Accelerat ing t he world's research.

Observational gait assessment tools in

paediatrics–A systematic review
Andrew Bateman

Cite this paper Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers Download a PDF Pack of t he best relat ed papers 

Bot ulinum t oxin assessment , int ervent ion and aft er-care for lower limb spast icit y in children …
Iona Novak, Cat hy Morgan

Effect iveness of Select ive Dorsal Rhizot omy in 2 Pat ient s Wit h Progressive Spast icit y Due t o Neurod…
Jules Becher

Effect iveness of Select ive Dorsal Rhizot omy In Adolescent s/Young Adult s wit h Cerebral Palsy
David Roye, Benjamin Roye
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
 Author's personal copy

Gait & Posture 40 (2014) 279–285

Contents lists available at ScienceDirect

Gait & Posture

journal homepage: www.elsevier.com/locate/gaitpost

Review

Observational gait assessment tools in paediatrics – A systematic

review
Chandrasekar Rathinam a,*, Andrew Bateman b,1, Janet Peirson c, Jane Skinner d,2
a
Block 9, Physiotherapy Department, Ida Darwin, Fulbourn, Cambridge CB21 5EE, United Kingdom
b
Oliver Zangwill Centre for Neuropsychological Rehabilitation, Princess of Wales Hospital, Ely CB6 1DN, United Kingdom
c
Block 13, Ida Darwin, Fulbourn, Cambridge CB21 5EE, United Kingdom
d
Norwich Medical School, Faculty of Medicine and Health Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom

A R T I C L E I N F O A B S T R A C T

Article history: Instrumented gait analysis (IGA) is an expensive technique used to objectively detect gait abnormalities
Received 3 December 2013 in children. Observational gait assessment is considered as a cost effective alternate for IGA in regular
Received in revised form 13 March 2014 clinical practice. This article is aimed at systematically reviewing the available paediatric gait analysis
Accepted 10 April 2014
tools and examines their reliability and validity compared to IGA. This review also examines the
structure of these tools, their clinical use and limitations. Articles were searched from PubMed, CINHL,
Keywords: AMED, BNI, EMBASE, PEDro and Cochrane library from the earliest record on the database to December
Observational gait analysis
2012. Hand searches were carried out in a few journals. Studies that examined children’s gait using a
Cerebral palsy
Video
structured assessment tool were included and analysed for their quality, reliability and validity. Pre-
Validity established criteria were used to judge the quality of methodology and reliability and validity. Five
Reliability observational gait tools for children with Cerebral Palsy (CP) and one for children with Downs Syndrome
were identified. Nine studies related to children with CP were enrolled for this review. None of the tools
have accomplished the level of IGA’s consistency. Edinburgh Visual Gait Score (EVGS) was found to have
better reliability and validity than the other tools. Very limited studies were available for most of the gait
assessment tools therefore their clinical use cannot be judged based on the existing evidence. EVGS was
found to have better concurrent validity and reliability and it should be considered to assess CP gait in
regular practice. Future work to investigate the use of low cost technology to improve observers’
accuracy of EVGS is suggested.
ß 2014 Elsevier B.V. All rights reserved.

1. Introduction measures, electromyography activity and 3 dimensional joint

kinematic and kinetic values to detect gait abnormality with
Gait poses a significant challenge to those with neurological accuracy [2–5]. IGA has been widely used to assess the gait of
disorders and analysing the deviation of gait is an integral part of a children with cerebral palsy (CP) and helps the clinician to
visual assessment. Gait assessment assists in determining the formulate a management plan and evaluate the outcome of the
degree and cause of abnormality and it can be used as an outcome intervention [6]. A gait laboratory requires considerable capital
measure to evaluate the effectiveness of intervention [1]. investment, trained personnel, and is not often readily accessible
Instrumented gait analysis (IGA), the gold standard for evaluation for routine clinical work [7,8]. Each session which includes the
of movement, uses sophisticated technology that enables objective assessment and interpretation of the results by experts takes
analysis of patients’ mobility in laboratory environments. IGA approximately three to six hours [6].
technology involves the assessment of video recordings, clinical Clinicians require simple and cost-effective outcome measures
to analyse the kinematic parameters of gait in their day-to-day
practice. Visual diagnosis of a patient’s gait in real time is
* Corresponding author. Tel.: +44 1223 883382. subjective, lacks accuracy and relies on the clinician’s training
E-mail addresses: [email protected] (C. Rathinam), and experience [9]. Visual gait analysis using a structured
[email protected] (A. Bateman), [email protected] (J. Peirson),
proforma has been suggested as an alternative to the IGA [5].
[email protected] (J. Skinner).
1
Tel.: +44 1353 652169; fax: +44 1353 652164. Observational gait tools are often and widely used as an essential
2
Tel.: +44 01603 59 3120. tool for an assessment of gait problems of children with CP [8]. In

