisbasedon.
Prismglasses
Medicaldeviceusedincorrectingeyeabnormalities.
Prognosis
Aprobablecourseoroutcomeofadisease.Prognosticfactorsarepatientor
diseasecharacteristicsthatinfluencethecourse.Goodprognosisis
associatedwithlowrateofundesirableoutcomes;poorprognosisis
associatedwithahighrateofundesirableoutcomes.
Proprioceptive
Individualssenseoftherelativepositionofneighbouringpartsofthebody.
Prospectivestudy
Astudyinwhichpeopleareenteredintotheresearchandthenfollowedup
overaperiodoftimewithfutureeventsrecordedastheyhappen.This
contrastswithstudiesthatareretrospective.
Psychometric
Relatedtothetheoryandtechniqueofeducationalmeasurementand
psychologicalmeasurement,whichincludesthemeasurementofknowledge,
abilities,attitudes,andpersonalitytraits.
Publicationbias
Alsoknownasreportingbias.Abiascausedbyonlyasubsetofallthe
relevantdatabeingavailable.Thepublicationofresearchcandependonthe
NationalClinicalGuidelineCentre,2013.
568
natureanddirectionofthestudyresults.Studiesinwhichaninterventionis
notfoundtobeeffectivearesometimesnotpublished.Becauseofthis,
systematicreviewsthatfailtoincludeunpublishedstudiesmayoverestimate
thetrueeffectofanintervention.Inaddition,apublishedreportmight
presentabiasedsetofresults(forexampleonlyoutcomesorsubgroups
whereastatisticallysignificantdifferencewasfound.
Pvalue
Theprobabilitythatanobserveddifferencecouldhaveoccurredbychance,
assumingthatthereisinfactnounderlyingdifferencebetweenthemeansof
theobservations.Iftheprobabilityislessthan1in20,thePvalueislessthan
0.05;aresultwithaPvalueoflessthan0.05isconventionallyconsideredto
bestatisticallysignificant.
Quadrantanopia
Referstolossofvisionaffectingaquarterofthefieldofvision.Itcanbe
associatedwithalesionofanopticradiation.
Qualityoflife
SeeHealthrelatedqualityoflife.
Qualityadjustedlifeyear
(QALY)
Anindexofsurvivalthatisadjustedtoaccountforthepatientsqualityoflife
duringthistime.QALYshavetheadvantageofincorporatingchangesinboth
quantity(longevity/mortality)andquality(morbidity,psychological,
functional,socialandotherfactors)oflife.Usedtomeasurebenefitsincost
utilityanalysis.TheQALYsgainedarethemeanQALYsassociatedwithone
treatmentminusthemeanQALYsassociatedwithanalternativetreatment.
Quasirandomisedtrial
Atrialusingaquasirandommethodofallocatingparticipantstodifferent
formsofcare.Thereisagreaterriskofselectionbiasinquasirandomtrials
whereallocationisnotadequatelyconcealedcomparedwithrandomised
controlledtrialswithadequateallocationconcealment.
QuickReferenceGuide
Randomeffectsmeta
analysis
AnabridgedversionofNICEguidance,whichpresentsthekeyprioritiesfor
implementationandsummarisestherecommendationsforthecoreclinical
audience.
Therandomeffectsmodelassumesthatthetruetreatmenteffectsinthe
individualstudiesmaybedifferentfromeachother.Thatmeansthereisno
singlenumbertoestimateinthemetaanalysis,butadistributionof
numbers.Themostcommonrandomeffectsmodelalsoassumesthatthese
differenttrueeffectsarenormallydistributed.Themetaanalysistherefore
estimatesthemeanandstandarddeviationofthedifferenteffects.
Randomisation
Allocationofparticipantsinaresearchstudytotwoormorealternative
groupsusingachanceprocedure,suchascomputergeneratedrandom
numbers.Thisapproachisusedinanattempttoensurethereisaneven
distributionofparticipantswithdifferentcharacteristicsbetweengroupsand
thusreducesourcesofbias.
Randomisedcontrolledtrial
(RCT)
Acomparativestudyinwhichparticipantsarerandomlyallocatedto
interventionandcontrolgroupsandfolloweduptoexaminedifferencesin
outcomesbetweenthegroups.
RCT
SeeRandomisedcontrolledtrial.
Relativerisk(RR)
Thenumberoftimesmorelikelyorlesslikelyaneventistohappeninone
groupcomparedwithanother(calculatedastheriskoftheeventingroup
A/theriskoftheeventingroupB).
Reportingbias
Seepublicationbias.
Resourceimplication
Thelikelyimpactintermsoffinance,workforceorotherNHSresources.
Retrospectivestudy
Aretrospectivestudydealswiththepresent/pastanddoesnotinvolve
studyingfutureevents.Thiscontrastswithstudiesthatareprospective.
Reviewquestion
Inguidelinedevelopment,thistermreferstothequestionsabouttreatment
andcarethatareformulatedtoguidethedevelopmentofevidencebased
recommendations.
ReyOsterreithTest
Neuropsychologicalassessmentinwhichexamineesareaskedtoreproducea
NationalClinicalGuidelineCentre,2013.
569
complicatedlinedrawing,firstbycopyingandthenfrommemory.
RivermeadADL
Scaledevelopedtoassessactivitiesofdailylivinginstrokepatients.
RivermeadMobilityIndex
Measureofdisabilityrelatedtobodilymobility.Itdemonstratesthepatient's
abilitytomoveherorhisownbody.
RivermeadPerceptual
AssessmentBattery(RPAB)
Preliminaryassessmentofone'slevelofvisualperceptualabilitypriorto
therapy.Theresultsmaybeusedtoplananappropriatetherapyprogramme.
Screening
Aprocessofidentifyingpeoplewithparticularimpairments.Peoplecanthen
beofferedinformation,furtherassessmentandappropriatetreatment.
Screeningmaybeperformedasaprecursortomoredetailedassessment.
Searchstrategy
Themethodsusedtoidentifystudiesincludinghandsearchingrelevant
journals,searchingelectronicdatabases,contactingdrugcompanies,other
formsofpersonalcontactandcheckingreferencelists.
Secondaryoutcome
Anoutcomeusedtoevaluateadditionaleffectsoftheinterventiondeemeda
prioriasbeinglessimportantthantheprimaryoutcomes.
Selectionbias
Asystematicbiasinselectingparticipantsforstudygroups,sothatthe
groupshavedifferencesinprognosisand/ortherapeuticsensitivitiesat
baseline.Randomisation(withconcealedallocation)ofpatientsprotects
againstthisbias.
ShortForm36(SF36)
Multipurpose,shortformhealthsurveywiththirtysixquestions.Ityieldsan
eightscaleprofileoffunctionalhealthandwellbeingscoresaswellas
psychometricallybasedphysicalandmentalhealthsummarymeasuresanda
preferencebasedhealthutilityindex.
Significance(clinical)
Inmedicineandpsychology,thisreferstoeitheroftworelatedbutslightly
dissimilarconceptswherebycertainfindingsordifferences,evenif
measurableorstatisticallyconfirmed,eithermayormaynothaveadditional
significance,eitherby(1)beingofamagnitudethatconveyspractical
relevance(ausagethatconflatespracticalandclinicalsignificance
interchangeably),or(2)moretechnicallyandrestrictively,addresseswhether
aninterventionortreatmentmayormaynotfullycorrectthefinding.
Significance(statistical)
Aresultisdeemedstatisticallysignificantiftheprobabilityoftheresult
occurringbychanceislessthan1in20(p<0.05).
Singleblind
Also,singlemasked.Theinvestigatorisawareofthetreatment/intervention
theparticipantisgetting,buttheparticipantisunaware.Seealsoblinding,
doubleblind,tripleblind.
Spasticity
Muscularhypertonicitywithincreasedtendonreflexes
Spatialneglect
SeeHemiagnosia
Stakeholder
Thosewithaninterestintheuseoftheguideline.Stakeholdersinclude
manufacturers,sponsors,healthcareprofessionals,andpatientandcarer
groups.
Standardisedmean
difference
Thedifferencebetweentwomeansdividedbyanestimateofthewithin
groupstandarddeviation.Whenanoutcome(suchaspain)ismeasuredina
varietyofwaysacrossstudies(usingdifferentscales)itmaynotbepossible
directlytocompareorcombinestudyresultsinasystematicreview.By
expressingtheeffectsasastandardisedvaluetheresultscanbecombined
sincetheyhavenounits.Standardisedmeandifferencesaresometimes
referredtoasadindex.
Statisticalpower
Theprobabilitythatthenullhypothesiswillberejectedifitisindeedfalse.In
studiesoftheeffectivenessofhealthcareinterventions,powerisameasure
ofthecertaintyofavoidingafalsenegativeconclusionthataninterventionis
noteffectivewhenintruthitiseffective.Thepowerofastudyisdetermined
byhowlargeitis(thenumberofparticipants),thenumberofevents(for
examplestrokes)orthedegreeofvariationinacontinuousoutcome(suchas
weight),howsmallaneffectonebelievesisimportant(i.e.thesmallest
NationalClinicalGuidelineCentre,2013.
570
differenceinoutcomesbetweentheinterventionandthecontrolgroupsthat
isconsideredtobeimportant),andhowcertainonewantstobeofavoiding
afalsepositiveconclusion(i.e.thecutoffthatisusedforstatistical
significance).
Strengthandfinger
extension
Measurementofthestrengthusedtoopenthehand,stretchingallthe
fingers.
StrokeSelfEfficacy
Questionnaire
Questionnaireusedtomeasureselfefficacyjudgementsinspecificdomains
offunctioningrelevanttoindividualsfollowingstroke.
Strokeimpactscale
Strokespecific,selfreport,healthstatusmeasure,designedtoassess
multidimensionalstrokeoutcomes,includingstrength,handfunction,
activitiesofdailyliving/instrumentalactivitiesofdailyliving,mobility,
communication,emotion,memoryandthinking,andparticipation.
Anenvironmentinwhichmultidisciplinarystroketeamsdeliverstrokecarein
StrokeUnit
adedicatedwardwhichhasabedarea,diningarea,gym,andaccessto
assessmentkitchens.
StrokeRehabilitationService Astrokeservicedesignedtodeliverstrokerehabilitationeitherinhospitalor
inthecommunity.
Aservicedesignedtodeliverarangeofactivitiesincludingassessmentin
StrokeService
casualty,deliveryofacutecare,followupofoutpatientreview,community
services.
Strokeunit
Anenvironmentinwhichmultidisciplinarystroketeamsdeliverstrokecarein
adedicatedwardwhichhasabedarea,diningarea,gym,andaccessto
assessmentkitchens.
Stylusmazetest
Spatialmemorytaskthoughttobesensitivetofrontalandparietaldamage.
