CLINICAL GERONTOLOGIST

2022, VOL. 45, NO. 4, 777–797

https://doi.org/10.1080/07317115.2019.1694117

A Systematic Review and Meta-Analysis of Dyadic Psychological Interventions

for BPSD, Quality of Life and/or Caregiver Burden in Dementia or MCI
Emma Poon, MSc
Clinical Education Development and Research (CEDAR) Group, Psychology: College of Life and Environmental Sciences, University of Exeter,
Exeter, UK

ABSTRACT KEYWORDS
Objectives: This systematic review and meta-analysis assesses the effectiveness of psychological Dementia; Alzheimer’s
interventions that involve people with dementia or mild cognitive impairment (MCI) and their disease; mild cognitive
informal caregivers, and target improvements in the management of the behavioral and psycho- impairment; informal
logical symptoms of dementia (BPSD); quality of life; and/or burden reduction for people with caregivers; quality of life;
psychological interventions;
either dementia or MCI and their informal caregivers. dyadic interventions;
Methods: Studies were identified through database searches (Cochrane Library, CENTRAL, systematic review
CINAHL, EMBASE, MEDLINE and PsychINFO) and clinical trials registers (ClinicalTrials.gov and
http://apps.who.int/trialsearch/). Data were pooled for meta-analysis.
Results: Database and reference list searches identified 1,878 references, of which fourteen
studies were included. Positive effects were found on the anxiety symptoms of people with
dementia on the RAID scale; on the quality of life of people with dementia on the self-rated QoL-
AD scale; and on informal caregiver burden on the Zarit Burden Interview.
Conclusions: Psychological interventions involving whole dyads have some promise for both
people with dementia and informal caregivers, but are still far from uniformly effective across
BPSD, quality of life, and caregiver burden. Further research directions are discussed.
Clinical Implications: The results suggest that clinicians should routinely involve both halves of
the dyad when delivering psychological interventions targeting anxiety or quality of life for
people with dementia, or burden for informal caregivers.

Introduction disease is the most common form of dementia

(Alzheimer’s Association, 2017).
Dementia is an increasingly important area for
Broadly, the behavioral and psychological
research, with the number of people living with
symptoms of dementia (BPSD) comprise disturbed
the disorder worldwide estimated to increase
behavior, mood, thought or perception. They are
from 47 million in 2015 to 132 million in 2050
relatively common, with previous estimates of the
(Prince et al., 2015). The World Health
prevalence of BPSD ranging from 61% to 88%
Organisation (2016) estimates that there are
among elderly populations in community settings
nearly 10 million new cases of dementia
(Lyketsos et al., 2000; Mega, Cummings, Fiorello,
every year.
& Gornbein, 1996).
Dementia is diagnosed when cognitive or beha-
There are currently no disease-modifying thera-
vioral symptoms interfere with an individual’s
pies available for dementia. While pharmacological
ability to function at work or at usual activities.
treatments can temporarily improve behavioral
This must represent a decline from previous levels
and cognitive symptoms by targeting neurotrans-
of functioning and performing; must not be
mitters in the brain to help maintain short-term
explained by delirium or a major psychiatric dis-
daily functioning (National Institute on Aging,
order; and could involve personality change or
2016), they do not reverse the effects of dementia.
memory, visuospatial, reasoning, or language
The identification and development of effective
impairments (McKhann et al., 2011). Alzheimer’s
psychological interventions for dementia,

CONTACT Emma Poon [email protected] Clinical Education Development and Research (CEDAR) Group, Psychology: College of Life and
Environmental Sciences, University of Exeter, Exeter, UK
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/wcli.
© 2019 Taylor & Francis Group, LLC
778 E. POON

therefore, remains extremely important, and in would enable interventions to build on the satisfaction
fact, studies have shown that interventions provid- caregivers can feel with providing care, which includes
ing psychological and psychosocial support can “the sense of personal accomplishment and gratifica-
both improve management of BPSD for people tion, feelings of mutuality in a dyadic relationship, an
with dementia and reduce the high burden on increase of family cohesion and functionality, and
caregivers (Chiatti et al., 2013; Testad et al., 2014). a sense of personal growth and purpose in life” (Yu,
This high caregiver burden is seen in research Cheng, & Wang, 2018). The current paper begins to
showing dementia caregiving to be associated with address this deficit.
negative mental health consequences, including This systematic review and meta-analysis assesses
depression and anxiety (Tremont, 2011). the evidence for the effectiveness of psychological
Moreover, the impact on informal caregivers, interventions for dementia that involve people with
such as relatives and friends, is particularly acute. dementia or mild cognitive impairment (MCI) and
Research shows that BPSD both place a heavy their informal caregivers and that target improve-
burden on family caregivers (Machnicki, Allegri, ments in the management of BPSD, quality of life
Dillon, Serrano, & Taragano, 2009; Pinquart & and/or burden reduction for people with either
Sörensen, 2003) and predict strongly caregiver dementia or MCI and their informal caregivers.
decisions to institutionalize people with dementia
(Chan, Kasper, Black, & Rabins, 2003; de Vugt
et al., 2005). The importance of informal caregiv- Methods
ing is further illustrated by the fact that it accounts
Eligibility criteria
for the single biggest dementia-related cost in
high-income countries, ahead of both formal social To be eligible for inclusion, studies had to:
care (40%) and direct medical costs (15%; World
Health Organization [WHO], 2012). (1) be published in English between
The importance of psychological interventions January 1997 and March 2018;
for dementia and the burden faced by, in particu- (2) involve at least one psychosocial interven-
lar, informal caregivers, forms the context for the tion for dementia;
current review. Systematic reviews have shown (3) have at least one control group;
psychological and psychosocial interventions to (4) include as participants adults diagnosed
be effective when delivered to people with demen- with dementia or Alzheimer’s disease, in
tia and when delivered to caregivers (Abraha et al., line with the DSM-IV, ICD-10 or compar-
2017; Vandepitte et al., 2016), but such interven- able, or adults diagnosed with mild cogni-
tions have often treated people with dementia and tive impairment (MCI), and their informal
caregivers separately (Judge, Yarry, Looman, & caregivers; and
Bass, 2013). This reveals a potential weakness in (5) measure at least one of the following:
the current evidence, as it is not clear whether a. one or more behavioral and psychologi-
many psychological interventions involving both cal symptoms of dementia (BPSD),
people with dementia and informal caregivers defined as disturbed behavior, mood,
exist and how many of these interventions are thought or perception;
effective for either or both halves of the dyad. b. the quality of life (QoL) of the adult
Psychological interventions for dementia that participants diagnosed with dementia,
involve both halves of the dyad potentially possess Alzheimer’s disease or MCI;
several benefits, including: home delivery and better c. caregiver burden or distress attributed to
integration with existing routines in a more supportive managing BPSD; or
atmosphere; delivery by a longer-term caregiver with d. caregiver QoL attributed to managing
greater consistency in applying the intervention; and BPSD.
greater caregiver acceptance if the intervention is per-
ceived to meet whole dyad needs (Judge et al., 2013). Psychosocial interventions could include, for exam-
This latter benefit could be particularly important as it ple, psychoeducational and cognitive behavioral
 CLINICAL GERONTOLOGIST 779

approaches, counseling, and family-based therapies, further assessment against the inclusion and exclu-
while control groups could include different types of sion criteria. Both reviewers screened the full-text
comparator, such as treatment as usual, waiting list, studies and five disagreements (2.4%) were dis-
attention placebo and no treatment. Studies could be cussed until consensus was reached.
randomized or non-randomized.
Excluded from this review were studies that
primarily focused on non-psychological interven- Data collection process
tions, defined as interventions involving: medica-
tion; music therapy; art and drama therapy; Data were extracted by the first reviewer using
reminiscence therapy; and/or exercise. Studies a data extraction form adapted from the Cochrane
with both psychological and non-psychological Handbook for Systematic Reviews of Interventions
interventions were included if the interventions (Higgins & Deeks, 2008). The data extraction form
were applied to separate groups of participants. was piloted by the first reviewer using a random
sample of included studies and adapted further to
summarize the research aims, study designs, parti-
Information sources cipants, interventions and outcomes. The second
reviewer checked half of the extracted data.
Eligible studies were identified through
Disagreements were discussed until consensus was
a comprehensive search of relevant databases,
reached. Consensus was reached in all cases.
comprising: the Cochrane Dementia and
The characteristics of each included study are
Cognitive Improvement Group’s Specialized
summarized in Table 1.
Register; the CENTRAL, CINAHL, EMBASE,
MEDLINE and PsychINFO databases; and clinical
trials registers (ClinicalTrials.gov and http://apps.
who.int/trialsearch/). Risk of bias in individual studies
Following the initial database searches, reference
Risk of bias was evaluated independently by both
lists from systematic reviews and meta-analyses
reviewers using guidelines from the Cochrane
related to BPSD and MCI were also scanned.
Collaboration (Higgins & Green, 2011; see Table 1).
The risk of bias assessment encompasses:
Search
a. random sequence generation;
The following search terms were input into all the b. allocation concealment;
above listed databases: c. blinding of outcome assessment;
(“dementia” OR “Alzheimer” OR “MCI”) AND d. incomplete outcome data; and
(“caregiver” OR “carer”) AND (“treatment” OR e. selective reporting.
“therapy” OR “intervention” OR “support” OR
“support group” OR “psycho-social” OR “psychoso-
Disagreements between the reviewers were
cial” OR “psychological” OR “psychotherapeutic)
AND (“behavioural” OR “psychological” OR “neu- resolved by discussion until consensus was achieved.
ropsychiatric” OR “anxiety” OR “aggression” OR A score between zero and two was given to each of
“agitation” OR “apathy” OR “depression” OR “dis- the five areas of risk of bias. A score of zero indicated
turbing behaviour” OR “sleep” OR “hallucination” a high risk of bias; a score of one an unclear risk of
OR “euphoria” OR “delusion” OR “disinhibition”). bias; and a score of two a low risk of bias. This
provided a total score ranging from zero to 10, with
studies considered of low (0–3), moderate (4–6) and
Study selection
high (7–10) methodological quality.
EP and a second reviewer independently screened To increase the validity of the review, only stu-
the list of references to identify studies that met dies with a score of four or above – i.e. deemed as
the eligibility criteria, based on titles and abstracts. being of moderate or high methodological quality –
Full papers for studies were then obtained for were included in the final analysis.
 780