http://dx.doi.org/10.1016/j.gaitpost.2014.04.187
0966-6362/ß 2014 Elsevier B.V. All rights reserved.
 Author's personal copy

280 C. Rathinam et al. / Gait & Posture 40 (2014) 279–285

observational gait analysis the examiner visually assesses the gait 2.2. Inclusion and exclusion criteria
pattern with the aid of video recordings using various scales that
describe gait abnormalities in different joints and planes [5]. 2.2.1. Participants
Video recording of gait in the clinical setting is relatively easy and Gait abnormality in children (0–18 years) with neurological,
a preferred method of examination for clinicians. Computer- neuromuscular, orthopaedic and other developmental delay due to
based video image analysis systems are able to provide an genetic disorders. Studies that involved the adult population were
interface for precise recording, quantifying and the analysing of excluded.
events [10]. They can be applied to capture abnormal posture and
movements, to reduce them to basic parameters such as joint 2.2.2. Tools
angles, swing distances and curvatures [10]. However, Toro et al. Only observational and video gait analysis tools that assess
quoted several authors in saying that observational gait assess- either reliability, validity or both compared against IGA were
ment is relatively subjective in nature and that it may lead to included. Studies describing IGA data alone on gait and as gait
poor validity, reliability, sensitivity, and specificity compared to index based on IGA were excluded.
IGA [11].
In order to overcome the shortfalls of the real time gait 2.2.3. Study type
examination, various gait assessment tools have been developed Any type of study that reported observational gait analysis
over the past two decades to assess children’s gait from video including commentaries and case studies in journals was included.
recorded files [2,3,12–14]. Examination of the reliability and Dissertations, conference abstracts and other sources of unpub-
validity of those tools has given a variable range of results which lished data were not included.
has been influenced by the experience of examiners and their
professional background [1,2,4,5,7,9,12,15–17]. Despite the 2.3. Data extraction
variable level of accuracy and reliability of the visual gait
assessment tools there continues to be a dependency on this The reviewers preliminarily screened the titles and abstracts of
method [2,7]. the references formed by the literature search based on the
A simple gait tool is needed for clinicians to identify and inclusion and exclusion criteria and duplicated independently. Full
quantify changes in a walking pattern at individual anatomical articles that met the review criteria were gathered for further
levels. It is required to reflect the deviation from the normal gait evaluation.
parameters both in the stance and swing phase, and truly reflect
functional gait problems. This review is aimed at identifying the
2.4. Appraisal of reliability and validity
variety of paediatric gait analysis tools that have been reported in
the literature, and examine their reliability and validity compared
Research concerning the reliability and validity of the clinical
to IGA. It considers the structure of these tools, their quality,
tools are judged by their psychometric properties and high
clinical use and their limitations.
methodological qualities [18]. A checklist was designed by Brink
and Louw to appraise the quality of reliability or validity and the
combined reliability and validity of studies. It can also be used to
2. Materials and methods
assess the reporting quality of objective clinical tools. Bellet et al.
modified this further to assess the responsiveness element [19].
2.1. Search strategy