Systematicreview
Researchthatsummarisestheevidenceonaclearlyformulatedquestion
accordingtoapredefinedprotocolusingsystematicandexplicitmethodsto
identify,selectandappraiserelevantstudies,andtoextract,collateand
reporttheirfindings.Itmayormaynotusestatisticalmetaanalysis.
TangentScreenExamination Visualfieldtestusedtoanalyseapatient'svisualfield.
ThePacedAuditorySerial
AdditionTest(PASAT)
Measureofcognitivefunctionthatspecificallyassessesauditoryinformation
processingspeedandflexibility,aswellascalculationability.
Timehorizon
Thetimespanoverwhichcostsandhealthoutcomesareconsideredina
decisionanalysisoreconomicevaluation.
TranscranialMagnetic
Stimulation(TMS)
Noninvasivemethodthatuseselectromagneticinductiontoinduceweak
electriccurrentsusingarapidlychangingmagneticfieldallowingthe
functioningandinterconnectionsofthebraintobestudied.
Treatmentallocation
Assigningaparticipanttoaparticulararmofthetrial.
Unilateralneglect
SeeHemiagnosia
Univariate
Analysiswhichseparatelyexploreseachvariableinadataset.
Utility
Ameasureofthestrengthofanindividualspreferenceforaspecifichealth
stateinrelationtoalternativehealthstates.Theutilityscaleassigns
numericalvaluesonascalefrom0(death)to1(optimalorperfecthealth).
Healthstatescanbeconsideredworsethandeathandthushaveanegative
value.
VisualAnalogueMoodScale
Scaleusedtomeasuretheinternalmoodstateinneurologicallyimpaired
patients.
WechslerAdultIntelligence
ScaleRevisedDigitSpan
Primaryclinicalinstrumentusedtomeasureadultandadolescent
intelligence;itconsistsofsixverbalandfiveperformancesubtests.The
verbaltestsare:Information,Comprehension,Arithmetic,DigitSpan,
Similarities,andVocabulary.ThePerformancesubtestswere:Picture
Arrangement,PictureCompletion,BlockDesign,ObjectAssembly,andDigit
Symbol.Verbal,performanceandfullscaleIntelligenceQuotientscoreswere
alsoobtained.
NationalClinicalGuidelineCentre,2013.
571
Wechslermemoryscale
Testdesignedtomeasuredifferentmemoryfunctionsinaperson.
WesternAphasiaBattery
(WAB)
Instrumentusedtoassessthelanguagefunctionofadults,abletodiscernthe
presence,degree,andtypeofaphasia.
Wiener
Determinationsgerat
Computerassistedreactiontraining,whichmeasuresalertness.
WolfMotorFunctionTest
(WMFT)
Testusedtoquantifyupperextremitymotorabilitythroughtimedand
functionaltasks.
WunndtJastrowIssusion
andreading
Testusedtoassessneglect.
YaleSingleQuestion(YSQ)
AssessmentusedfordepressionthatentailsaskingpatientstheYaleSingle
Question:Doyoufrequentlyfeelsadordepressed?
ZahlenVerbindungsTest
Germanlanguagefreeintelligencetest,thatusesnumberconnectionteststo
assessparticipants.
NationalClinicalGuidelineCentre,2013.
572
18 Referencelist
1
Collaborativesystematicreviewoftherandomisedtrialsoforganisedinpatient(strokeunit)
careafterstroke.StrokeUnitTrialists'Collaboration.BMJ(ClinicalResearchEd).1997;
314(7088):11511159
2
AbenL,HeijenbrokKalMH,VanLoonEMP,GroetE,PondsRWHM,BusschbachJJVetal.
Trainingmemoryselfefficacyinthechronicstageafterstroke:Arandomizedcontrolledtrial.
NeurorehabilitationandNeuralRepair.2012;Epub
3
AidarFJ,SilvaAJ,ReisVM,CarneiroA,CarneiroCottaS.Astudyonthequalityoflifein
ischaemicvascularaccidentsanditsrelationtophysicalactivity.RevNeurol.2007;45(9):518522
4
AitkenPD,RodgersH,FrenchJM,BatesD,JamesOFW.Generalmedicalorgeriatricunitcare
foracutestroke?Acontrolledtrial.Age&Ageing.1993;22(Suppl2):45
5
AlaszewskiA,AlaszewskiH,PotterJ.Thebereavementmodel,strokeandrehabilitation:a
criticalanalysisoftheuseofapsychologicalmodelinprofessionalpractice.Disabilityand
Rehabilitation.2004;26(18):10671078
6
AlonG,LevittAF,McCarthyPA.Functionalelectricalstimulationenhancementofupper
extremityfunctionalrecoveryduringstrokerehabilitation:apilotstudy.Neurorehabilitationand
NeuralRepair.2007;21(3):207215
7
AlonG,LevittAF,McCarthyPA.Functionalelectricalstimulation(FES)maymodifythepoor
prognosisofstrokesurvivorswithseveremotorlossoftheupperextremity:apreliminarystudy.
AmericanJournalofPhysicalMedicineandRehabilitation.2008;87(8):627636
8
AndersonC,MhurchuCN,RubenachS,ClarkM,SpencerC,WinsorA.Homeorhospitalfor
strokerehabilitation?ResultsofarandomizedcontrolledtrialII:costminimizationanalysisat6
months.Stroke.2000;31(5):10321037
9
AndersonC,NiMC,BrownPM,CarterK.Strokerehabilitationservicestoacceleratehospital
dischargeandprovidehomebasedcare:anoverviewandcostanalysis.Pharmacoeconomics.2002;
20(8):537552
10
AndersonC,RubenachS,MhurchuCN,ClarkM,SpencerC,WinsorA.Homeorhospitalfor
strokerehabilitation?resultsofarandomizedcontrolledtrial:I:healthoutcomesat6months.
Stroke.2000;31(5):10241031
11
AndreassenS,WyllerTB.Patients'experienceswithselfreferraltoinpatientrehabilitation:
aqualitativeinterviewstudy.DisabilityandRehabilitation.2005;27(21):13071313
12
AskimT,RohwederG,LydersenS,IndredavikB.Evaluationofanextendedstrokeunitservice
withearlysupporteddischargeforpatientslivinginaruralcommunity.Arandomizedcontrolledtrial.
ClinicalRehabilitation.2004;18(3):238248
13
BakheitAM,ShawS,BarrettL,WoodJ,CarringtonS,GriffithsSetal.Aprospective,
randomized,parallelgroup,controlledstudyoftheeffectofintensityofspeechandlanguage
therapyonearlyrecoveryfrompoststrokeaphasia.ClinicalRehabilitation.2007;21(10):885894
14
BaleM,StrandLI.Doesfunctionalstrengthtrainingoftheleginsubacutestrokeimprove
physicalperformance?Apilotrandomizedcontrolledtrial.ClinicalRehabilitation.2008;22(10
11):911921
NationalClinicalGuidelineCentre,2013.
573
15
BarbeauH,VisintinM.Optimaloutcomesobtainedwithbodyweightsupportcombinedwith
treadmilltraininginstrokesubjects.ArchivesofPhysicalMedicineandRehabilitation.2003;
84(10):14581465
16
BarkerCS,FeiginVL,LawesCM,ParagV,SeniorH,RodgersA.Reducingattentiondeficits
afterstrokeusingattentionprocesstraining:arandomizedcontrolledtrial.Stroke.2009;
40(10):32933298
17
BarkerWH,MulloolyJP.Strokeinadefinedelderlypopulation,19671985.Alesslethaland
disablingbutnolesscommondisease.Stroke.1997;28(2):284290
18
BasaranA,EmreU,KaradavutK,BalbalogluO,BulmusN.Handsplintingforpoststroke
spasticity:Arandomizedcontrolledtrial.TopicsinStrokeRehabilitation.2012;19(4):329337
19
BatemanA,CulpanFJ,PickeringAD,PowellJH,ScottOM,GreenwoodRJ.Theeffectof
aerobictrainingonrehabilitationoutcomesafterrecentseverebraininjury:arandomizedcontrolled
evaluation.ArchivesofPhysicalMedicineandRehabilitation.2001;82(2):174182
20
BautzHoltertE,SveenU,RyghJ,RodgersH,WyllerTB.Earlysupporteddischargeofpatients
withacutestroke:arandomizedcontrolledtrial.DisabilityandRehabilitation.2002;24(7):348355
21
BeckermanH.Walkingabilityofstrokepatients:efficacyoftibialnerveblockinganda
polypropyleneanklefootorthosis.ArchivesofPhysicalMedicineandRehabilitation.1996;
77(11):11441151
22
BeechR,RuddAG,TillingK,WolfeCD.Economicconsequencesofearlyinpatientdischarge
tocommunitybasedrehabilitationforstrokeinaninnerLondonteachinghospital.Stroke.1999;
30(4):729735
23
BendzM.Thefirstyearofrehabilitationafterastrokefromtwoperspectives.Scandinavian
JournalofCaringSciences.2003;17(3):215222
24
BeninatoM,GillBodyKM,SallesS,StarkPC,BlackSchafferRM,SteinJ.Determinationofthe
minimalclinicallyimportantdifferenceintheFIMinstrumentinpatientswithstroke.Archivesof
PhysicalMedicineandRehabilitation.2006;87(1):3239
25
BhatnagarP,ScarboroughP,SmeetonNC,AllenderS.Theincidenceofallstrokeandstroke
subtypeintheUnitedKingdom,1985to2008:asystematicreview.BMCPublicHealth.2010;10:539
26
Biomedical.com.
27
BlennerhassettJ,DiteW.Additionaltaskrelatedpracticeimprovesmobilityandupperlimb
functionearlyafterstroke:arandomisedcontrolledtrial.AustralianJournalofPhysiotherapy.2004;
50(4):219224
28
BoutinLesterP,GibsonRW.Patients'perceptionsofhomehealthoccupationaltherapy.
AustralianOccupationalTherapyJournal.2002;49(3):146154
29
BowenA,HeskethA,PatchickE,YoungA,DaviesL,VailAetal.Clinicaleffectiveness,cost
effectivenessandserviceusers'perceptionsofearly,wellresourcedcommunicationtherapy
followingastroke:arandomisedcontrolledtrial(theACTNoWStudy).HealthTechnolAssess.2012;
16(26):1160
30
BradyMC,KellyH,GodwinJ,EnderbyP.Speechandlanguagetherapyforaphasiafollowing
stroke.CochraneDatabaseofSystematicReviews(Online).2012;5:CD000425
31
BrazzelliM,SaundersDH,GreigCA,MeadGE.Physicalfitnesstrainingforstrokepatients.