Table 1. Overview data for studies included for qualitative analysis.

Relevant Outcome
Study Sample (Dyads) Method Intervention Patient and Caregiver Roles Measures Risk of Bias
Amieva et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
al., 2016 n = 655 RCT Caregiver education and skills-training/group 78.7/N/A NPI, QoL-AD, (b) Unclear
E. POON

cognitive stimulation/individual cognitive MADRS, AI (c) Low

stimulation (d) Low
(e) Low
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 9
1. AD diagnosis, according to NINCDS- Typical Cognition 39/N/A ZBI
ADRDA criteria care Duration/frequency Caregiver relationship
24-month duration Relative
Months 1-3: Caregiver participation
Weekly sessions Supported cognitive rehabilitation
90 minutes per session
Months 4-24:
6-weekly sessions
Session length N/A
Gitlin et al., Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
2010 n = 272 RCT Caregiver education and skills-training 82.1/66.3 RMBPC (b) Low
(c) Low
(d) Unclear
(e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 8
1. Caregivers living with individuals with a No Neuropsychiatric behaviours/caregiver distress 82.4/62.2 ZBI 12-item/
physician diagnosis of dementia or MMSE treatment Duration/frequency Caregiver relationship Perceived Change
score > 24 24-week duration N/A – were living with patient Index
2. Caregivers aged ≥ 21 years Weeks 1-16: Caregiver participation
3. Not actively seeking nursing home Up to 11 sessions Reviewed problem behaviours with
placement Session length N/A occupational therapists; instructed in
4. Managing problem behaviours Weeks 16-24: stress reduction and self-care techniques
5. Reporting upset (> 5 on a 10-point scale) 8-week duration
3 brief telephone calls
Joosten- Size Design Type Mean age (patients/caregivers) Patient outcomes (a) High
Weyn n = 93 NRCT Group CBT/patient and caregiver education 70.1/N/A RAND-36 Health (b) High
Banningh Survey/GDS-15 (c) Unclear
et al., (d) Unclear
2010 (e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 6
1. MCI diagnosis Waitlist Memory problems and psychosocial 52/N/A N/A
2. Aged > 50 years consequences Caregiver relationship
3. Partner, spouse, relative, or close friend Duration/frequency Relative or friend
willing to participate 10-week duration Caregiver participation
Weekly sessions Received training and promoted mutual
2 hours per session support
(Continued )
Table 1. (Continued).
Relevant Outcome
Study Sample (Dyads) Method Intervention Patient and Caregiver Roles Measures Risk of Bias
Judge et al., Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Unclear
2013 n = 118 RCT Patient and caregiver education and skills- 77.2/65.4 N/A (b) Unclear
training/CBT (c) Low
(d) High
(e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 5
1. Diagnosis of dementia or memory loss No Managing and coping with dementia-related 44/65 QoL-AD
2. MMSE scores ≥ 7 treatment issues Caregiver relationship
3. Assisted by a primary family caregiver Duration Relative
11-week duration Caregiver participation
6 sessions Helped patient practise and model skills
90 minutes per session
Logsdon et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) High
al., 2010 n = 142 RCT Structured patient and caregiver peer support 74.9/70.5 RMBPC/QoL-AD/ (b) High
group/Patient and caregiver education Medical Outcome (c) Unclear
Study Short Form (d) Unclear
SF-36/GDS-15 (e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 6
1. Diagnosis of dementia Waitlist Daily living skills 51/32 N/A
2. MMSE score ≥ 18 Duration/frequency Caregiver relationship
3. Aware of memory loss and able to 9-week duration Relative
communicate Weekly sessions Caregiver participation
4. Able to participate independently in a 90 minutes per session (average) Received training and supported
group setting patients
McCurry et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
al., 2005 n = 36 RCT Behaviour therapy/caregiver education and 77.8/N/A RMBPC/CSDD (b) Unclear
skills-training (c) Low
(d) Low
(e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 8
1. Diagnosis of AD Contact Insomnia 56/N/A N/A
2. ≥ 2 sleep problems on NPI behaviour care Duration/frequency Caregiver relationship
scale occurring ≥ 3 times per week 2-month duration Relative
3. Ambulatory, with no existing diagnosis 6 sessions Caregiver participation
of a primary sleep disorder 1 hour per session Involved in NITE-AD intervention and
trained to develop sleep treatment plans
Orrell et al., Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
2017 n = 356 RCT Caregiver education and skills-training/ N/A/N/A NPI, QoL-AD, (b) Low
caregiver-led individual cognitive stimulation DEMQOL, GDS-15 (c) Low
therapy (d) High
CLINICAL GERONTOLOGIST

(e) Low
(Continued )
781
Table 1. (Continued).
782

Relevant Outcome
Study Sample (Dyads) Method Intervention Patient and Caregiver Roles Measures Risk of Bias
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 8
1. DSM-IV criteria for dementia Typical Cognition/QoL 54/27 SF-12, EQ-5D
E. POON

2. Mini Mental State Examination score ≥ care Duration/frequency Caregiver relationship

10 25-week duration Relative or friend
3. Some ability to communicate and Up to 3 sessions per week Caregiver participation
understand 30 minutes per session Delivered iCST
4. Have caregiver (relative or friend) able to
deliver intervention and act as an
informant for assessments
Phung et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
al., 2013 n = 330 RCT Psychosocial counselling and support 76.5/66.5 NPI-Q/EQ-VAS/QoL- (b) Low
programme AD/CSDD (c) Low
(d) High
(e) Low
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 8
1. Home-living patients diagnosed with AD No Patient and caregiver QoL through prevention 47/33 EQ-VAS
≤ 12 months ago or mixed AD with treatment of depressive symptoms Caregiver relationship
vascular component or Lewy body Duration/frequency N/A – responsible for ‘informal care’
dementia 8-12-month duration Caregiver participation
2. Meet DSM-IV criteria for dementia, 7 session Participated in counselling sessions
NINCDS-ADRDA criteria for probable AD or Session length N/A
McKeith criteria for Lewy body dementia 5-8 telephone calls
3. Aged ≥ 50 years Call length N/A
4. MMSE score ≥ 20
5. Had caregiver for informal care with
minimum weekly contact
Quayhagen Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Unclear
et al., n = 103 RCT Multi-component – counselling/support 74.5/71.8 MBPC, Part A (b) Unclear
2000 groups/caregiver skills-training/caregiver-led (c) Low
individual cognitive stimulation (d) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes (e) Unclear
1. Diagnosis of mild to moderate AD Waitlist Behavioural symptoms of dementia 63/37 MBPC, Part B Total score: 6
2. Mattis Dementia Rating Scale > 100 Duration/frequency Caregiver relationship
8-weeks duration Spouse
Weekly sessions Caregiver participation
3 of 4 programmes 1.5 hours per session Involved in all interventions; assessed on
4th programme 4 hours per session (caregivers marital interaction, emotional status,
limited to 2 sessions) physical health, stress, coping, and social
support
Quinn et al., Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
2016 n = 24 RCT pilot Structured patient and caregiver peer support 75.2/67.0 EQ-5D-3L, ICECAP-O, (b) Unclear
study group/patient and caregiver education and HADS (c) Low
skills-training (d) Low
(e) Low
(Continued )
Table 1. (Continued).
Relevant Outcome
Study Sample (Dyads) Method Intervention Patient and Caregiver Roles Measures Risk of Bias
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 9
1. ICD-10 diagnosis of AD, vascular Typical Self-efficacy in managing dementia 75/21 N/A
dementia, or mixed AD and vascular care Duration/frequency Caregiver relationship
dementia 8-week duration Relative or friend
2. In early-stages of dementia, determined Weekly sessions Caregiver participation
by an MMSE score ≥ 20 90 minutes per session Supported patient participation and
3. Taking a stable dose of cholinesterase contributed perceptions of group
inhibitors or Memantine or not taking sessions
either medication, with no plan to change
during study
4. Have caregiver willing to participate
Samus et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
al., 2014 n = 303 RCT pilot Multi-component – patient and caregiver 83.9/66.7 NPI-Q/QoL-AD/ (b) Unclear
study education and skills-training/individualized ADRQL-40/CSDD (c) High
care planning and monitoring/referral and (d) High
linkage to services (e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 4
1. Aged ≥ 70 Typical Unmet dementia-related care needs 34/25 N/A
2. Have a reliable study partner care Duration/frequency Caregiver relationship
3. Dementia or COG DO NOS consistent 18-month duration Informal caregiver
with DSM-IV-TR criteria 2 sessions per month (average) Caregiver participation
4. > 1 unmet care need on JHDCNA Participated in study visits
Spector et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
al., 2015 n = 50 RCT pilot CBT 78.0/N/A NPI, QoL-AD, HADS, (b) Low
study RAID, CSDD (c) Low
(d) High
(e) Low
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 8
1. DSM-IV for mild to moderate dementia, Typical Anxiety 40/44 NPI
determined by a CDR score of 0.5, 1 or 2 care Duration/frequency Caregiver relationship
2. Score of ≥ 11 the RAID scale, with or 10-week duration Relative
without comorbid depression Up to 10 sessions Caregiver participation
3. Have caregiver willing to participate in Weekly sessions Supported patient in implementing
therapy 1 hour per session strategies
4. Willing to engage in therapy involving
discussion of thoughts and feelings
Stanley et Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Unclear
al., 2013 n = 32 RCT pilot Group CBT 78.6/63 QoL-AD/NPI-A/RAID/ (b) Unclear
study GAI/GDS (c) Unclear
(d) Low
CLINICAL GERONTOLOGIST