The modified checklist has seventeen items (4 reliability, 4 validity,
1 responsiveness and 8 generic items) with specific scoring criteria
In December 2012, a comprehensive computerised biblio-
which ensure a ‘yes’, ‘no’ or ‘not applicable’ response. Although this
graphic databases search was performed in the following
tool does not report a quality score, the studies scoring more than
database: PubMed (1966 to current), Cumulative Index to Nursing
60% of positive responses were considered to be of good quality
and Allied Health Literature (CINHL; 1982 to 2012), Allied and
[19] (Table 1). The quality of the extracted articles, their reliability,
Complementary Medicine (AMED; 1985 to current), British
validity and responsiveness were assessed using the checklist. The
Nursing Index (BNI; 1992 to current) and Excerpta Medica
examiners (1st and 3rd author) had a good understanding of the
Database (EMBASE; 1980 to current). We extended our search
scoring method, they assessed the quality of the articles
to the Cochrane library and Physiotherapy Evidence Database
independently and reached a consensus through discussion.
(PEDro). Scopus database was separately accessed to find relevant
citations and articles. 3. Results
The following search terms were used: (observation* OR
The search strategy resulted in 1508 citations and after removing duplicates this
Video*) AND (Gait* OR Walk* OR GAIT) AND (Analy* OR Examinat* was narrowed down to 961 citations. The preliminary selection, based on the title
OR Assess*). A broad search strategy included free-text words, and abstracts contained 58 citations, with information about gait tools and gait
medical subject heading and all thesauruses subject terms in the variations, reliability and validity. We collected full text articles for all of the
citations. Eight articles that fulfilled the criteria were chosen for the review (Fig. 1).
database wherever applicable. In order to limit the number of
The corresponding authors of all of the selected articles were contacted to identify
results the searches were confined to humans only, and the any additional related work published or unpublished, which had been carried out
paediatric population (age group 0–18 years; different variations on the scales.
denoting the age limit which included child, infant, pre-school and Over the past two decades five different gait assessment tools to assess the gait of
adolescent) wherever possible. No limit was set for language and children with CP and one tool to assess children with Downs Syndrome have been
developed (Fig. 2). The authors of some gait scales [1,2,9,15,17,20,21] have
all were included.
confirmed that they have not carried out further studies on their tool and reported
In addition to the electronic database search, a hand search was that they were not aware of any other studies by associates. In order to maintain
carried out in specific journals [Archives of Physical Medicine and homogeneity only the gait tools related to children with CP were included for this
Rehabilitation (1995–2012), Gait and Posture (1995–2012), review.
Developmental Medicine and Child Neurology (1989–2012) and In every study that has been included in this review, the authors have
acknowledged that the IGA is the accepted gold standard and used the IGA result to
Paediatric Physical Therapy (1989–2012)]. A further search establish their tools’ validity. All the tools were compared to IGA. The studies also
tracking citation of all the primary studies were scanned and indicated that the examiners extracted the data from pre-recorded video footage of
examined for inclusion. gait from the gait lab and compared them with their corresponding IGA data. The
 Author's personal copy