CochraneDatabaseSystRev.2011;11:CD003316
NationalClinicalGuidelineCentre,2013.
574
32
BritishSocietyofRehabilitationMedicine.Vocationalassessmentandrehabilitationforlong
termneurologicalconditions:recommendationsforbestpractice.London.BritishSocietyof
RehabilitationMedicine,2010
33
BritishSocietyofRehabilitationMedicine,DepartmentforWorkandPensions,andRoyal
CollegeofPhysicians.Vocationalassessmentandrehabilitationafteracquiredbraininjury:inter
agencyguidelines.London.BSRM,2004
34
BronnumHansenH,DavidsenM,ThorvaldsenP.Longtermsurvivalandcausesofdeath
afterstroke.Stroke.2001;32(9):21312136
35
CabralNL,MoroC,SilvaGR,ScolaRH,WerneckLC.Studycomparingthestrokeunitoutcome
andconventionalwardtreatment:arandomizedstudyinJoinville,Brazil.ArquivosDeNeuro
Psiquiatria.2003;61(2A):188193
36
CarnabyG,HankeyGJ,PizziJ.Behaviouralinterventionfordysphagiainacutestroke:a
randomisedcontrolledtrial.TheLancetNeurology.:ElsevierLimited.2006;5(1):31
37
CarterLT,HowardBE,O'NeilWA.Effectivenessofcognitiveskillremediationinacutestroke
patients.AmericanJournalofOccupationalTherapy.1983;37(5):320326
38
CauraughJ,LightK,KimS,ThigpenM,BehrmanA.Chronicmotordysfunctionafterstroke:
recoveringwristandfingerextensionbyelectromyographytriggeredneuromuscularstimulation.
Stroke.2000;31(6):13601364
39
CauraughJH,KimS.Twocoupledmotorrecoveryprotocolsarebetterthanone:
electromyogramtriggeredneuromuscularstimulationandbilateralmovements.Stroke.2002;
33(6):15891594
40
CentreforBiologicalInformationTechnology(UniversityofQueensland).
41
ChaeJ,BethouxF,BohineT,DobosL,DavisT,FriedlA.Neuromuscularstimulationforupper
extremitymotorandfunctionalrecoveryinacutehemiplegia.Stroke;aJournalofCerebral
Circulation.1998;29(5):975979
42
ChanMK,TongRK,ChungKY.Bilateralupperlimbtrainingwithfunctionalelectric
stimulationinpatientswithchronicstroke.NeurorehabilitationandNeuralRepair.2009;23(4):357
365
43
ChiuCWY,ManDWK.Theeffectoftrainingolderadultswithstroketousehomebased
assistivedevices.OTJR:Occupation,Participation&Health.2004;24(3):113120
44
ClaessonL,GosmanHedstromG,JohannessonM,FagerbergB,BlomstrandC.Resource
utilizationandcostsofstrokeunitcareintegratedinacarecontinuum:A1yearcontrolled,
prospective,randomizedstudyinelderlypatients.TheGoteborg70+strokestudy.Stroke.2000;
31:25692577:25692577
45
ClisbyC.Visualassessmentofpatientswithcerebrovascularaccidentontheelderlycare
wards.BritishOrthopticJournal.1995;52:3840
46
CombsSA,KellySP,BartonR,IvaskaM,NowakK.Effectsofanintensive,taskspecific
rehabilitationprogramforindividualswithchronicstroke:acaseseries.DisabilityandRehabilitation.
2010;32(8):669678
47
CookeEV,TallisRC,ClarkA,PomeroyVM.Efficacyoffunctionalstrengthtrainingon
restorationoflowerlimbmotorfunctionearlyafterstroke:phaseIrandomizedcontrolledtrial.
NeurorehabilitationandNeuralRepair.2010;24(1):8896
NationalClinicalGuidelineCentre,2013.
575
48
CorrS,BayerA.Occupationaltherapyforstrokepatientsafterhospitaldischargea
randomizedcontrolledtrial.ClinicalRehabilitation.1995;9(4):291296
49
CottCA.Clientcentredrehabilitation:clientperspectives.DisabilityandRehabilitation.2004;
26(24):14111422
50
CurtisL.UnitCostsofHealthandSocialCare.Kent:UK.PersonalSocialServicesResearch
Unit,2010
51
CurtisL.UnitCostsofHealthandSocialCare2012,2012
52
CuvielloPalmerED.EffectoftheKinetronIIongaitandfunctionaloutcomeinhemiplegic
subjectsTexasWomensUniversity;1988.
53
daCunhaIT,Jr.,LimPA,QureshyH,HensonH,MongaT,ProtasEJ.Gaitoutcomesafteracute
strokerehabilitationwithsupportedtreadmillambulationtraining:arandomizedcontrolledpilot
study.ArchivesofPhysicalMedicineandRehabilitation.2002;83(9):12581265
54
DahlAE,AskimT,StockR,LangorgenE,LydersenS,IndredavikB.Shortandlongterm
outcomeofconstraintinducedmovementtherapyafterstroke:arandomizedcontrolledfeasibility
trial.ClinicalRehabilitation.2008;22(5):436447
55
DanielsR,WindingK,BorellL.Experiencesofocuupationaltherapistsinstrokerehabilitation:
dilemmasofsomeoccupationaltherapistsininpatientstrokerehabilitation.ScandinavianJournalof
OccupationalTherapy.2002;9:167175
56
DavidR,EnderbyP,BaintonD.Treatmentofacquiredaphasia:speechtherapistsand
volunteerscompared.JournalofNeurology,NeurosurgeryandPsychiatry.1982;45(11):957961
57
deWitDC,BuurkeJH,NijlantJM,IjzermanMJ,HermensHJ.Theeffectofananklefoot
orthosisonwalkingabilityinchronicstrokepatients:arandomizedcontrolledtrial.Clinical
Rehabilitation.2004;18(5):550557
58
DeanCM,RichardsCL,MalouinF.Taskrelatedcircuittrainingimprovesperformanceof
locomotortasksinchronicstroke:arandomized,controlledpilottrial.ArchivesofPhysicalMedicine
andRehabilitation.2000;81(4):409417
59
DenesG,PerazzoloC,PianiA,PiccioneF.Intensiveversusregularspeechtherapyinglobal
aphasia:Acontrolledstudy.Aphasiology.1996;10(4):385394
60
DepartmentofHealth.TheNationalServiceFrameworkforLongtermConditions.London.
DepartmentofHealth,2005
61
DepartmentofHealth.Nationalstrokestrategy.London.DH,2007
62
DepartmentofHealth.Startactivestayactive:areportonphysicalactivitiyfromthefour
countries'ChiefMedicalOfficers.London.DH,2011
63
DePippoKL,HolasMA,RedingMJ,MandelFS,LesserML.Dysphagiatherapyfollowing
stroke:acontrolledtrial.Neurology.1994;44(9):16551660
64
DiasD,LainsJ,PereiraA,NunesR,CaldasJ,AmaralCetal.Canweimprovegaitskillsin
chronichemiplegics?Arandomisedcontroltrialwithgaittrainer.EuropaMedicophysica.2007;
43(4):499504
65
DicksonS,BarbourRS,ClarkAM,PatonG.Patients'experiencesofdisruptionsassociated
withpoststrokedysarthria.InternationalJournalofLanguageandCommunicationDisorders.2008;
43(2):135153
NationalClinicalGuidelineCentre,2013.
576
66
DoesborghSJ,vandeSandtKoendermanMW,DippelDW,vanHF,KoudstaalPJ,VischBrink
EG.Effectsofsemantictreatmentonverbalcommunicationandlinguisticprocessinginaphasiaafter
stroke:arandomizedcontrolledtrial.Stroke;aJournalofCerebralCirculation.2004;35(1):141146
67
DoesborghSJC,vandeSandtKoendermanMWME,DippelDWJ,vanHF,KoudstaalPJ,Visch
BrinkEG.Cuesonrequest:Theefficacyofmulticue,acomputerprogramforwordfindingtherapy.
Aphasiology.2004;18(3):213222
68
DonaldsonC,TallisR,MillerS,SunderlandA,LemonR,PomeroyV.Effectsofconventional
physicaltherapyandfunctionalstrengthtrainingonupperlimbmotorrecoveryafterstroke:a
randomizedphaseIIstudy.NeurorehabilitationandNeuralRepair.2009;23(4):389397
69
DonnellyM,PowerM,RussellM,FullertonK.Randomizedcontrolledtrialofanearly
dischargerehabilitationservice:theBelfastcommunitystroketrial.Stroke.2004;35:127133
70
DoornheinK,DeHaanEHF.CognitiveTrainingforMemoryDeficitsinStrokePatients.
NeuropsychologicalRehabilitation.:PsychologyPress.1998;8(4):393400
71
DriversMedicalGroup.AtaglanceGuidetothecurrentMedicalStandardsofFitnessto
Drive.Swansea.DriverandVechicleLicensingAgency,2011
72
DromerickAW,EdwardsDF,HahnM.Doestheapplicationofconstraintinducedmovement
therapyduringacuterehabilitationreducearmimpairmentafterischemicstroke?Stroke.2000;
31(12):29842988
73
DuncanPW,SullivanKJ,BehrmanAL,AzenSP,WuSS,NadeauSEetal.Bodyweight
supportedtreadmillrehabilitationafterstroke.NewEnglandJournalofMedicine.2011;
364(21):20262036
74
EichHJ,MachH,WernerC,HesseS.AerobictreadmillplusBobathwalkingtrainingimproves
walkinginsubacutestroke:arandomizedcontrolledtrial.ClinicalRehabilitation.2004;18(6):640651
75
EllisG,RodgerJ,McAlpineC,LanghorneP.Theimpactofstrokenursespecialistinputonrisk
factormodification:arandomisedcontrolledtrial.AgeandAgeing.2005;34(4):389392
76
ElmanRJ,BernsteinEllisE.Theefficacyofgroupcommunicationtreatmentinadultswith
chronicaphasia.JournalofSpeech,Language,andHearingResearch:JSLHR.1999;42(2):411419
77
ErelS,UygurF,EnginS,I,YakutY.Theeffectsofdynamicanklefootorthosesinchronic
strokepatientsatthreemonthfollowup:arandomizedcontrolledtrial.ClinicalRehabilitation.2011;
25(6):515523
78
FagerbergB,ClaessonL,GosmanHedstromG,BlomstrandC.Effectofacutestrokeunitcare
integratedwithcarecontinuumversusconventionaltreatment:Arandomized1yearstudyofelderly
patients:theGoteborg70+StrokeStudy.Stroke.2000;31(11):25782584
79
FalkeP,AbelaBM,Jr.,KrakauCE,LiljaB,LindgardeF,MalyPetal.Highfrequencyof
asymptomaticvisualfielddefectsinsubjectswithtransientischaemicattacksorminorstrokes.