(e) Unclear
(Continued )
783
 784
E. POON

Table 1. (Continued).
Relevant Outcome
Study Sample (Dyads) Method Intervention Patient and Caregiver Roles Measures Risk of Bias
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 6
1. Diagnosis of dementia Typical Anxiety 41/13 NPI-A – caregiver
2. NPI-A score ≥ 4 care Duration/frequency Caregiver relationship distress
3. CDR score from 0.5 to 2.0 6-month duration Friend or family member who provided
4. Had a friend or family member who Months 1-3: ≥ 8 hours care per week
provided ≥ 8 or more hours care per week 12 sessions Caregiver participation
Weekly sessions Participated as coach
Session length N/A
Months 4-6:
≤ 8 brief telephone calls
Teri et al., Size Design Type Mean age (patients/caregivers) Patient outcomes (a) Low
1997 n = 72 RCT Behaviour therapy/caregiver skills-training 76.4/66.9 Hamilton Depression (b) Unclear
Rating Scale/CSDD/ (c) Low
BDI (d) Unclear
(e) Unclear
Inclusion criteria Control Target Male % (patients/caregivers) Caregiver outcomes Total score: 7
1. Probably AD on NINCDS-ADRDA Typical Depression/behavioural disturbances 53/31 ZBI
2. 6-month+ history of cognitive problems care and Duration/frequency Caregiver relationship
3. Live with caregivers waitlist 9-week duration N/A – were living with patient
4. RDC and DSM-III-R Weekly sessions Caregiver participation
5. Hamilton Depression Rating Scale ≥ 10 60-minutes per session Received training in and delivered
Behavior Therapy-Pleasant Events
intervention
 CLINICAL GERONTOLOGIST 785

Figure 1. PRISMA flow diagram showing the inclusion and exclusion of studies.

Synthesis of results type of outcome measures. The statistical hetero-

Due to the diversity of interventions across stu- geneity of the included studies was assessed using
dies, a random-effects model was used to present the Chi square test (Deeks, Altman, & Bradburn,
overall estimated effects. Only data from studies 2001) on Review Manager 5.3 (2014).
with the most comparable outcome measures, as Heterogeneity was considered statistically signifi-
judged by the first reviewer and corroborated by cant at p < .10 and was quantified by using the I2
the second, were pooled for the meta-analyses, statistic. For the interpretation of heterogeneity, I2
with data pooled if they were deemed to measure values of 25%, 50% and 75%, respectively, were
the same underlying construct. All meta-analyses deemed to represent low, moderate, and high
were conducted on Review Manager 5.3 (2014). levels of inconsistency of findings (Higgins,
Standardized mean differences were computed Thompson, Deeks, & Altman, 2003).
based on differences in post-treatment continuous
outcome measurements for the intervention and
control groups, using Hedges’ g. Results
Narrative summary
Study selection
Assessment of heterogeneity
Literature searches in March 2018 identified 1,878
Study characteristics were evaluated based on the references. After eliminating duplicates, 1,859
included population, type of interventions and remained, from which 1,650 were excluded after
786 E. POON

Intervention Category Number of Studies

Caregiver education or skills-training with behavior therapy Two

Caregiver education or skills-training with cognitive stimulation therapy Two

Caregiver education and skills-training (on its own) One

CBT (on its own) Two

Dyad education or skills-training with CBT Two

Dyad education or skills-training with dyad peer support groups Two

Multi-component interventions (i.e. three or more) Two

Psychosocial counseling and support One

Figure 2. Categories of study interventions.

assessing the title and abstract. This left 209 stu- The mean age of the people with dementia or
dies for which the full text was obtained, and of MCI ranged from 70.1 years (Joosten-Weyn
these, 195 were excluded as they did not meet the Banningh et al., 2011) to 83.9 years (Samus et al.,
review criteria. Thus, 14 studies were eligible for 2014). Excepting one study for which only median
inclusion (see Figure 1) and are presented accord- values were reported (Spector et al., 2015), and two
ing to PRISMA guidelines (Moher, Liberati, for which values were not reported and supplemen-
Tetzlaff, & Altman, 2009). tary data were not available (Amieva et al., 2016;
Joosten-Weyn Banningh et al., 2011), the mean age
of caregivers ranged from 63.0 years (Stanley et al.,
Study characteristics
2013) to 71.8 years (Quayhagen et al., 2000). Within
A total of 2,586 dyads, each comprising one person
each study, the mean age of caregivers was lower
with dementia or mild cognitive impairment (MCI)
than the mean age of people with dementia or MCI.
and one informal caregiver, participated in the 14
The mean gender balance among people with
studies selected for analysis. The studies spanned five
dementia or MCI was 49.6% male to 50.4% female.
countries and had a mean number of dyads of 185 per
Again excepting two studies for which values were
study, with a range of 24 (Quinn et al., 2016) to 655
not reported and supplementary data were not
(Amieva et al., 2016).
available (Amieva et al., 2016; Joosten-Weyn
Banningh et al., 2011), the mean gender balance
Design and participants of caregivers was 27.6% male to 72.4% female.
Thirteen of the studies were randomized con-
trolled trials (RCTs) and one was a non- Interventions
randomized controlled trial (NRCT). Four were The 14 studies assessed relatively diverse types of
pilot studies (Quinn et al., 2016; Samus et al., intervention. From the descriptions provided in
2014; Spector et al., 2015; Stanley et al., 2013). each study, the interventions can be divided into
The 13 RCTs all involved individuals diagnosed eight broad categories (see Figure 2).
with dementia or Alzheimer’s disease (n = 2,493), Three studies had two or more intervention
while the one NRCT involved patients diagnosed groups (Amieva et al., 2016; Quayhagen et al.,
with MCI (n = 93). The caregivers were relatives 2000; Teri, Logsdon, Uomoto, & McCurry, 1997).
or friends of the person with dementia or MCI in One of these groups, which received reminiscence
ten studies (n = 1,609), generic informal caregivers therapy (in Amieva et al., 2016), was excluded
in two studies (n = 633) and individuals specified from analysis. Intervention duration across the 14
as living with the patient in two studies (n = 344). studies averaged 27 weeks and ranged from eight
All participants were living in the community. weeks (Quayhagen et al., 2000) to 104 weeks
 CLINICAL GERONTOLOGIST 787