C. Rathinam et al. / Gait & Posture 40 (2014) 279–285 281

Table 1
Critical appraisal of articles for reliability and validity.

Tool Authors 1 2 3a 4b 5b 6b 7a 8b 9a 10 11a 12 13 14c 15 16 17 %Y

Observational Gait Scale [12] Araujo et al. Y Y Y N N Y Y Y Y Y Y Y Y n/a Y N Y 82

Visual Gait Score [1] Kawamura et al. Y Y Y Y n/a n/a Y Y Y Y Y Y Y n/a Y N Y 94
Salford Gait Tool [9] Toro et al. Y Y Y Y n/a n/a Y Y Y Y Y Y Y n/a Y N Y 94
Salford Gait Tool [3] Toro et al. Y Y N Y Y Y Y Y Y Y Y Y Y n/a Y N Y 88
Edinburgh Visual Gait Scale [14] Read et al. Y Y Y N N Y Y Y Y Y Y Y Y n/a Y N Y 82
Edinburgh Visual Gait Scale [16] Ong et al. Y Y N N N Y Y Y Y Y Y Y Y n/a Y N Y 76
Edinburgh Visual Gait Scale [23] Viehweger Y Y Y N Y Y Y Y Y Y Y Y Y n/a Y N Y 88
Observational Gait Analysis* [15] Mackey et al. Y Y Y Y Y Y Y Y Y Y Y Y Y n/a Y N Y 94
Physician Rating Scale* [17] Wren et al. Y Y Y N N Y Y Y Y Y Y Y Y n/a Y N Y 82
Visual Gait Assessment Scale* [7] Dickens and Smith Y Y Y Y N n/a Y Y Y Y Y Y Y n/a Y N Y 88
Visual Gait Assessment Scale* [4] Brown et al. Y Y Y Y Y Y Y Y Y Y Y Y Y n/a Y N Y 94
Physician Rating Scale* [24] Corry et al. Y Y Y n/a n/a n/a Y Y Y Y Y Y Y n/a Y N Y 94
a
Questions relevant to validity.
b
Questions relevant to reliability.
c
Questions relevant to responsiveness.
Y, yes; N, no; n/a – not applicable.
Bellet’s questionnaire items:
(1) If human subjects were used, did the authors give a detailed description of the sample of subjects used to perform the (index) test?
(2) Did the authors clarify the qualification, or competence of the rater(s) who performed the (index) test?
(3) Was the reference standard explained?
(4) If interrater reliability was tested, were raters blinded to the findings of other raters?
(5) If intrarater reliability was tested, were raters blinded to their own prior findings of the test under evaluation?
(6) Was the order of examination varied?
(7) If human subjects were used, was the time period between the reference standard and the index test short enough to be reasonably sure that the target condition did not
change between the two tests?
(8) Was the stability (or theoretical stability) of the variable being measured taken into account when determining the suitability of the time interval between repeated
measures?
(9) Was the reference standard independent of the index test? Validity studies.
(10) Was the execution of the (index) test described in sufficient detail to permit replication of the test?
(11) Was the execution of the reference standard described in sufficient detail to permit its replication?
(12) Were withdrawals from the study explained?
(13) Were the statistical methods appropriate for the purpose of the study?
(14) Does the instrument capture clinical change?
(15) Were the subjects selected either randomly or consecutively?
(16) Was the number of subjects either >50 or was a sample size calculation provided?
(17) Did subjects give consent prior to testing?

[(Fig._1)TD$IG]
Reasons for exclusion

Potentially relevant citations after

electronic search (n = 961)
First screening of titles and abstracts

903 abstracts excluded as

clearly not relevant

Second screening of collected full

articles (n = 58)

19 - Technology related
17 - General Gait description
and others
6 - Gait index

Third screening of full articles after

including hand searched and reference
checking- consensus meeting (n = 22)

8 – Not compared against

Instrumented gait analysis
3 – Comparison of tools
1 - Examined
responsiveness
1 – Children with Downs
syndrome

Nine relevant articles entered for final

review

Fig. 1. Flow of studies in systematic review.

 Author's personal copy

[(Fig._2)TD$IG]
282 C. Rathinam et al. / Gait & Posture 40 (2014) 279–285

Tool PRS SF-GT OGA OGS EVGS

Plane Sagi Trans Cor Sagi Trans Cor Sagi Trans Cor Sagi Trans Cor Sagi Trans Cor
Phase St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St Sw St S
w
Ankle
Knee
Hip
Pelvis
Trunk
Sagi – Sagittal
Trans – Transverse
Cor – Coronal
St – Stance
Sw – Swing
PRS – Physician Rating Scale
SF-GT – Salford Gait Tool
OGA – Observational Gait Analysis
OGS – Observational Gait Scale
EVGS – Edinburgh Visual Gait Score
– Item included in the tool
- Items not included in the tool

Fig. 2. General structure of Gait scales in different planes and phases.

examiners reported their assessment findings of individual joints using their gait to maintain uniformity, only the mean value data from the original cited articles
tools and compared them to the same measures in the IGA. Although all the studies were given in Table 2 [1,3,4,7,9,12,14–16].
examined the hip, knee and ankle joints, the assessment criteria and the number of
items in each joint examined at stance and swing phase of the gait cycle were not 3.1. Description of gait tools
the same. The statistical methods used also varied. In order to describe the validity,
all the related papers reported how far the examiners’ findings are in agreement The authors of four of five tools, developed and standardised their tool by video
with the IGA results. For ease of understanding the reliability and validity result and recording children walking a set distance at their self-selected speed in an

Table 2
Critical appraisal of articles for reliability and validity.