JournalofInternalMedicine.1991;229(6):521525
80
FanthomeY,LincolnNB,DrummondA,WalkerMF.Thetreatmentofvisualneglectusing
feedbackofeyemovements:apilotstudy.DisabilityandRehabilitation.1995;17(8):413417
81
FeldmanDJ,LeePR,UntereckerJ,LloydK,RuskHA,TooleA.Acomparisonoffunctionally
orientatedmedicalcareandformalrehabilitationinthemanagementofpatientswithhemiplegia
duetocerebrovasculardisease.JournalofChronicDiseases.1962;15:297310
NationalClinicalGuidelineCentre,2013.
577
82
FjaeartoftH,IndredavikB,JohnsenR,LydersenS.Acutestrokeunitcarecombinedwithearly
supporteddischarge.Longtermeffectsonqualityoflife.Arandomizedcontrolledtrial.Clinical
Rehabilitation.2004;18(5):580586
83
FjaertoftH,IndredavikB,MagnussenJ,JohnsenR.Earlysupporteddischargeforstroke
patientsimprovesclinicaloutcome.Doesitalsoreduceuseofhealthservicesandcosts?Oneyear
followupofarandomizedcontrolledtrial.CerebrovascularDiseases.2005;19:376383
84
FlansbjerUB,MillerM,DownhamD,LexellJ.Progressiveresistancetrainingafterstroke:
effectsonmusclestrength,muscletone,gaitperformanceandperceivedparticipation.Journalof
RehabilitationMedicine.2008;40(1):4248
85
FoyeSJ,KirschnerKL,BradyWagnerLC,StockingC,SieglerM.Ethicalissuesinrehabilitation:
aqualitativeanalysisofdilemmasidentifiedbyoccupationaltherapists.TopicsinStroke
Rehabilitation.2002;9(3):89101
86
FranceschiniM,CardaS,AgostiM,AntenucciR,MalgratiD,CisariCetal.Walkingafter
stroke:whatdoestreadmilltrainingwithbodyweightsupportaddtoovergroundgaittrainingin
patientsearlyafterstroke?:asingleblind,randomized,controlledtrial.Stroke.2009;40(9):3079
3085
87
FrassinettiF,AngeliV,MeneghelloF,AvanziS,LadavasE.Longlastingameliorationof
visuospatialneglectbyprismadaptation.Brain.2002;125(Pt3):608623
88
GalvinR,CusackT,O'GradyE,MurphyTB,StokesE.Familymediatedexerciseintervention
(FAME):evaluationofanovelformofexercisedeliveryafterstroke.Stroke;aJournalofCerebral
Circulation.2011;42(3):681686
89
GaronBR,EngleM,OrmistonC.Arandomizedcontrolstudytodeterminetheeffectsof
unlimitedoralintakeofwaterinpatientswithidentifiedaspiration.JournalofNeurologic
Rehabilitation.1997;11(3):139148
90
GarrawayWM,AkhtarAJ,HockeyL,PrescottRJ.Managementofacutestrokeintheelderly:
followupofacontrolledtrial.BritishMedicalJournal.1980;281(6244):827829
91
GilbertsonL,LanghorneP,WalkerA,AllenA,MurrayGD.Domiciliaryoccupationaltherapy
forpatientswithstrokedischargedfromhospital:randomisedcontrolledtrial.BritishMedical
Journal.2000;320(7235):603606
92
GilhotraJS,MitchellP,HealeyPR,CummingRG,CurrieJ.Homonymousvisualfielddefects
andstrokeinanolderpopulation.Stroke.2002;33(10):24172420
93
GlasserL.Effectsofisokinetictrainingontherateofmovementduringambulationin
hemipareticpatients.PhysicalTherapy.1986;66(5):673676
94
GlobasC,BeckerC,CernyJ,LamJM,LindemannU,ForresterLWetal.Chronicstroke
survivorsbenefitfromhighintensityaerobictreadmillexercise:arandomizedcontroltrial.
NeurorehabilitationandNeuralRepair.2012;26(1):8595
95
GoodmanC.ADelphiSurveyofclinicalnursingresearchprioritieswithinaregionalhealth
authorityUniversityofLondon;1986.
96
GordonEE,KohnKH.Evaluationofrehabilitationmethodsinthehemiplegicpatient.Journal
ofChronicDiseases.1966;19(1):316
97
GordonTJ.Currentmethodsoffutureresearch.InsititutefortheFuture;1971
NationalClinicalGuidelineCentre,2013.
578
98
GRADEWorkingGroup.GradingofRecommendationsAssessment,Developmentand
Evaluationworkinggroup.2011.
99
GrayCS,FrenchJM,BatesD,CartlidgeNE,VenablesGS,JamesOF.Recoveryofvisualfieldsin
acutestroke:homonymoushemianopiaassociatedwithadverseprognosis.AgeandAgeing.1989;
18(6):419421
100
HaleLA.UsingGoalAttainmentScalinginphysiotherapeutichomebasedstroke
rehabilitation.AdvancesinPhysiotherapy.2010;12(3):142149
101
HaleLA,PiggotJ.Exploringthecontentofphysiotherapeutichomebasedstroke
rehabilitationinNewZealand.ArchivesofPhysicalMedicineandRehabilitation.2005;86(10):1933
1940
102
HammerAM,LindmarkB.Effectsofforceduseonarmfunctioninthesubacutephaseafter
stroke:Arandomized,clinicalpilotstudy.PhysicalTherapy.2009;89(6):526539
103
HankeyGJ,DeleoD,StewartWynneEG.Strokeunits:anAustralianperspective.Australian
andNewZealandJournalofMedicine.1997;27(4):437438
104
HaraY,MuraokaY.Ahomeprogramofpowerassistedfunctionalelectricalstimulationwith
upperlimbtaskpracticeinchronicstrokepatients.InternationalJournalofStroke.2008;3(Suppl
1):342
105
HaraY,OgawaS,TsujiuchiK,MuraokaY.Ahomebasedrehabilitationprogramforthe
hemiplegicupperextremitybypowerassistedfunctionalelectricalstimulation.Disabilityand
Rehabilitation.2008;30(4):296304
106
HardenA,BruntonG,FletcherA,OakleyA,BurchettH,andBackhansM.Youngpeople,
pregnancyandsocialexclusion:asystematicsynthesisofresearchevidencetoidentifyandeffective,
approriateandpromisingapproachesforpreventionandsupport.London:UK.EPPICentre,Social
ScienceResearchUnit,InstituteofEducation,UniversityofLondon,2006
107
HartmanJ,LandauWM.Comparisonofformallanguagetherapywithsupportivecounseling
foraphasiaduetoacutevascularaccident.ArchivesofNeurology.1987;44(6):646649
108
HerMajesty'sStationeryOffice.MentalCapacityAct2005(c.9).London.HMSO,2005
109
HesseS,BerteltC,SchaffrinA,MalezicM,MauritzKH.Restorationofgaitinnonambulatory
hemipareticpatientsbytreadmilltrainingwithpartialbodyweightsupport.ArchivesofPhysical
MedicineandRehabilitation.1994;75(10):10871093
110
HidlerJ,NicholsD,PelliccioM,BradyK,CampbellDD,KahnJHetal.Multicenterrandomized
clinicaltrialevaluatingtheeffectivenessoftheLokomatinsubacutestroke.Neurorehabilitationand
NeuralRepair.2009;23(1):513
111
HigginsJ,SalbachNM,WoodDauphineeS,RichardsCL,CoteR,MayoNE.Theeffectofa
taskorientedinterventiononarmfunctioninpeoplewithstroke:arandomizedcontrolledtrial.
ClinicalRehabilitation.2006;20(4):296310
112
HoffmannT,McKennaK,WorrallL,ReadSJ.Randomisedtrialofacomputergenerated
tailoredwritteneducationpackageforpatientsfollowingstroke.AgeandAgeing.2007;36(3):280
286
113
HolmgrenE,GosmanHedstromG,LindstromB,WesterP.Whatisthebenefitofahigh
intensiveexerciseprogramonhealthrelatedqualityoflifeanddepressionafterstroke?A
randomizedcontrolledtrial.AdvancesinPhysiotherapy.2010;12(3):125133
NationalClinicalGuidelineCentre,2013.
579
114
HornbyTG,CampbellDD,KahnJH,DemottT,MooreJL,RothHR.Enhancedgaitrelated
improvementsaftertherapistversusroboticassistedlocomotortraininginsubjectswithchronic
stroke:Arandomizedcontrolledstudy.Stroke.2008;39(6):17861792
115
HoyerE,JahnsenR,StanghelleJK,StrandLI.Bodyweightsupportedtreadmilltrainingversus
traditionaltraininginpatientsdependentonwalkingassistanceafterstroke:arandomizedcontrolled
trial.DisabilityandRehabilitation.2012;34(3):210219
116
HsiehYW,WangCH,WuSC,ChenPC,SheuCF,HsiehCL.Establishingtheminimalclinically
importantdifferenceoftheBarthelIndexinstrokepatients.NeurorehabilitationandNeuralRepair.
2007;21(3):233238
117
HsuSS,HuMH,WangYH,YipPK,ChiuJW,HsiehCL.DoseResponseRelationBetween
NeuromuscularElectricalStimulationandUpperExtremityFunctioninPatientsWithStroke.Stroke.
2010;41(4):821824
118
HusemannB,MullerF,KrewerC,HellerS,KoenigE.Effectsoflocomotiontrainingwith
assistanceofarobotdrivengaitorthosisinhemipareticpatientsafterstroke:arandomized
controlledpilotstudy.Stroke.2007;38(2):349354
119
IanesP,VaraltaV,GandolfiM,PicelliA,CornoM,DiMatteoAetal.Stimulatingvisual
explorationoftheneglectedspaceintheearlystageofstrokebyhemifieldeyepatching:a
randomizedcontrolledtrialinpatientswithrightbraindamage.EuropeanJournalofPhysicaland
RehabilitationMedicine.2012;48(2):189196
120
IndredavikB,BakkeF,SolbergR,RoksethR,HaaheimLL,HolmeI.Benefitofastrokeunit:a
randomizedcontrolledtrial.Stroke;aJournalofCerebralCirculation.1991;22(8):10261031
121
IndredavikB,FjaertoftH,EkebergG,LogeAD,MorchB.Benefitofanextendedstrokeunit
servicewithearlysupporteddischarge:Arandomized,controlledtrial.Stroke.2000;31(12):2989
2994
122
IntercollegiateStrokeWorkingParty.Nationalclinicalguidelineforstroke:Thirdedition(July
2008).RoyalCollegeofPhysicians,2008
123
IntercollegiateStrokeWorkingParty.NationalSentinelStrokeClinicalAudit2010(Round7):
publicreportforEngland,WalesandNorthernIreland.London.RoyalCollegeofPhysicians,2011
124
JaeschkeR,SingerJ,GuyattGH.Measurementofhealthstatus.Ascertainingtheminimal
clinicallyimportantdifference.ControlledClinicalTrials.1989;10(4):407415
125
JinH,JiangY,WeiQ,WangB,MaG.Intensiveaerobiccyclingtrainingwithlowerlimb
weightsinChinesepatientswithchronicstroke:discordancebetweenimprovedcardiovascular
fitnessandwalkingability.DisabilityandRehabilitation.2012;34(19):16651671
126
JubyLC,LincolnNB,BermanP.Theeffectofa
strokerehabilitationunitonfunctionalandpsychologicaloutcome.Arandomisedcontrolledtrial.