(Amieva et al., 2016). The timing of follow ups using Alexopoulos, Abrams, Young, and
ranged from a single follow up at three months Shamoian (1988) Cornell Scale for Depression in
(Quayhagen et al., 2000), to four follow ups at Dementia (CSDD; n = 5); Yesavage, Brink, and
three, six, 12 and 36 months (Phung et al., 2013). Rose (2000) Geriatric Depression Scale (GDS) full
Caregivers were involved in all the interven- version (n = 1) or Sheikh and Yesavage (1986)
tions, but to varying degrees. Examples of lim- GDS-15 (n = 3); Zigmond and Snaith (1983)
ited caregiver involvement include caregivers Hospital Anxiety and Depression Scale (HADS;
briefly attending patient intervention sessions n = 2); Hamilton’s (1967) Hamilton Depression
and implementing everyday strategies (Spector Rating Scale (n = 1); Beck, Ward, Mendelson,
et al., 2015); caregivers supporting the participa- Mock, and Erbaugh (1961) Beck Depression
tion of people with dementia in the study and Inventory (BDI; n = 1); Montgomery and
offering their perceptions (Quinn et al., 2016); Åsberg (1979) Montgomery-Åsberg Depression
and caregivers supporting therapists to devise Rating Scale (MADRS; n = 1). Anxiety was mea-
cognitive rehabilitation strategies (Amieva et al., sured using Zigmond and Snaith (1983) HADS (n
2016). More extensive caregiver involvement in = 2); Shankar, Walker, Frost, and Orrell (1999)
other studies included caregivers receiving train- Rating Anxiety in Dementia Scale (RAID; n = 2);
ing and delivering interventions to people with Pachana et al.’s (2007) Geriatric Anxiety
dementia (Teri et al., 1997); caregivers partici- Inventory (GAI; n = 1); Cummings et al.’s
pating fully in education and skills-training (1994) Neuropsychiatric Inventory-Anxiety (NPI-
(Joosten-Weyn Banningh et al., 2011); and care- A) subscale (n = 1). Apathy was measured using
givers providing cognitive behavioral therapy Robert et al.’s (2002) Apathy Inventory (AI;
based coaching (Stanley et al., 2013). n = 1).
The quality of life of the person with dementia
Comparisons or MCI was assessed using Logsdon, Gibbons,
The types of control used in the studies were also McCurry, and Teri (2002) Quality of Life in
varied. Control groups received typical care con- Alzheimer’s Disease (QoL-AD) self-rated (n
trol (n = 6); waitlist control (n = 3); typical care = 2), both self- and caregiver-rated (n = 4) and
and waitlist control that involved two control unspecified (n = 1); Stewart, Hays, and Ware
groups (n = 1); contact care control (n = 1); and (1988) Medical Outcome Study short form SF-
no treatment (n = 3). 36 (n = 1); Van der Zee, Sanderman, Heyink,
and de Haes (1996) RAND-36 Health Survey
Outcomes (Dutch version; n = 1); the EQ-VAS visual ana-
All studies reported outcomes immediately after logue scale (EuroQol, 1990; n = , p. 1); Kasper,
the end of the intervention, with the exception Black, Shore, and Rabins (2009) Alzheimer’s
of Joosten-Weyn Banningh et al. (2011), who Disease Related Quality of Life-40 items
reported within two weeks; Judge et al. (2013), (ADRQL-40; n = 1); EQ-5D-3L (EuroQol, 1990;
who reported after an average of three and a half n = , p. 1); Coast et al.’s (2008) ICECAP-O (n
weeks; Spector et al. (2015), who reported within = 1); Smith, Murray, et al.’s (2005) Dementia-
five weeks; and Quinn et al. (2016), who Specific Quality of Life (DEMQOL; n = 1).
reported within one month of the intervention. Caregiver burden and distress was assessed
Behavioral and psychological symptoms of using the Zarit Burden Interview (ZBI) 22-item
dementia were measured using a variety of stan- version (Zarit, Todd, & Zarit, 1986; n = , p. 2)
dardized instruments: Neuropsychiatric Inventory and 12-item version (Bedard et al., 2001; n = ,
(NPI) Cummings et al.’s (1994) version (n = 3) or p. 1), NPI (n = 2); Cummings et al.’s (1994) NPI-
Kaufer et al.’s (2000) NPI-Q (n = 2); Teri et al.’s A subscale (n = 1); Zarit et al.’s (1985) MBPC, Part
(1992) Revised Memory and Behavior Problems B (n = 1).
Checklist (RMBPC; n = 3); Zarit, Orr, and Zarit Caregiver quality of life was assessed using
(1985) Memory and Behavior Problems Checklist EuroQoL Group’s (1990) EQ-5D-5L (n = 2); Gitlin,
(MBPC), Part A (n = 1). Depression was assessed Winter, Dennis, and Hauck (2006) Perceived
788 E. POON

Change Index (n = 1); Ware, Kosinski, and Keller RAID scale (two studies, 65 participants, SMD
(1996) Short Form-12 Health Survey (SF-12; n = 1); −0.66, 95% CI [−1.16, −0.15]), but had no effect
Logsdon et al.’s (2002) QoL-AD (n = 1). on depression (eight studies, 1,474 participants,
SMD −0.17, 95% CI [−0.37, 0.03]). There was
Risk of bias assessment moderate heterogeneity between studies in the lat-
The risk of bias was found to be low to moderate ter analysis (I2 = 65%).
(see Table 1). The methods of randomization and
allocation concealment, however, were often Quality of life of people with dementia or MCI
unclear. Few studies had published protocols or For the self-rated quality of life for the person with
were registered on clinical trial registries, so the dementia or MCI, data were pooled from six stu-
risk of selective reporting cannot be ruled out. dies (Logsdon et al., 2010; Orrell et al., 2017;
Most studies reported good levels of attrition Phung et al., 2013; Samus et al., 2014; Spector
(< 30%). Furthermore, nine studies used outcome et al., 2015; Stanley et al., 2013), and for caregiver-
assessors who were blinded to treatment alloca- rated patient quality of life, data were pooled from
tion, three studies were unclear, and one used three studies (Phung et al., 2013; Samus et al.,
unblinded outcome assessors. 2014; Spector et al., 2015) based on the QoL-AD
scale. Psychological interventions had a significant
effect on self-rated patient quality of life (six stu-
Meta-analysis
dies, 1046 participants, SMD 0.16, 95% CI [0.03,
Synthesis of results 0.28]; Figure 7) but had no effect on caregiver-
The following time point measurements were rated quality of life (three studies, 616 participants,
included in the meta-analyses (see Figure 3). SMD 0.08, 95% CI [−0.08, 0.24]; Figure 8) on the
For greater completeness, data for the study by QoL-AD measure.
Phung et al. (2013) were taken as reported in
Waldorff et al. (2012). Caregiver burden
For caregiver burden based on the Zarit Burden
Behavioral and psychological symptoms of Interview (ZBI), three studies (Amieva et al., 2016;
dementia Gitlin, Winter, Dennis, Hodgson, & Hauck, 2010;
The first meta-analysis on the effect of psycholo- Teri et al., 1997) were pooled for analysis. Results
gical interventions for the behavioral and psycho- favored psychological interventions in reducing
logical symptoms of dementia, as measured by the caregiver burden (three studies, 646 participants,
Neuropsychiatric Inventory (NPI; Figure 4), SMD −0.19; 95% CI [−0.35, −0.03]; Figure 9). This
pooled data from five studies (Amieva et al., analysis was strongly influenced by Amieva et al.
2016; Orrell et al., 2017; Phung et al., 2013; (2016), which was the largest study, although there
Samus et al., 2014; Spector et al., 2015). This was no heterogeneity between studies (I2 = 0%).
included 1,241 participants and showed that psy-
chological interventions had no effect on BPSD Caregiver quality of life
rated on the NPI (five studies, 1,241 participants, For caregiver self-rated quality of life, data were pooled
standardized mean difference (SMD) 0.02; 95% from two studies based on the EQ-5D-3L measures
confidence interval (CI) [−0.10, 0.15]). (Orrell et al., 2017; Phung et al., 2013). Psychological
Meta-analyses on anxiety (Figure 5) and depres- interventions had no effect on caregiver quality of life
sion (Figure 6) pooled data, respectively, from two (two studies, 546 participants, SMD −0.03; 95% CI
studies based on the RAID scale (Spector et al., [−0.25, 0.19]; Figure 10) on the EQ-5D instrument.
2015; Stanley et al., 2013) and eight studies
(Amieva et al., 2016; Logsdon et al., 2010; Orrell Heterogeneity
et al., 2017; Phung et al., 2013; Samus et al., 2014; No evidence of heterogeneity was detected in the
Spector et al., 2015; Stanley et al., 2013; Teri et al., pooled studies using the Chi2 test, with the exception
1997). Psychological interventions had of the analysis on depression, which had moderate
a significant effect on anxiety measured on the heterogeneity (65%). Analyses on BPSD and caregiver
 CLINICAL GERONTOLOGIST 789

Study Measurement Time point

Amieva et al. (2016) Caregiver burden 24 months

Person with Dementia: BPSD and quality of life 24 months

Gitlin et al. (2010) Caregiver burden 24 weeks

Judge et al. (2013) Caregiver quality of life Treatment end, av. 14.5 wks

Logsdon et al. (2010) Person with Dementia: BPSD and quality of life Nine weeks

Orrell et al. (2017) Caregiver quality of life 26 weeks

Person with Dementia: BPSD and quality of life 26 weeks

Phung et al. (2013) Caregiver quality of life 12 months

Person with Dementia: BPSD and quality of life 12 months

Samus et al. (2014) Person with Dementia: BPSD and quality of life 18 months

Spector et al. (2015) Person with Dementia: BPSD and quality of life 15 weeks

Stanley et al. (2013) Person with Dementia: BPSD and quality of life Six months

Teri et al. (1997) Caregiver burden Nine weeks

Person with Dementia: BPSD Nine weeks

Figure 3. Study time point measurements included in the meta-analyses.

Figure 4. Psychological treatment versus control. Outcome: 1.1 BPSD NPI.

Figure 5. Psychological treatment versus control. Outcome: 1.2 Anxiety RAID.

Figure 6. Psychological treatment versus control. Outcome: 1.3 Depression.

790 E. POON

Figure 7. Psychological treatment versus control. Outcome: 2.1 Quality of life QoL-AD (self-ratings).

Figure 8. Psychological treatment versus control. Outcome: 2.2 Quality of life QoL-AD (caregiver ratings).

Figure 9. Psychological treatment versus control. Outcome: 3.1 Caregiver burden ZBI.

Figure 10. Psychological treatment versus control. Outcome: 4.1 Caregiver quality of life EQ-5D-3L.

quality of life had low (14%) and moderate (41%) dementia (BPSD); caregiver burden; or quality
heterogeneity, respectively, but were not statistically of life for either or both halves of the dyad.
significant. The 14 studies collectively involved a total of
2,586 dyads. Thirteen studies involved indivi-
duals with a dementia or Alzheimer’s Disease
Discussion diagnosis (n = 2,493), whilst one study recruited
MCI diagnosed participants (n = 93). All parti-
Summary of evidence
cipants lived in the community and not in resi-
From the currently available published literature, dential homes.
this review found 14 studies that involved dyads, Data from nine studies with comparable out-
comprising people with dementia or mild cogni- come measures were pooled across seven meta-
tive impairment (MCI) and their informal care- analyses to assess the efficacy of the studies’
givers, and evaluated the efficacy of psychological interventions. The results were
psychological interventions in managing: the mixed, showing that psychological interventions
behavioral and psychological symptoms of have some promise for supporting people with
 CLINICAL GERONTOLOGIST 791

dementia or MCI and their informal caregivers, these interventions did not primarily target care-
but are still far from uniformly effective across giver burden, the findings suggest that involving
BPSD, quality of life, and caregiver burden. both people with dementia and informal care-
Across five studies (Amieva et al., 2016; Orrell givers may have indirect benefits over targeting
et al., 2017; Phung et al., 2013; Samus et al., 2014; only half of the dyad.
Spector et al., 2015), no effect was found from Findings from this review also tentatively
psychological interventions on BPSD as measured suggest that psychological interventions can be
by the Neuropsychiatric Inventory (NPI), while effective for improving the quality of life of
across eight studies (Amieva et al., 2016; Logsdon people with dementia. Results pooled from six
et al., 2010; Orrell et al., 2017; Phung et al., 2013; studies (Logsdon et al., 2010; Orrell et al., 2017;
Samus et al., 2014; Spector et al., 2015; Stanley Phung et al., 2013; Samus et al., 2014; Spector
et al., 2013; Teri et al., 1997), no effect was found et al., 2015; Stanley et al., 2013) showed that
on depression. More promisingly, however, there psychological interventions improve the quality
was an improvement in anxiety symptoms mea- of life of people with dementia as indicated on
sured on the RAID scale, based on data from two the self-rated QoL-AD scale. The interventions
studies (Spector et al., 2015; Stanley et al., 2013), tested in these studies encompassed: peer group
both of which involved cognitive behavioral ther- support sessions to learn from other newly
apy (CBT) based activity. The common CBT base diagnosed individuals (Logsdon et al., 2010);
of these studies and the strong, positive effect size CBT sessions with the informal caregiver in
found by the meta-analysis suggests that CBT a coaching role (Stanley et al., 2013); individual
based interventions can offer a productive route and dyad counseling and skills-training (Phung
for dyads involving people with dementia who also et al., 2013); skills-training with individualized
experience anxiety. care planning, care monitoring and linkage to
Similarly, across two studies (Orrell et al., services (Samus et al., 2014); CBT-based ther-
2017; Phung et al., 2013), meta-analysis found apy targeting anxiety (Spector et al., 2015); and
no intervention effect on self-rated caregiver home-based, caregiver-led cognitive stimulation
quality of life, but this contrasts with results therapy (Orrell et al., 2017). These diverse
showing that psychological interventions signif- interventions could suggest that dyads have
icantly reduced informal caregiver burden on a range of effective options for psychological
the Zarit Burden Interview in data pooled from interventions that improve the quality of life
three studies (Amieva et al., 2016; Gitlin et al., of people with dementia. This conclusion, how-
2010; Teri et al., 1997). The psychological ever, should be treated with particular caution
interventions of these three studies varied as, curiously, when data were pooled from three
from behavioral strategies and systematic pro- of the above studies (Phung et al., 2013; Samus
blem-solving techniques to standardized cogni- et al., 2014; Spector et al., 2015), the interven-
tive training and rehabilitation, but all focused tions had no effect on the quality of life of the
on both improving symptoms for the person person with dementia as rated by the caregiver,
with dementia and providing some form of based on the QoL-AD scale. It is possible that
caregiver education and skills training. differences in these findings can be explained
The findings of this meta-analysis reinforce by the different sets of factors that influence the
existing evidence showing that psychological perceptions of quality of life of people with
interventions can reduce informal caregiver bur- dementia and of caregivers, as indicated in pre-
den (Chiatti et al., 2013) and the diversity of vious studies (Black et al., 2012; Römhild et al.,
these interventions suggests that multiple options 2018). These include the construct of quality of
may be available to caregivers. Furthermore, as life (Smith, Lamping, et al., 2005); the quality of
792 E. POON

the caregiver-patient relationship; and caregiver Conclusion

distress and quality of life (Huang, Chang,
This systematic review and meta-analysis shows
Tang, Chiu, & Weng, 2008).
that psychological interventions have some pro-
mise for supporting people with dementia and
their informal caregivers, but are still far from
Limitations uniformly effective across BPSD, quality of life,
and caregiver burden. Findings go part of the
Among the limitations of this review is the inclu-
way to addressing a gap in the literature around
sion and analysis of only published studies. It is
psychosocial interventions for dementia involving
known that negative findings from clinical
both individuals with dementia or MCI and their
research can be withheld from publication
informal caregivers.
(Joober, Schmitz, Annable, & Boksa, 2012), and
the inclusion only of published studies means This is an important area of research, as inter-
publication bias cannot be ruled out. ventions that involve both people with dementia
The inconsistent depth of reporting across and informal caregivers possess several potential
the 14 included studies limited the ability of benefits. These include home delivery and better
this review to include a greater number of integration with existing routines in a more sup-
studies in each meta-analysis, where this may portive atmosphere; delivery by a longer-term
otherwise have been feasible. An example saw caregiver with greater consistency in applying the
one study (Logsdon et al., 2010) reporting self- intervention; and greater caregiver acceptance if
but not carer-rated quality of life for the per- the intervention is perceived to meet whole dyad
son with dementia, and another (Amieva et al., needs (Judge et al., 2013). The global cost of
2016) not providing details of the rater, leading dementia was estimated to be US$604 billion in
to their exclusion from the meta-analyses. 2010, and informal caregiving accounts for the
Similarly, limited information on the patient- single largest part of this cost in high-income
caregiver relationship across the studies makes countries (WHO, 2012).
firm conclusions difficult to draw. Precise car- The review found that psychological interven-
egiving roles, including in interventions, were tions can have a positive effect on the anxiety
unclear in most studies, so the review is unable symptoms of people with dementia as measured
to draw conclusions on how caregiver-led inter- on the RAID scale, and the quality of life of people
ventions differ from clinician-led interventions, with dementia as indicated on the self-rated QoL-
or whether the level of caregiver involvement AD scale; and can reduce informal caregiver bur-
has an impact on patient and/or caregiver out- den on the Zarit Burden Interview. Moreover, the
comes. Similarly, since only one eligible study types of intervention involved begin to suggest
included participants with MCI, the review was that CBT based interventions could be effective
unable to assess the effectiveness of dyadic psy- for reducing anxiety of people with dementia,
chological interventions for this population. and interventions involving education and skills
Moreover, the diversity of interventions and training could be effective for reducing caregiver
outcome measures in the 14 included studies burden. The review, however, found insufficient
meant that, even where reporting was compre- evidence to draw firm conclusions.
hensive and detailed, the review had to make More research is required to gain a better
necessarily imperfect comparisons between dif- understanding of dyadic caregiver-person with
ferent outcome measures and populations. The dementia relationships, and to identify compo-
comparisons that were made, however, were, in nents within psychological interventions that can
this author’s judgment, made between studies potentially combine to form multi-component
with sufficiently similar and robust outcome interventions that are clinically effective and
measures to lead to valid conclusions, albeit meet whole dyadic needs. Given the potential
ones which should still be treated with caution. shown by the psychological interventions
 CLINICAL GERONTOLOGIST 793