CerebrovascularDiseases.1996;6:106110
127
KaganA,BlackSE,DuchanFJ,SimmonsMackieN,SquareP.Trainingvolunteersas
conversationpartnersusing"SupportedConversationforAdultswithAphasia"(SCA):acontrolled
trial.JournalofSpeechLanguage&HearingResearch.2001;44(3):624638
128
KalraL,DaleP,CromeP.Improvingstrokerehabilitation.Acontrolledstudy.Stroke.1993;
24(10):14621467
NationalClinicalGuidelineCentre,2013.
580
129
KalraL,EadeJ.Roleofstrokerehabilitationunitsinmanagingseveredisabilityafterstroke.
Stroke;aJournalofCerebralCirculation.1995;26(11):20312034
130
KalraL,PerezI,GuptaS,WittinkM.Theinfluenceofvisualneglectonstrokerehabilitation.
Stroke.1997;28(7):13861391
131
KangHK,KimY,ChungY,HwangS.Effectsoftreadmilltrainingwithopticflowonbalance
andgaitinindividualsfollowingstroke:randomizedcontrolledtrials.ClinicalRehabilitation.2012;
26(3):246255
132
KasteM,PalomakiH,SarnaS.Whereandhowshouldelderlystrokepatientsbetreated?A
randomizedtrial.Stroke.1995;26(2):249253
133
KatzRC,WertzRT.Theefficacyofcomputerprovidedreadingtreatmentforchronicaphasic
adults.JournalofSpeechLanguage&HearingResearch.1997;40(3):493507
134
KatzLeurerM,CarmeliE,ShochinaM.Theeffectofearlyaerobictrainingonindependence
sixmonthspoststroke.ClinicalRehabilitation.2003;17(7):735741
135
KeeneyS,McKennaH,HassonF.TheDelphiTechniqueinnursingandhealthresearch.1st
edition.Chichester:JohnWileyandSonsLtd;2011
136
KimCM,EngJJ,MacIntyreDL,DawsonAS.Effectsofisokineticstrengthtrainingonwalkingin
personswithstroke:adoubleblindcontrolledpilotstudy.JournalofStrokeandCerebrovascular
Diseases.2001;10(6):265273
137
KimberleyTJ,LewisSM,AuerbachEJ,DorseyLL,LojovichJM,CareyJR.Electricalstimulation
drivingfunctionalimprovementsandcorticalchangesinsubjectswithstroke.ExperimentalBrain
Research.2004;154(4):450460
138
KosakMC,RedingMJ.Comparisonofpartialbodyweightsupportedtreadmillgaittraining
versusaggressivebracingassistedwalkingpoststroke.NeurorehabilitationandNeuralRepair.2000;
14(1):1319
139
KuysSS,BrauerSG,AdaL.Higherintensitytreadmillwalkingduringrehabilitationafter
strokeinfeasibleandnotdetrimentaltowalkingpatternorquality:apilotrandomizedtrial.Clinical
Rehabilitation.2011;25(4):316326
140
KwakkelG,WagenaarRC,TwiskJW,LankhorstGJ,KoetsierJC.Intensityoflegandarm
trainingafterprimarymiddlecerebralarterystroke:arandomisedtrial.Lancet.1999;
354(9174):191196
141
LangCE,EdwardsDF,BirkenmeierRL,DromerickAW.Estimatingminimalclinicallyimportant
differencesofupperextremitymeasuresearlyafterstroke.ArchivesofPhysicalMedicineand
Rehabilitation.2008;89(9):16931700
142
LanghammerB,LindmarkB,StanghelleJK.Strokepatientsandlongtermtraining:Isit
worthwhile?Arandomizedcomparisonoftwodifferenttrainingstrategiesafterrehabilitation.
ClinicalRehabilitation.2007;21(6):495510
143
LanghammerB,StanghelleJK.Exerciseonatreadmillorwalkingoutdoors?Arandomized
controlledtrialcomparingeffectivenessoftwowalkingexerciseprogrammeslateafterstroke.
ClinicalRehabilitation.2010;24(1):4654
144
LanninNA,CusickA,McCluskeyA,HerbertRD.Effectsofsplintingonwristcontractureafter
stroke:arandomizedcontrolledtrial.Stroke.2007;38(1):111116
NationalClinicalGuidelineCentre,2013.
581
145
LanninNA,HerbertRD.Ishandsplintingeffectiveforadultsfollowingstroke?Asystematic
reviewandmethodologiccritiqueofpublishedresearch.ClinicalRehabilitation.2003;17(8):807816
146
LarsenT,OlsenTS,SorensenJ.Earlyhomesupporteddischargeofstrokepatients:ahealth
technologyassessment.InternationalJournalofTechnologyAssessmentinHealthCare.2006;
22(3):313320
147
LauferY,DicksteinR,ChefezY,MarcovitzE.Theeffectoftreadmilltrainingonthe
ambulationofstrokesurvivorsintheearlystagesofrehabilitation:arandomizedstudy.Journalof
RehabilitationResearch&Development.2001;38(1):6978
148
LaursenSO,HenriksenIO,DonsU,JacobsenB,GundertofteL.[Intensiverehabilitationafter
apoplexyacontrolledpilotstudy].UgeskrLaeger.1995;157(14):19961999
149
LawlerJ,DowswellG,HearnJ,ForsterA,YoungJ.Recoveringfromstroke:aqualitative
investigationoftheroleofgoalsettinginlatestrokerecovery.JournalofAdvancedNursing.1999;
30(2):401409
150
LeachE,CornwellP,FlemingJ,HainesT.Patientcenteredgoalsettinginasubacute
rehabilitationsetting.DisabilityandRehabilitation.2010;32(2):159172
151
LeggLA,DrummondAE,LanghorneP.Occupationaltherapyforpatientswithproblemsin
activitiesofdailylivingafterstroke.CochraneDatabaseofSystematicReviews.2006;(4):CD003585
152
LennonO,CareyA,GaffneyN,StephensonJ,BlakeC.Apilotrandomizedcontrolledtrialto
evaluatethebenefitofthecardiacrehabilitationparadigmforthenonacuteischaemicstroke
population.ClinRehabil.2008;22(2):125133
153
LinKC,ChangYF,WuCY,ChenYA.Effectsofconstraintinducedtherapyversusbilateral
armtrainingonmotorperformance,dailyfunctions,andqualityoflifeinstrokesurvivors.
NeurorehabilitationandNeuralRepair.2009;23(5):441448
154
LinKC,WuCY,WeiTH,LeeCY,LiuJS.Effectsofmodifiedconstraintinducedmovement
therapyonreachtograspmovementsandfunctionalperformanceafterchronicstroke:A
randomizedcontrolledstudy.ClinicalRehabilitation.2007;21(12):10751086
155
LinZ,YanT.Longtermeffectivenessofneuromuscularelectricalstimulationforpromoting
motorrecoveryoftheupperextremityafterstroke.JournalofRehabilitationMedicine.2011;
43(6):506510
156
LincolnNB,McGuirkE,MulleyGP,LendremW,JonesAC,MitchellJR.Effectivenessofspeech
therapyforaphasicstrokepatients.Arandomisedcontrolledtrial.Lancet.1984;1(8388):11971200
157
LincolnNB,PickersgillMJ,HankeyAI,HiltonCR.Anevaluationofoperanttrainingandspeech
therapyinthelangaugerehabilitationofmoderateaphasics.BehaviouralPsychotherapy.1982;
10(2):162178
158
LoganPA,AhernJ,GladmanJR,LincolnNB.ArandomizedcontrolledtrialofenhancedSocial
Serviceoccupationaltherapyforstrokepatients.ClinicalRehabilitation.1997;11(2):107113
159
LoweD,SharmaA,LeathleyM.TheCareFileproject:afeasibilitystudytoexaminetheeffects
ofanindividualisedinformationbookletonpatientsafterstroke.AgeandAgeing.2007;36(1):8389
160
LuftAR,MackoRF,ForresterLW,VillagraF,IveyF,SorkinJDetal.Treadmillexercise
activatessubcorticalneuralnetworksandimproveswalkingafterstroke:arandomizedcontrolled
trial.Stroke.2008;39(12):33413350
NationalClinicalGuidelineCentre,2013.
582
161
LundstromE,TerentA,BorgJ.Prevalenceofdisablingspasticity1yearafterfirsteverstroke.
EurJNeurol.2008;15(6):533539
162
MaRH,WangYJ,QuH,YangZH.Assessmentoftheearlyeffectivenessofastrokeunitin
comparisontothegeneralward.ChinMedJ(Engl).2004;117(6):852855
163
MacKayS,HolmesDW,GersumkyAT.Methodstoassessaphasicstrokepatients.Geriatric
Nursing.1988;9(3):177179
164
MaitraKK,ErwayF.Perceptionofclientcenteredpracticeinoccupationaltherapistsand
theirclients.AmericanJournalofOccupationalTherapy:OfficialPublicationoftheAmerican
OccupationalTherapyAssociation.2006;60(3):298310
165
MajorK,WalkerA.Economicsofstrokecareunit.In:LanghorneP,MartinD(eds),Stroke
units:anevidencebasedapproach,London:BMJ,1998:5665
166
MancusoM,PaciniM,GemignaniP,BartaliniB,AgostiniB,FerroniLetal.Clinicalapplication
ofprismaticlensesintherehabilitationofneglectpatients.Arandomizedcontrolledtrial.European
JournalofPhysicalandRehabilitationMedicine.2012;48(2):197208
167
MangoldS,SchusterC,KellerT,ZimmermannSchlatterA,EttlinT.Motortrainingofupper
extremitywithfunctionalelectricalstimulationinearlystrokerehabilitation.Neurorehabilitationand
NeuralRepair.2009;23(2):184190
168
MannGE,BurridgeJH,MaloneLJ,StrikePW.Apilotstudytoinvestigatetheeffectsof
electricalstimulationonrecoveryofhandfunctionandsensationinsubacutestrokepatients.