analyzed in this review, it may also be beneficial References

for clinicians to routinely involve both halves of
Abraha, I., Rimland, J., Trotta, F., Dell’Aquila, G., Cruz-Jentoft,
the dyad when delivering psychological interven- A., Petrovic, M., … Cherubini, A. (2017). Systematic review of
tions targeting anxiety or quality of life for peo- systematic reviews of non-pharmacological interventions to
ple with dementia, or burden for informal treat behavioural disturbances in older patients with dementia.
caregivers. Furthermore, the initial evidence The SENATOR-OnTop series. BMJ Open, 7(3), e012759.
this review provides, set against the known glo- doi:10.1136/bmjopen-2016-012759
Alexopoulos, G. S., Abrams, R. C., Young, R. C., &
bal cost of dementia and informal caregiving,
Shamoian, C. A. (1988). Cornell scale for depression in
suggests policy-makers may wish to explore the dementia. Biological Psychiatry, 23(3), 271–284. doi:10.1016/
potential long-term cost benefits of dyadic, 0006-3223(88)90038-8
multi-component interventions for dementia. Alzheimer’s Association. (2017). What is dementia? Retrieved
from https://www.alz.org/what-is-dementia.asp
Amieva, H., Robert, P. H., Grandoulier, A. S., Meillon, C., De
Clinical Implications Rotrou, J., Andrieu, S., … Joël, M. E. (2016). Group and
individual cognitive therapies in Alzheimer’s disease: The
● Clinicians should routinely involve both ETNA3 randomized trial. International Psychogeriatrics, 28
halves of the dyad when delivering psy- (5), 707–717. doi:10.1017/S1041610215001830
chological interventions targeting anxiety Beck, A., Ward, C., Mendelson, M., Mock, J., & Erbaugh, J.
or quality of life for people with dementia, (1961). An inventory for measuring depression. Archives of
or burden for informal caregivers. General Psychiatry, 4(6), 561–571. doi:10.1001/archpsyc.19
61.01710120031004
● Further research is required to gain
Bedard, M., Molloy, D. W., Squire, L., Dubois, S., Lever, J. A.,
a better understanding of dyadic care- & O’Donnell, M. (2001). The Zarit Burden interview:
giver-person with dementia relationships, A new short version and screening version. The
and to identify components within psy- Gerontologist, 41(5), 652–657. doi:10.1093/geront/41.5.652
chological interventions that can poten- Black, B. S., Johnston, D., Morrison, A., Rabins, P. V.,
tially combine to form multi-component Lyketsos, C. G., & Samus, Q. M. (2012). Quality of life of
community-residing persons with dementia based on
interventions that are clinically effective self-rated and caregiver-rated measures. Quality of Life
and meet whole dyadic needs Research, 21(8), 1379–1389. doi:10.1007/s11136-011-0044-z
● Given the known global cost of dementia Chan, D.-C., Kasper, J. D., Black, B. S., & Rabins, P. V.
and informal caregiving, the long-term (2003). Presence of behavioral and psychological symp-
cost benefits of dyadic, psychological toms predicts nursing home placement in
interventions for dementia should be community-dwelling elders with cognitive impairment in
univariate but not multivariate analysis. Journals of
explored, as these types of interventions Gerontology Series A: Biological Sciences and Medical
could be particularly useful in low Science, 58, 548–554. doi:10.1093/gerona/58.6.M548
resource settings. Chiatti, C., Masera, F., Rimland, J. M., Cherubini, A., Scarpino, O.,
Spazzafumo, L., & Lattanzio, F. (2013). The UP-TECH project,
an intervention to support caregivers of Alzheimer’s disease
patients in Italy: Study protocol for a randomized controlled
trial. Trials, 14, 155. doi:10.1186/1745-6215-14-155
Coast, J., Flynn, T. N., Natarajan, L., Sproston, K., Lewis, J.,
Louviere, J. J., & Peters, T. J. (2008). Valuing the ICECAP
Acknowledgments capability index for older people. Social Science & Medicine,
I would like to thank Henry Clarke for his input during the 67(5), 874–882. doi:10.1016/j.socscimed.2008.05.015
review, and Janet Smithson and Ian Frampton for their Cummings, J., Mega, M., Gray, K., Rosenberg-Thompson, S.,
insightful comments. Carusi, D. A., & Gornbein, J. (1994). The neuropsychiatric
inventory: Comprehensive assessment of psychopathology
in dementia. Neurology, 44, 2308–2314. doi:10.1212/
WNL.44.12.2308
Disclosure statement de Vugt, M., Stevens, F., Aalten, P., Lousberg, R., Jaspers, N.,
& Verhey, F. (2005). A prospective study of the effects of
No potential conflict of interest was reported by the author. behavioral symptoms on the institutionalization of patients
794 E. POON

with dementia. International Psychogeriatratics, 17, Kasper, J. D., Black, B. S., Shore, A. D., & Rabins, P. V. (2009).
577–589. doi:10.1017/S1041610205002292 Evaluation of the validity and reliability of the Alzheimer’s
Deeks, J., Altman, D., & Bradburn, M. (2001). Statistical Disease-Related Quality of Life (ADRQL) assessment instru-
methods for examining heterogeneity and combining ment. Alzheimer Disease and Associated Disorders, 23(3), 275.
results from several studies in meta-analysis. In M. Egger, doi:10.1097/WAD.0b013e31819b02bc
D. G. Altman, & M. J. Bradburn (Eds.), Systematic reviews Kaufer, D. I., Cummings, J. L., Ketchel, P., Smith, V.,
in health care: Meta-analysis in context (pp. 285–312). MacMillan, A., Shelley, T., … DeKosky, S. T. (2000).
London, UK: BMJ Publishing Group. Validation of the NPI-Q, a brief clinical form of the neuropsy-
EuroQol, G. (1990). EuroQol-a new facility for the measurement chiatric inventory. The Journal of Neuropsychiatry and Clinical
of health-related quality of life. Health Policy, 16(3), 199–208. Neurosciences, 12(2), 233–239. doi:10.1176/jnp.12.2.233
Gitlin, L. N., Winter, L., Dennis, M. P., & Hauck, W. W. Logsdon, R. G., Gibbons, L. E., McCurry, S. M., & Teri, L.
(2006). Assessing perceived change in the well-being of (2002). Assessing quality of life in older adults with cog-
family caregivers: Psychometric properties of the perceived nitive impairment. Psychosomatic Medicine, 64(3),
change index and response patterns. American Journal of 510–519. doi:10.1097/00006842-200205000-00016
Alzheimer’s Disease & Other Dementias, 21(5), 304–311. Logsdon, R. G., Pike, K. C., McCurry, S. M., Hunter, P.,
doi:10.1177/1533317506292283 Maher, J., Snyder, L., & Teri, L. (2010). Early-stage mem-
Gitlin, L. N., Winter, L., Dennis, M. P., Hodgson, N., & ory loss support groups: Outcomes from a randomized
Hauck, W. W. (2010). Targeting and managing behavioral controlled clinical trial. Journals of Gerontology Series B:
symptoms in individuals with dementia: A randomized Psychological Sciences and Social Sciences, 65(6), 691–697.
trial of a nonpharmacological intervention. Journal of the doi:10.1093/geronb/gbq054
American Geriatrics Society, 58(8), 1465–1474. Lyketsos, C., Steinberg, M., Tschanz, J., Norton, M.,
doi:10.1111/j.1532-5415.2010.02971.x Steffens, D., & Breitner, J. (2000). Mental and behavioral
Hamilton, M. A. X. (1967). Development of a rating scale for disturbances in dementia: Findings from the cache county
primary depressive illness. British Journal of Social and Clinical study on memory in aging. American Journal of Psychiatry,
Psychology, 6(4), 278–296. doi:10.1111/bjc.1967.6.issue-4 157, 708–714. doi:10.1176/appi.ajp.157.5.708
Higgins, J., & Deeks, J. (2008). Chapter 7: Selecting studies Machnicki, G., Allegri, R., Dillon, C., Serrano, C., & Taragano, F.
and collecting data. In J. P. T. Higgins & S. Green (Eds.), (2009). Cognitive, functional and behavioral factors associated
Cochrane handbook for systematic reviews of interventions with the burden of caring for geriatric patients with cognitive
version 5.0.1. London, UK: The Cochrane Collaboration. impairment or depression: Evidence from a South American
Retrieved from www.cochrane-handbook.org sample. International Journal of Geriatratric Psychiatry, 24,
Higgins, J., & Green, S. (2011). Cochrane handbook for sys- 382–389. doi:10.1002/gps.2133
tematic reviews of interventions version 5.1.0. London, UK: McCurry, S. M., Gibbons, L. E., Logsdon, R. G., Vitiello, M. V.,
The Cochrane Collaboration. Retrieved from http://www. & Teri, L. (2005). Nighttime insomnia treatment and educa-
cochrane-handbook.org tion for Alzheimer’s disease: A randomized, controlled trial.
Higgins, J., Thompson, S., Deeks, J., & Altman, D. (2003). Journal of the American Geriatrics Society, 53(5), 793–802.
Measuring inconsistency in meta-analyses. British Medical doi:10.1111/(ISSN)1532-5415
Journal, 327(7414), 557–560. doi:10.1136/bmj.327.7414.557 McKhann, G., Knopman, D., Chertkow, H., Hyman, B., Jack, C.,
Huang, H. L., Chang, M. Y., Tang, J. S. H., Chiu, Y. C., & Kawas, C., … Phelps, C. (2011). The diagnosis of dementia due
Weng, L. C. (2008). Determinants of the discrepancy in to Alzheimer’s disease: Recommendations from the National
patient- and caregiver-rated quality of life for persons with Institute on Aging-Alzheimer’s Association workgroups on
dementia. Journal of Clinical Nursing, 18, 3107–3117. diagnostic guidelines for Alzheimer’s disease. Alzheimer’s &
doi:10.1111/j.1365-2702.2008.02537.x Dementia: the Journal of the Alzheimer’s Association, 7,
Joober, R., Schmitz, N., Annable, L., & Boksa, P. (2012). 263–269. doi:10.1016/j.jalz.2011.03.005
Publication bias: What are the challenges and can they be Mega, M., Cummings, J., Fiorello, T., & Gornbein, J. (1996).
overcome? Journal of Psychiatry & Neuroscience: JPN, 37 The spectrum of behavioral changes in Alzheimer’s dis-
(3), 149–152. doi:10.1503/jpn.110175 ease. Neurology, 46, 130–135. doi:10.1212/WNL.46.1.130
Joosten-Weyn Banningh, L. W., Prins, J. B., Vernooij- Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009).
Dassen, M. J., Wijnen, H. H., Rikkert, M. G. O., & Preferred reporting items for systematic reviews and
Kessels, R. P. (2011). Group therapy for patients with meta-analyses: The PRISMA statement. Annals of
mild cognitive impairment and their significant others: Internal Medicine, 151(4), 264–269. doi:10.7326/0003-
Results of a waiting-list controlled trial. Gerontology, 57 4819-151-4-200908180-00135
(5), 444–454. doi:10.1159/000315933 Montgomery, S. A., & Åsberg, M. A. (1979). A new depression
Judge, K. S., Yarry, S. J., Looman, W. J., & Bass, D. M. (2013). scale designed to be sensitive to change. The British Journal
Improved strain and psychosocial outcomes for caregivers of Psychiatry, 134(4), 382–389. doi:10.1192/bjp.134.4.382
of individuals with dementia: Findings from project National Institute on Aging. (2016). Alzheimer’s disease fact
ANSWERS. The Gerontologist, 53(2), 280–292. sheet. Retrieved from https://www.nia.nih.gov/health/alz
doi:10.1093/geront/gns076 heimers-disease-fact-sheet
 CLINICAL GERONTOLOGIST 795

Orrell, M., Yates, L., Leung, P., Kang, S., Hoare, Z., independence at home (MIND) pilot randomized trial. The
Whitaker, C., … Pearson, S. (2017). The impact of indivi- American Journal of Geriatric Psychiatry, 22(4), 398–414.
dual Cognitive Stimulation Therapy (iCST) on cognition, doi:10.1016/j.jagp.2013.12.175
quality of life, caregiver health, and family relationships in Shankar, K. K., Walker, M., Frost, D., & Orrell, M. W. (1999).
dementia: A randomised controlled trial. PLoS Medicine, The development of a valid and reliable scale for rating
14(3), e1002269. doi:10.1371/journal.pmed.1002269 anxiety in dementia (RAID). Aging & Mental Health, 3(1),
Pachana, N. A., Byrne, G. J., Siddle, H., Koloski, N., Harley, E., 39–49. doi:10.1080/13607869956424
& Arnold, E. (2007). Development and validation of the Sheikh, J. I., & Yesavage, J. A. (1986). Geriatric Depression Scale
geriatric anxiety inventory. International Psychogeriatrics, (GDS): Recent evidence and development of a shorter
19(1), 103–114. doi:10.1017/S1041610206003504 version. Clinical Gerontologist: the Journal of Aging and
Phung, K. T., Waldorff, F. B., Buss, D. V., Eckermann, A., Mental Health, 5(1–2), 165–173. doi:10.1300/J018v05n01_09
Keiding, N., Rishøj, S., … Vogel, A. (2013). A three-year Smith, S. C., Lamping, D. L., Banerjee, S., Harwood, R., Foley, B.,
follow-up on the efficacy of psychosocial interventions for Smith, P., … Mann, A. (2005). Measurement of health-related
patients with mild dementia and their caregivers: The quality of life for people with dementia: Development of a new
multicentre, rater-blinded, randomised Danish Alzheimer instrument (DEMQOL) and an evaluation of current
Intervention Study (DAISY). BMJ Open, 3(11), e003584. methodology. Health Technology Assessment, 9(10), 1–93.
doi:10.1136/bmjopen-2013-003584 doi:10.3310/hta9100
Pinquart, M., & Sörensen, S. (2003). Associations of stressors Smith, S. C., Murray, J., Banerjee, S., Foley, B., Cook, J. C.,
and uplifts of caregiving with caregiver burden and depres- Lamping, D. L., … Mann, A. (2005). What constitutes
sive mood: A metaanalysis. Journals of Gerontology Series health-related quality of life in dementia? Development of
B: Psychological Sciences and Social Sciences, 58, 112–128. a conceptual framework for people with dementia and their
doi:10.1093/geronb/58.2.P112 carers. International Journal of Geriatric Psychiatry, 20(9),
Prince, M., Wimo, A., Guerchet, M., Ali, G., Wu, Y., & Prina, M. 889–895. doi:10.1002/(ISSN)1099-1166
(2015). World Alzheimer’s Report 2015, The Global Impact of Spector, A., Charlesworth, G., King, M., Lattimer, M.,
Dementia: An analysis of prevalence, incidence, cost and Sadek, S., Marston, L., … Orrell, M. (2015). Cognitive–
trends. London, UK: Alzheimer’s Disease International. Behavioural therapy for anxiety in dementia: Pilot rando-
Quayhagen, M. P., Quayhagen, M., Corbeil, R. R., mised controlled trial. The British Journal of Psychiatry,
Hendrix, R. C., Jackson, J. E., Snyder, L., & Bower, D. 206(6), 509–516. doi:10.1192/bjp.bp.113.140087
(2000). Coping with dementia: Evaluation of four nonphar- Stanley, M. A., Calleo, J., Bush, A. L., Wilson, N., Snow, A. L.,
macologic interventions. International Psychogeriatrics, 12 Kraus-Schuman, C., … Williams, S. P. (2013). The peace-
(2), 249–265. doi:10.1017/S1041610200006360 ful mind program: A pilot test of a cognitive–Behavioral
Quinn, C., Toms, G., Jones, C., Brand, A., Edwards, R. T., therapy–Based intervention for anxious patients with
Sanders, F., & Clare, L. (2016). A pilot randomized con- Dementia. The American Journal of Geriatric Psychiatry,
trolled trial of a self-management group intervention for 21(7), 696–708. doi:10.1016/j.jagp.2013.01.007
people with early-stage dementia (The SMART study). Stewart, A. L., Hays, R. D., & Ware, J. E. (1988). The MOS
International Psychogeriatrics, 28(5), 787–800. short-form general health survey: Reliability and validity in
doi:10.1017/S1041610215002094 a patient population. Medical Care, 26(7), 724–735.
Review Manager (RevMan) [Computer program] Version doi:10.1097/00005650-198807000-00007
5.3. (2014). Copenhagen: The Nordic Cochrane Centre. Teri, L., Logsdon, R. G., Uomoto, J., & McCurry, S. M.
The Cochrane Collaboration. (1997). Behavioral treatment of depression in dementia
Robert, P. H., Clairet, S., Benoit, M., Koutaich, J., Bertogliati, C., patients: A controlled clinical trial. The Journals of
Tible, O., … Bedoucha, P. (2002). The apathy inventory: Gerontology Series B: Psychological Sciences and Social
Assessment of apathy and awareness in Alzheimer’s disease, Sciences, 52(4), 159–166. doi:10.1093/geronb/52B.4.P159
Parkinson’s disease and mild cognitive impairment. Teri, L., Truax, P., Logsdon, R., Uomoto, J., Zarit, S., &
International Journal of Geriatric Psychiatry, 17(12), Vitaliano, P. P. (1992). Assessment of behavioral problems
1099–1105. doi:10.1002/(ISSN)1099-1166 in dementia: The revised memory and behavior problems
Römhild, J., Fleischer, S., Meyer, G., Stephan, A., Zwakhalen, S., checklist. Psychology and Aging, 7(4), 622. doi:10.1037/
Leino-Kilpi, H., … Hallberg, I. R. (2018). Inter-rater agree- 0882-7974.7.4.622
ment of the Quality of Life-Alzheimer’s Disease (QoL-AD) Testad, I., Corbett, A., Aarsland, D., Lexow, K., Fossey, J.,
self-rating and proxy rating scale: Secondary analysis of Woods, B., & Ballard, C. (2014). The value of personalized
RightTimePlaceCare data. Health and Quality of Life psychosocial interventions to address behavioral and psycho-
Outcomes, 16(1), 131. doi:10.1186/s12955-018-0959-y logical symptoms in people with dementia living in care home
Samus, Q. M., Johnston, D., Black, B. S., Hess, E., Lyman, C., settings: A systematic review. International Psychogeriatrics,
Vavilikolanu, A., … Lyketsos, C. G. (2014). 26(7), 1083–1098. doi:10.1017/S1041610214000131
A multidimensional home-based care coordination interven- Tremont, G. (2011). Family caregiving in dementia. Rhode
tion for elders with memory disorders: The maximizing Island: Medicine and Health, 94(2), 36–38.
796 E. POON