Neuromodulation.2005;8(3):193202
169
MarshK,BertranouE,SuominenH,andVenkatachalamM.Aneconomicevaluationof
speechandlanguagetherapy.London.MatrixEvidence,2010
170
MayoNE,WoodDauphineeS,CoteR,GaytonD,CarltonJ,ButteryJetal.There'snoplace
likehome:anevaluationofearlysupporteddischargeforstroke.Stroke.2000;31(5):10161023
171
McGrathJR,AdamsL.Patientcenteredgoalplanning:asystemicpsychologicaltherapy?
TopicsinStrokeRehabilitation.1999;6:4350
172
McNameeP,ChristensenJAKO,SoutterJENN,RodgersH,CraigN,PearsonPetal.Cost
analysisofearlysupportedhospitaldischargeforstroke.AgeandAgeing.1998;27(3):345351
173
MiltnerWH,BauderH,SommerM,DettmersC,TaubE.Effectsofconstraintinduced
movementtherapyonpatientswithchronicmotordeficitsafterstroke:areplication.Stroke.1999;
30(3):586592
174
ModdenC,BehrensM,DamkeI,EilersN,KastrupA,HildebrandtH.Arandomizedcontrolled
trialcomparing2interventionsforvisualfieldlosswithstandardoccupationaltherapyduring
inpatientstrokerehabilitation.NeurorehabilitationandNeuralRepair.2012;26(5):463469
175
MonaghanJ,ChannellK,McDowellD,SharmaAK.Improvingpatientandcarer
communication,multidisciplinaryteamworkingandgoalsettinginstrokerehabilitation.Clinical
Rehabilitation.2005;19(2):194199
176
MoodieM,CadilhacD,PearceD,MihalopoulosC,CarterR,DavisSetal.Economicevaluation
ofAustralianstrokeservices:aprospective,multicenterstudycomparingdedicatedstrokeunitswith
othercaremodalities.Stroke.2006;37(11):27902795
NationalClinicalGuidelineCentre,2013.
583
177
MooreJL,RothEJ,KillianC,HornbyTG.Locomotortrainingimprovesdailysteppingactivity
andgaitefficiencyinindividualspoststrokewhohavereacheda"plateau"inrecovery.Stroke.2010;
41(1):129135
178
MorelandJD,GoldsmithCH,HuijbregtsMP,AndersonRE,PrenticeDM,BruntonKBetal.
Progressiveresistancestrengtheningexercisesafterstroke:asingleblindrandomizedcontrolled
trial.ArchivesofPhysicalMedicineandRehabilitation.2003;84(10):14331440
179
MoroneG,IosaM,BragoniM,DeAD,VenturieroV,CoiroPetal.Whomayhavedurable
benefitfromroboticgaittraining?:A2yearfollowuprandomizedcontrolledtrialinpatientswith
subacutestroke.Stroke.2012;43(4):11401142
180
MudgeS,BarberPA,StottNS.Circuitbasedrehabilitationimprovesgaitendurancebutnot
usualwalkingactivityinchronicstroke:arandomizedcontrolledtrial.ArchPhysMedRehabil.2009;
90(12):19891996
181
MurphyMK,BlackNA,LampingDL,McKeeCM,SandersonCF,AskhamJetal.Consensus
developmentmethods,andtheiruseinclinicalguidelinedevelopment.HealthTechnology
Assessment.1998;2(3):188
182
MyintJM,YuenGF,YuTK,KngCP,WongAM,ChowKKetal.Astudyofconstraintinduced
movementtherapyinsubacutestrokepatientsinHongKong.ClinicalRehabilitation.2008;22(2):112
124
183
NationalAuditOffice.ProgressinImprovingStrokeCare:reportbytheComptrollerand
AuditorGeneral.London.NationalAuditOffice,2010Availablefrom:
http://www.nao.org.uk/publications/0910/stroke.aspx
184
NationalHealthService.CriticalAppraisalSkillsProgramme(CASP).2003.Availablefrom:
http://www.phru.nhs.uk/Pages/PHD/CASP.htm
185
NationalInstituteforHealthandClinicalExcellence.Guidetothemethodsoftechnology
appraisal.London.NICE,2008
186
NationalInstituteforHealthandClinicalExcellence.Socialvaluejudgements:principlesof
thedevelopmentofNICEguidance(2ndEd.).London.NICE,2008
187
NationalInstituteforHealthandClinicalExcellence.TheGuidelinesManual2009.London.
NationalInstituteforHealthandClinicalExcellence,2009Availablefrom:
http://www.nice.org.uk/aboutnice/howwework/developingniceclinicalguidelines/clinicalguidelinede
velopmentmethods/GuidelinesManual2009.jsp
188
NationalInstituteforHealthandClinicalExcellence.Ongoingrehabilitation.Qualitystandard
forstroke,London:NICE,2010:1516
189
2010
NationalInstituteforHealthandClinicalExcellence.Strokequalitystandard.London.NICE,
190
NilssonL,CarlssonJ,DanielssonA,FuglMeyerA,HellstromK,KristensenLetal.Walking
trainingofpatientswithhemiparesisatanearlystageafterstroke:acomparisonofwalkingtraining
onatreadmillwithbodyweightsupportandwalkingtrainingontheground.ClinicalRehabilitation.
2001;15(5):515527
191
NysGM,deHaanEH,KunnemanA,deKortPL,DijkermanHC.Acuteneglectrehabilitation
usingrepetitiveprismadaptation:arandomizedplacebocontrolledtrial.RestorativeNeurologyand
Neuroscience.2008;26(1):112
NationalClinicalGuidelineCentre,2013.
584
192
OfficeforNationalStatistics.InterimlifetablesforEnglandandWales20072009.UK.Office
forNationalStatistics,2009Availablefrom:
http://www.statistics.gov.uk/statbase/Product.asp?vlnk=14459
193
OlawaleO,JajaS,AnigboguC,AppiahKubiK,JonesOkaiD.Exercisetrainingimproves
walkingfunctioninanAfricangroupofstrokesurvivors:arandomizedcontrolledtrial.Clinical
Rehabilitation.2011;25(5):442450
194
OrganisationforEconomicCooperationandDevelopment.
195
OuelletteMM,LeBrasseurNK,BeanJF,PhillipsE,SteinJ,FronteraWRetal.Highintensity
resistancetrainingimprovesmusclestrength,selfreportedfunction,anddisabilityinlongterm
strokesurvivors.Stroke.2004;35(6):14041409
196
OzdemirF,BirtaneM,TabatabaeiR,KokinoS,EkukluG.Comparingstrokerehabilitation
outcomesbetweenacuteinpatientandnonintensehomesettings.ArchivesofPhysicalMedicineand
Rehabilitation.2001;82(10):13751379
197
PageSJ,LevineP,LeonardA,SzaflarskiJP,KisselaBM.Modifiedconstraintinducedtherapy
inchronicstroke:Resultsofasingleblindedrandomizedcontrolledtrial.PhysicalTherapy.2008;
88(3):333340
198
PalmerR,EnderbyP,CooperC,LatimerN,JuliousS,PatersonGetal.Computertherapy
comparedwithusualcareforpeoplewithlongstandingaphasiapoststroke:apilotrandomized
controlledtrial.Stroke.2012;43(7):19041911
199
ParkerCJ,GladmanJR,DrummondAE,DeweyME,LincolnNB,BarerDetal.Amulticentre
randomizedcontrolledtrialofleisuretherapyandconventionaloccupationaltherapyafterstroke.
TOTALStudyGroup.TrialofOccupationalTherapyandLeisure.ClinicalRehabilitation.2001;
15(1):4252
200
ParryRH.Communicationduringgoalsettinginphysiotherapytreatmentsessions.Clinical
Rehabilitation.2004;18(6):668682
201
PeacockPB,RileyCHB,LamptonTD,RafelSS,WalkerJS.In:StewartGT(eds),Trendsin
epidemiology:theBirminghamstroke,epidemiologyandrehabilitationstudy.,Springfield,Illinois:
Thomas,1972:231345
202
PerryJ,GarrettM,GronleyJK,MulroySJ.Classificationofwalkinghandicapinthestroke
population.Stroke.1995;26(6):982989
203
PeuralaSH,AiraksinenO,HuuskonenP,JakalaP,JuhakoskiM,SandellKetal.Effectsof
intensivetherapyusinggaittrainerorfloorwalkingexercisesearlyafterstroke.Journalof
RehabilitationMedicine.2009;41(3):166173
204
PeuralaSH,TarkkaIM,PitkanenK,SiveniusJ.Theeffectivenessofbodyweightsupported
gaittrainingandfloorwalkinginpatientswithchronicstroke.ArchivesofPhysicalMedicineand
Rehabilitation.2005;86(8):15571564
205
PhippsS,RichardsonP.Occupationaltherapyoutcomesforclientswithtraumaticbrain
injuryandstrokeusingtheCanadianOccupationalPerformanceMeasure.AmericanJournalof
OccupationalTherapy:OfficialPublicationoftheAmericanOccupationalTherapyAssociation.2007;
61(3):328334
206
PloughmanM,CorbettD.Canforcedusetherapybeclinicallyappliedafterstroke?An
exploratoryrandomizedcontrolledtrial.ArchivesofPhysicalMedicineandRehabilitation.2004;
85(9):14171423
NationalClinicalGuidelineCentre,2013.
585
207
PodsiadloD,RichardsonS.Thetimed"Up&Go":atestofbasicfunctionalmobilityforfrail
elderlypersons.JournaloftheAmericanGeriatricsSociety.1991;39(2):142148
208
PohlM,MehrholzJ,RitschelC,RuckriemS.Speeddependenttreadmilltrainingin
ambulatoryhemipareticstrokepatients:arandomizedcontrolledtrial.[seecomment].Stroke.2002;
33(2):553558
209
PohlM,WernerC,HolzgraefeM,KroczekG,MehrholzJ,WingendorfIetal.Repetitive
locomotortrainingandphysiotherapyimprovewalkingandbasicactivitiesofdailylivingafterstroke:
asingleblind,randomizedmulticentretrial(DEutscheGAngtrainerStudie,DEGAS).Clinical
Rehabilitation.2007;21(1):1727
210
PopovicMB,PopovicDB,SinkjaerT,StefanovicA,SchwirtlichL.Clinicalevaluationof
FunctionalElectricalTherapyinacutehemiplegicsubjects.JournalofRehabilitationResearch&
Development.2003;40(5):443453
211
PotempaK,LopezM,BraunLT,SzidonJP,FoggL,TincknellT.Physiologicaloutcomesof
aerobicexercisetraininginhemipareticstrokepatients.Stroke.1995;26(1):101105
212
PowellJ,CameronM,GranatM,PandyanD,StottDJ.Electricalstimulationatthewristin
poststrokehemiplegia.AgeandAgeing.1998;27(Suppl2):49
213
PowellJ,PandyanAD,GranatM,CameronM,StottDJ.Electricalstimulationofwrist
extensorsinpoststrokehemiplegia.Stroke.1999;30(7):13841389
214
RobertsPS,VegherJA,GilewskiM,BenderA,RiggsRV.Clientcenteredoccupationaltherapy
usingconstraintinducedtherapy.JournalofStrokeandCerebrovascularDiseases.2005;14:115121
215
RobertsonIH,GrayJM,PentlandB,WaiteLJ.Microcomputerbasedrehabilitationfor
unilateralleftvisualneglect:arandomizedcontrolledtrial.ArchivesofPhysicalMedicineand
Rehabilitation.1990;71(9):663668
216
RobertsonIH,McMillanTM,MacleodE,EdgeworthJ,BrockD.Rehabilitationbylimb
activationtrainingreducesleftsidedmotorimpairmentinunilateralneglectpatients:asingle
randomisedcontroltrial.NeuropsychologicalRehabilitation.2002;12(5):439454
217
RochonE,LairdL,BoseA,ScofieldJ.Mappingtherapyforsentenceproductionimpairments
innonfluentaphasia.NeuropsychologicalRehabilitation.2005;15(1):136
218
RodgersH,AtkinsonC,BondS,SuddesM,DobsonR,CurlessR.Randomizedcontrolledtrial
ofacomprehensivestrokeeducationprogramforpatientsandcaregivers.Stroke.1999;30(12):2585
2591
219
RodgersH,SoutterJ,KaiserW,PearsonP,DobsonR,SkilbeckCetal.Earlysupported
hospitaldischargefollowingacutestroke:pilotstudyresults.ClinicalRehabilitation.1997;11(4):280
287
220
RosewilliamS,MalhotraS,RoffeC,JonesP,PandyanAD.Cansurfaceneuromuscular
electricalstimulationofthewristandhandcombinedwithroutinetherapyfacilitaterecoveryofarm
functioninpatientswithstroke?ArchivesofPhysicalMedicineandRehabilitation.2012;
93(10):17151721
221
RosewilliamS,RoskellCA,PandyanAD.Asystematicreviewandsynthesisofthequantitative
andqualitativeevidencebehindpatientcentredgoalsettinginstrokerehabilitation.Clinical
Rehabilitation.2011;25(6):501514
222
RossiPW,KheyfetsS,RedingMJ.Fresnelprismsimprovevisualperceptioninstrokepatients
withhomonymoushemianopiaorunilateralvisualneglect.Neurology.1990;40(10):15971599
NationalClinicalGuidelineCentre,2013.
586
223
RoyalCollegeofSpeechandLanguageTherapists.RCSLTResourceManualfor
CommissioningandPlanningServiceforSLCN.RoyalCollegeofSpeechandLanguageTherapists,
2009Availablefrom:http://www.rcslt.org/speech_and_language_therapy/commissioning/dysphagia
224
RuddAG,WolfeCD,TillingK,BeechR.Randomisedcontrolledtrialtoevaluateearly
dischargeschemeforpatientswithstroke.[ErratumappearsinBMJ1998Feb7;316(7129):435].
BritishMedicalJournal.1997;315(7115):10391044
225
RyanT,EnderbyP,RigbyAS.Arandomizedcontrolledtrialtoevaluateintensityof
communitybasedrehabilitationprovisionfollowingstrokeorhipfractureinoldage.Clinical
Rehabilitation.2006;20(2):123131
226
SackleyC,WadeDT,MantD,AtkinsonJC,YudkinP,CardosoKetal.Clusterrandomizedpilot
controlledtrialofanoccupationaltherapyinterventionforresidentswithstrokeinUKcarehomes.
Stroke.2006;37(9):23362341
227
SahinN,UgurluH,AlbayrakI.Theefficacyofelectricalstimulationinreducingthepost
strokespasticity:arandomizedcontrolledstudy.DisabilityandRehabilitation.2012;34(2):151156
228
SakaO,SerraV,SamyshkinY,McGuireA,WolfeCC.Costeffectivenessofstrokeunitcare
followedbyearlysupporteddischarge.Stroke.2009;40(1):2429
229
SalbachNM,MayoNE,WoodDauphineeS,HanleyJA,RichardsCL,CoteR.Ataskorientated
interventionenhanceswalkingdistanceandspeedinthefirstyearpoststroke:arandomized
controlledtrial.ClinicalRehabilitation.2004;18(5):509519
230
ScarboroughP,PetoV,BhatnagarP,KaurA,LealJ,LuengoFernandexRetal.Stroke
Statistics.UK.BritishHeartFoundation&TheStrokeAssociation,2009
231
SchunemannHJ,GuyattGH.CommentarygoodbyeM(C)ID!HelloMID,wheredoyoucome
from?HealthServicesResearch.2005;40(2):593597
232
SchunemannHJ,PuhanM,GoldsteinR,JaeschkeR,GuyattGH.Measurementpropertiesand
interpretabilityoftheChronicrespiratorydiseasequestionnaire(CRQ).COPD:JournalofChronic
ObstructivePulmonaryDisease.2005;2(1):8189
233
SchwartzI,SajinA,FisherI,NeebM,ShochinaM,KatzLeurerMetal.Theeffectivenessof
locomotortherapyusingroboticassistedgaittraininginsubacutestrokepatients:arandomized
controlledtrial.PM&R:theJournalofInjury,FunctionandRehabilitation.2009;1(6):516523
234
ScottishHealthInnovationsLtd.Mirrortherapyusingprismglassesforphantomlimbpain,
strokerehabilitation,chronicpain.2011.Availablefrom:
http://www.prismglasses.co.uk/buy_prism_glasses/index.html#ecwid:category=0&mode=category&
offset=0&sort=normal[Lastaccessed:17May2011]
235
SellarsC,HughesT,LanghorneP.Speechandlanguagetherapyfordysarthriaduetonon
progressivebraindamage.CochraneDatabaseofSystematicReviews.2005;(3):CD002088
236
ShewanCM,KerteszA.Effectsofspeechandlanguagetreatmentonrecoveryfromaphasia.
BrainandLanguage.1984;23(2):272299
237
ShindoK,FujiwaraT,HaraJ,ObaH,HottaF,TsujiTetal.Effectivenessofhybridassistive
neuromusculardynamicstimulationtherapyinpatientswithsubacutestroke:arandomized
controlledpilottrial.NeurorehabilitationandNeuralRepair.2011;25(9):830837
238
SimsJ,GaleaM,TaylorN,DoddK,JespersenS,JoubertLetal.Regenerate:assessingthe
feasibilityofastrengthtrainingprogramtoenhancethephysicalandmentalhealthofchronicpost
strokepatientswithdepression.InternationalJournalofGeriatricPsychiatry.2009;24(1):7683
NationalClinicalGuidelineCentre,2013.
587
239
SiveniusJ,PyoralaK,HeinonenOP,SalonenJT,RiekkinenP.Thesignificanceofintensityof
rehabilitationofstrokeacontrolledtrial.Stroke.1985;16(6):928931
240
SmaniaN,AgliotiSM,GirardiF,TinazziM,FiaschiA,CosentinoAetal.Rehabilitationoflimb
apraxiaimprovesdailylifeactivitiesinpatientswithstroke.Neurology.2006;67(11):20502052
241
SmithDS,GoldenbergE,AshburnA,KinsellaG,SheikhK,BrennanPJetal.Remedialtherapy
afterstroke:arandomisedcontrolledtrial.BritishMedicalJournal.1981;282(6263):517520
242
SmithGC,EgbertN,DellmanJenkinsM,NannaK,PalmieriPA.Reducingdepressioninstroke
survivorsandtheirinformalcaregivers:Arandomizedclinicaltrialofawebbasedintervention.
RehabilitationPsychology.2012;57(3):196206
243
SmithJ,ForsterA,YoungJ.Arandomizedtrialtoevaluateaneducationprogrammefor
patientsandcarersafterstroke.ClinicalRehabilitation.2004;18(7):726736
244
SmithPS,ThompsonM.Treadmilltrainingpoststroke:arethereanysecondarybenefits?A
pilotstudy.ClinRehabil.2008;22(1011):9971002
245
SmithardDG,O'NeillPA,ParksC,MorrisJ.Complicationsandoutcomeafteracutestroke.
Doesdysphagiamatter?Stroke.1996;27(7):12001204
246
SpitzynaGA,WiseRJ,McDonaldSA,PlantGT,KiddD,CrewesHetal.Optokinetictherapy
improvestextreadinginpatientswithhemianopicalexia:acontrolledtrial.Neurology.2007;
68(22):19221930
247
StevensRS,AmblerNR,WarrenMD.Arandomizedcontrolledtrialofastrokerehabilitation
ward.AgeandAgeing.1984;13(2):6575
248
StrandT,AsplundK,ErikssonS,HaggE,LithnerF,WesterPO.Anonintensivestrokeunit
reducesfunctionaldisabilityandtheneedforlongtermhospitalization.Stroke.1985;16(1):2934
249
StraussH,ZieglerH.TheDelphiTechniqueanditsusesinsocialsciencesresearch.Journalof
CreativeBehaviour.1975;9(4):253259
250
StrokeAssociation.Factsandfiguresaboutstroke,2012
251
StrokeUnitTrialists'Collaboration.Organisedinpatient(strokeunit)careforstroke.
CochraneDatabaseofSystematicReviews.2007;(4):CD000197
252
SuddickKM,DeSL.Therapists'experiencesandperceptionsofteamworkinneurological
rehabilitation:reasoningbehindtheteamapproach,structureandcompositionoftheteamand
teamworkingprocesses.PhysiotherapyResearchInternational:theJournalforResearchersand
CliniciansinPhysicalTherapy.2006;11(2):7283
253
SvenssonA,HarmsenP,WihelmsenLandGoteborgOstraStrokeTrialUnit.Personal
communication:2012
254
TaubE,UswatteG,KingDK,MorrisD,CragoJE,ChatterjeeA.Aplacebocontrolledtrialof
constraintinducedmovementtherapyforupperextremityafterstroke.Stroke.2006;37(4):1045
1049
255
TengJ,MayoNE,LatimerE,HanleyJ,WoodDauphineeS,CoteRetal.Costsandcaregiver
consequencesofearlysupporteddischargeforstrokepatients.Stroke.2003;34(2):528536
256
ThomasJ,SutcliffeK,HardenA,OakleyA,OliverS,ReesRetal.Childrenandhealthyeating:
asystematicreviewofbarriersandfacilitators.London:UK.EPPICentre,SocialScienceResearch
Unit,InstituteofEducation,UniversityofLondon,2003
NationalClinicalGuidelineCentre,2013.