Van der Zee, K. I., Sanderman, R., Heyink, J. W., & de Haes, H. symptoms in individuals with dementia: a randomized trial
(1996). Psychometric qualities of the RAND 36-Item Health of a nonpharmacological intervention. Journal of the
Survey 1.0: A multidimensional measure of general health American Geriatrics Society, 58, 8, pp.1465–1474.
status. International Journal of Behavioral Medicine, 3(2), 104. Joosten-Weyn Banningh, L.W., Prins, J.B., Vernooij-
doi:10.1207/s15327558ijbm0302_2 Dassen, M.J., Wijnen, H.H., Rikkert, M.G.O. and Kessels, R.
Vandepitte, S., Van Den Noortgate, N., Putman, K., P. (2011). Group therapy for patients with mild cognitive
Verhaeghe, S., Faes, K., & Annemans, L. (2016). Effectiveness impairment and their significant others: results of a waiting-
of supporting informal caregivers of people with dementia: list controlled trial. Gerontology, 57, 5, pp.444–454.
A systematic review of randomized and non-randomized con- Judge, K. S., Yarry, S. J., Looman, W. J. and Bass,
trolled trials. Journal of Alzheimer’s Disease, 52(3), 929–965. D. M. (2013). Improved Strain and Psychosocial Outcomes
doi:10.3233/JAD-151011 for Caregivers of Individuals with Dementia: Findings from
Waldorff, F. B., Buss, D. V., Eckermann, A., Rasmussen, M. L. H., Project ANSWERS. The Gerontologist, 53, 2, pp.280–292.
Keiding, N., Rishøj, S., … Waldemar, G. (2012). Efficacy of Logsdon, R.G., Pike, K.C., McCurry, S.M., Hunter, P.,
psychosocial intervention in patients with mild Alzheimer’s Maher, J., Snyder, L. and Teri, L. (2010). Early-stage memory
disease: The multicentre, rater blinded, randomised Danish loss support groups: outcomes from a randomized controlled
Alzheimer Intervention Study (DAISY). BMJ, 345, e.4693. clinical trial. Journals of Gerontology Series B: Psychological
doi:10.1136/bmj.e4693 Sciences and Social Sciences, 65 6, pp.691–697.
Ware, J. E., Jr., Kosinski, M., & Keller, S. D. (1996). A 12-item McCurry, S.M., Gibbons, L.E., Logsdon, R.G., Vitiello, M.
short-form health survey: Construction of scales and pre- V. and Teri, L. (2005). Nighttime insomnia treatment and educa-
liminary tests of reliability and validity. Medical Care, 34 tion for Alzheimer’s disease: a randomized, controlled trial.
(3), 220–233. doi:10.1097/00005650-199603000-00003 Journal of the American Geriatrics Society, 53, 5, pp.793–802.
World Health Organization. (2012). Dementia: A public Orrell, M., Yates, L., Leung, P., Kang, S., Hoare, Z., Whitaker,
health priority. Geneva, Switzerland: WHO Press. C., Burns, A., Knapp, M., Leroi, I., Moniz-Cook, E. and Pearson,
Retrieved from https://www.who.int/mental_health/publi S. (2017). The impact of individual Cognitive Stimulation
cations/dementia_report_2012/en/ Therapy (iCST) on cognition, quality of life, caregiver health,
World Health Organization. (2016). Dementia factsheet, and family relationships in dementia: A randomized controlled
December 2017. Retrieved from http://www.who.int/mediacen trial. PLoS medicine, 14, 3, p. e1002269.
tre/factsheets/fs362/en Phung, K.T., Waldorff, F.B., Buss, D.V., Eckermann, A.,
Yesavage, J. A., Brink, T. L., & Rose, T. (2000). Geriatric depres- Keiding, N., Rishøj, S., Siersma, V., Sørensen, J., Søgaard, R.,
sion scale (GDS). In Association, A.P., & Rush, A.J., (Eds.), Sørensen, L.V. and Vogel, A. (2013). A three-year follow-up
Handbook of psychiatric measures (pp. 544–546). Washington, on the efficacy of psychosocial interventions for patients with
DC: American Psychiatric Association. mild dementia and their caregivers: the multicentre, rater-
Yu, D., Cheng, S., & Wang, J. (2018). Unravelling positive blinded, randomized Danish Alzheimer Intervention Study
aspects of caregiving in dementia: An integrative review of (DAISY). BMJ open, 3, 11, p.e003584.
research literature. International Journal of Nursing Quayhagen, M.P., Quayhagen, M., Corbeil, R.R., Hendrix,
Studies, 79, 1–26. doi:10.1016/j.ijnurstu.2017.10.008 R.C., Jackson, J.E., Snyder, L. and Bower, D. (2000). Coping
Zarit, S., Orr, N. K., & Zarit, J. M. (1985). The hidden victims of with dementia: evaluation of four nonpharmacologic inter-
Alzheimer’s disease: Families under stress. New York, NY: NYU ventions. International Psychogeriatrics, 12, 2, pp.249–265.
Press. Samus, Q.M., Johnston, D., Black, B.S., Hess, E., Lyman,
Zarit, S. H., Todd, P. A., & Zarit, J. M. (1986). Subjective burden C., Vavilikolanu, A., Pollutra, J., Leoutsakos, J.M., Gitlin, L.
of husbands and wives as caregivers: A longitudinal study. The N., Rabins, P.V. and Lyketsos, C.G. (2014).
Gerontologist, 26(3), 260–266. doi:10.1093/geront/26.3.260 A multidimensional home-based care coordination interven-
Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety tion for elders with memory disorders: the maximizing inde-
and depression scale. Acta Psychiatrica Scandinavica, 67 pendence at home (MIND) pilot randomized trial. The
(6), 361–370. doi:10.1111/acp.1983.67.issue-6 American Journal of Geriatric Psychiatry, 22, 4, pp.398–414.
Spector, A., Charlesworth, G., King, M., Lattimer, M., Sadek,
S., Marston, L., Rehill, A., Hoe, J., Qazi, A., Knapp, M. and
Appendix A Orrell, M. (2015). Cognitive–behavioral therapy for anxiety in
dementia: pilot randomized controlled trial. The British Journal
The 14 studies included for qualitative analysis of Psychiatry, 206, 6, pp.509–516.
Amieva, H., Robert, P.H., Grandoulier, A.S., Meillon, C., De Stanley, M.A., Calleo, J., Bush, A.L., Wilson, N., Snow, A.
Rotrou, J., Andrieu, S., Berr, C., Desgranges, B., Dubois, B., L., Kraus-Schuman, C., Paukert, A.L., Petersen, N.J., Brenes,
Girtanner, C. and Joël, M.E. (2016). Group and individual cog- G.A., Schulz, P.E. and Williams, S.P. (2013). The Peaceful
nitive therapies in Alzheimer’s disease: the ETNA3 randomized Mind program: A pilot test of a cognitive–behavioral ther-
trial. International psychogeriatrics, 28, 5, pp.707–717. apy–based intervention for anxious patients with Dementia.
Gitlin, L.N., Winter, L., Dennis, M.P., Hodgson, N. and The American Journal of Geriatric Psychiatry, 21,7,
Hauck, W.W. (2010). Targeting and managing behavioral pp.696–708.
 CLINICAL GERONTOLOGIST 797

Teri, L., Logsdon, R.G., Uomoto, J. and McCurry, S.M. Quinn, C., Toms, G., Jones, C., Brand, A., Edwards, R.T.,
(1997). Behavioral treatment of depression in dementia Sanders, F. and Clare, L. (2016). A pilot randomized con-
patients: a controlled clinical trial. The Journals of trolled trial of a self-management group intervention for
Gerontology Series B: Psychological Sciences and Social people with early-stage dementia (The SMART study).
Sciences, 52, 4, pp.159–166. International psychogeriatrics, 28(5), pp.787–800.
Copyright of Clinical Gerontologist is the property of Taylor & Francis Ltd and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.