588
257
ThrasherTA,ZivanovicV,McIlroyW,PopovicMR.Rehabilitationofreachingandgrasping
functioninseverehemiplegicpatientsusingfunctionalelectricalstimulationtherapy.
NeurorehabilitationandNeuralRepair.2008;22(6):706714
258
TilsonJK,SullivanKJ,CenSY,RoseDK,KoradiaCH,AzenSPetal.Meaningfulgaitspeed
improvementduringthefirst60dayspoststroke:minimalclinicallyimportantdifference.Physical
Therapy.2010;90(2):196208
259
TimmermansAA,SeelenHA,WillmannRD,BakxW,deRB,LanfermannGetal.Armandhand
skills:trainingpreferencesafterstroke.DisabilityandRehabilitation.2009;31(16):13441352
260
TongRK,NgMF,LiLS.Effectivenessofgaittrainingusinganelectromechanicalgaittrainer,
withandwithoutfunctionalelectricstimulation,insubacutestroke:arandomizedcontrolledtrial.
ArchivesofPhysicalMedicineandRehabilitation.2006;87(10):12981304
261
TownendBS,SturmJW,PetsoglouC,O'LearyB,WhyteS,CrimminsD.Perimetric
homonymousvisualfieldlosspoststroke.JournalofClinicalNeuroscience.2007;14(8):754756
262
TrexlerLE,TrexlerLC,MalecJF,KlyceD,ParrottD.Prospectiverandomizedcontrolledtrialof
resourcefacilitationoncommunityparticipationandvocationaloutcomefollowingbraininjury.
JournalofHeadTraumaRehabilitation.2010;25(6):440446
263
TsangMH,SzeKH,FongKN.Occupationaltherapytreatmentwithrighthalffieldeye
patchingforpatientswithsubacutestrokeandunilateralneglect:arandomisedcontrolledtrial.
DisabilityandRehabilitation.2009;31(8):630637
264
TurtonAJ,O'LearyK,GabbJ,WoodwardR,GilchristID.Asingleblindedrandomised
controlledpilottrialofprismadaptationforimprovingselfcareinstrokepatientswithneglect.
NeuropsychologicalRehabilitation.2010;20(2):180196
265
TysonSF,RogersonL.Assistivewalkingdevicesinnonambulantpatientsundergoing
rehabilitationafterstroke:theeffectsonfunctionalmobility,walkingimpairments,andpatients'
opinion.ArchivesofPhysicalMedicineandRehabilitation.2009;90(3):475479
266
TysonSF,ThorntonHA.Theeffectofahingedanklefootorthosisonhemiplegicgait:
objectivemeasuresandusers'opinions.ClinicalRehabilitation.2001;15(1):5358
267
VanDePortIGL,WeversLEG,LindemanE,KwakkelG.Effectsofcircuittrainingasalternative
tousualphysiotherapyafterstroke:Randomisedcontrolledtrial.BMJ(ClinicalResearchEd).2012;
344(7859)
268
vanderGaagA,BrooksR.Economicaspectsofatherapyandsupportserviceforpeoplewith
longtermstrokeandaphasia.InternationalJournalofLanguageandCommunicationDisorders.
2008;43(3):233244
269
vanderLeeJH,BeckermanH,LankhorstGJ,BouterLM.TheresponsivenessoftheAction
ResearchArmtestandtheFuglMeyerAssessmentscaleinchronicstrokepatients.Journalof
RehabilitationMedicine.2001;33(3):110113
270
vanderLeeJH,WagenaarRC,LankhorstGJ,VogelaarTW,DevilleWL,BouterLM.Forceduse
oftheupperextremityinchronicstrokepatients:resultsfromasingleblindrandomizedclinicaltrial.
Stroke.1999;30(11):23692375
271
vanExelNJ,ScholteopReimerWJ,KoopmanschapMA.Assessmentofpoststrokequalityof
lifeincosteffectivenessstudies:theusefulnessoftheBarthelIndexandtheEuroQoL5D.Qualityof
LifeResearch.2004;13(2):427433
NationalClinicalGuidelineCentre,2013.
589
272
VisintinM,BarbeauH,KornerBitenskyN,MayoNE.Anewapproachtoretraingaitinstroke
patientsthroughbodyweightsupportandtreadmillstimulation.Stroke.1998;29(6):11221128
273
vonArbinM.,BrittonM,deFaireU,HelmersC,MiahK,MurrayV.Astudyofstrokepatients
treatedinanonintensivestrokeunitoringeneralmedicalwards.ActaMedicaScandinavica.1980;
208(12):8185
274
vonKochL,dePedroCuestaJ,KostulasV,AlmazanJ,HolmqvistLW.Randomizedcontrolled
trialofrehabilitationathomeafterstroke:oneyearfollowupofpatientoutcome,resourceuseand
cost.CerebrovascularDiseases.2001;12(2):131138
275
vonKochL,WidenHL,KostulasV,AlmazanJ,dePedroCuestaJ.Arandomizedcontrolled
trialofrehabilitationathomeafterstrokeinSouthwestStockholm:outcomeatsixmonths.
ScandinavianJournalofRehabilitationMedicine.2000;32(2):8086
276
WadeDT,deJongBA.Recentadvancesinrehabilitation.BMJ(ClinicalResearchEd).2000;
320(7246):13851388
277
WadeDT,HewerRL.Functionalabilitiesafterstroke:measurement,naturalhistoryand
prognosis.JNeurolNeurosurgPsychiatry.1987;50(2):177182
278
WalkerMF,GladmanJR,LincolnNB,SiemonsmaP,WhiteleyT.Occupationaltherapyfor
strokepatientsnotadmittedtohospital:arandomisedcontrolledtrial.Lancet.1999;354(9175):278
280
279
WalkerML,RinglebSI,MaihaferGC,WalkerR,CrouchJR,LunenBVetal.VirtualReality
EnhancedPartialBodyWeightSupportedTreadmillTrainingPoststroke:FeasibilityandEffectiveness
in6Subjects.ArchivesofPhysicalMedicineandRehabilitation.2010;91(1):115122
280
WardNS,CohenLG.Mechanismsunderlyingrecoveryofmotorfunctionafterstroke.
ArchivesofNeurology.2004;61(12):18441848
281
WernerC,vonFrankenbergS,TreigT,KonradM,HesseS.Treadmilltrainingwithpartial
bodyweightsupportandanelectromechanicalgaittrainerforrestorationofgaitinsubacutestroke
patients:arandomizedcrossoverstudy.Stroke.2002;33(12):28952901
282
WernerRA,KesslerS.Effectivenessofanintensiveoutpatientrehabilitationprogramfor
postacutestrokepatients.AmericanJournalofPhysicalMedicineandRehabilitation.1996;
75(2):114120
283
WertzRT,WeissDG,AtenJL,BrookshireRH,GarciaBunuelL,HollandALetal.Comparisonof
clinic,home,anddeferredlanguagetreatmentforaphasia.AVeteransAdministrationCooperative
Study.ArchivesofNeurology.1986;43(7):653658
284
WesterbergH,JacobaeusH,HirvikoskiT,ClevbergerP,OstenssonML,BartfaiAetal.
Computerizedworkingmemorytrainingafterstrokeapilotstudy.BrainInjury.2007;21(1):2129
285
WinsteinCJ,RoseDK,TanSM,LewthwaiteR,ChuiHC,AzenSP.Arandomizedcontrolled
comparisonofupperextremityrehabilitationstrategiesinacutestroke:Apilotstudyofimmediate
andlongtermoutcomes.ArchivesofPhysicalMedicineandRehabilitation.2004;85(4):620628
286
WolfSL,WinsteinCJ,MillerJP,TaubE,UswatteG,MorrisDetal.Effectofconstraintinduced
movementtherapyonupperextremityfunction3to9monthsafterstroke:theEXCITErandomized
clinicaltrial.JournaloftheAmericanMedicalAssociation.2006;296(17):20952104
287
WorrallL,SherrattS,RogersP,HoweT,HershD,FergusonAetal.Whatpeoplewithaphasia
want:TheirgoalsaccordingtotheICF.Aphasiology.2011;25(3):309322
NationalClinicalGuidelineCentre,2013.
590
288
WorrallL,YiuE.Effectivenessoffunctionalcommunicationtherapybyvolunteersforpeople
withaphasiafollowingstroke.Aphasiology.2000;14(9):911924
289
WressleE,EegOlofssonAM,MarcussonJ,HenrikssonC.Improvedclientparticipationinthe
rehabilitationprocessusingaclientcentredgoalformulationstructure.JournalofRehabilitation
Medicine:OfficialJournaloftheUEMSEuropeanBoardofPhysicalandRehabilitationMedicine.
2002;34(1):511
290
WressleE,ObergB,HenrikssonC.Therehabilitationprocessforthegeriatricstrokepatient
anexploratorystudyofgoalsettingandinterventions.DisabilityandRehabilitation.1999;21(2):80
87
291
WuCY,ChenCL,TangSF,LinKC,HuangYY.Kinematicandclinicalanalysesofupper
extremitymovementsafterconstraintinducedmovementtherapyinpatientswithstroke:a
randomizedcontrolledtrial.ArchivesofPhysicalMedicineandRehabilitation.2007;88(8):964970
292
WuCY,ChenCL,TsaiWC,LinKC,ChouSH.Arandomizedcontrolledtrialofmodified
constraintinducedmovementtherapyforelderlystrokesurvivors:changesinmotorimpairment,
dailyfunctioning,andqualityoflife.ArchivesofPhysicalMedicineandRehabilitation.2007;
88(3):273278
293
WuCY,LinKC,ChenHC,ChenIH,HongWH.Effectsofmodifiedconstraintinduced
movementtherapyonmovementkinematicsanddailyfunctioninpatientswithstroke:akinematic
studyofmotorcontrolmechanisms.NeurorehabilitationandNeuralRepair.2007;21(5):460466
294
WuC,ChuangL,LinK,ChenH,TsayP.Randomizedtrialofdistributedconstraintinduced
therapyversusbilateralarmtrainingfortherehabilitationofupperlimbmotorcontrolandfunction
afterstroke.NeurorehabilitationandNeuralRepair.2011;25(2):130139
295
YoungCA,ManmathanGP,WardJC.Perceptionsofgoalsettinginaneurological
rehabilitationunit:aqualitativestudyofpatients,carersandstaff.JournalofRehabilitationMedicine
:OfficialJournaloftheUEMSEuropeanBoardofPhysicalandRehabilitationMedicine.2008;
40(3):190194
NationalClinicalGuidelineCentre,2013.
591