Tool Authors Segment Mean Validity Mean Reliability Total number of

Intra-rater Inter-rater Examiners Children

Observational Gait Araujo et al. (2009) Ankle wk = 0.59 wk = 0.79 wk = 0.68 4 18 (6–12 years)
Scale [12] Knee wk = 0.64 wk = 0.77 wk = 0.65
Hip wk = 0.20 wk = 0.73 wk = 0.48
Pelvis wk = 0.23 wk = 0.59 wk = 0.30
Visual Gait Score [1] Kawamura et al. (2007) Ankle and foot k = 0.18 – k = 0.67 4 50 (<8 years)
Knee k = 0.39 k = 0.41
Hip k = 0.22 k = 0.37
Pelvis k = 0.37 k = 0.56
Salford Gait Tool [3,9] Toro et al. (2007, 2007) Ankle LSD = 12.09 MA = 73% MA = 75% 1st Ax = 23 13 (6–16 years)
Knee LSD = 14.43 MA = 78% MA = 81% 2nd Ax = 17
Hip LSD = 19.30 MA = 72% MA = 77%
Edinburgh Visual Read et al. (2003) Ankle PA = 68.3% – k = 0.49 5 5 (9–15 years)
Gait Scale [14] Knee PA = 55.0% k = 0.41
Hip PA = 71.5% k = 0.25
Pelvis PA = 62.0% k = 0.23
Trunk – k = 0.31
Edinburgh Visual Ong et al. (2008) Ankle PA = 54.7% – k = 0.36 6 5 (9–15 years)
Gait Scale [16] Knee PA = 42.3% k = 0.25
Hip PA = 60% k = 0.14
Pelvis PA = 54.5% k = 0.18
Trunk – k = 0.09
Observational Gait Mackey et al. (2003) Foot wk = 0.62 wk = 0.65 wk = 0.59 2 20 (6–12 years)
Analysis [15] Knee wk = 0.57 wk = 0.62 wk = 0.69
Base of support – wk = 0.38 wk = 0.38
Visual Gait Assessment Dickens and Smith (2008) Foot – k = 0.69 k = 0.54 2 31 (5–17 years)
Scale [7] Knee k = 0.20 k = 0.54 k = 0.69
Hip k = 0.23 k = 0.25 k = 0.73
Visual Gait Assessment Brown et al. (2008) Ankle – PA = 75 PA = 75.7 10 4 (No age range
Scale [4] Knee PA = 51.1 PA = 60.9 PA = 51.0 details given)
Hip PA = 56.6 PA = 62 PA = 49.4
Physician Rating Wren et al. (2005) Foot – wk = 0.65 wk = 0.64 4 30 (5–20 years)a
Scale [17] Ankle wk = 0.53 wk = 0.46
Knee wk = 0.68 wk = 0.62
Hip wk = 0.41 wk = 0.32
Crouch wk = 0.75 wk = 0.64
k, kappa; wk, weighted kappa; Ax, assessment; LSD, least significant difference; PA, percentage agreement.
a
Age range goes beyond 18 years.
 Author's personal copy

C. Rathinam et al. / Gait & Posture 40 (2014) 279–285 283

instrumented gait lab [1,3,12,14]. Raters were encouraged to assess the distribution of CP and compared those 10 items included in OGA against IGA data.
participants’ gait using computer monitors. They watched the participants gait a Their results showed higher inter-observer agreement and better criterion validity
number of times and were allowed to score at their own pace without time for knee flexion at initial contact and pelvic obliquity but the other 8 items were
restriction. found to be unreliable for visual observation.

3.1.1. Observational Gait Scale 3.1.4. Edinburgh Visual Gait Score

The Observational Gait Scale (OGS) was a nominal scale developed by Araujo The Edinburgh Visual Gait Score (EVGS) is a comprehensive video assessment
et al. for children with CP [12]. It accommodates observations of items which developed by Read et al. to assess the gait of children with CP [14]. EVGS provides a
deviate from the normal direction in all three planes. The authors also considered numerical value of 17 gait parameters in the foot, knee, pelvis and trunk in both the
the raters’ variable level of experience and training level while developing OGS. This stance and swing phase. These selected items represent key features of the
scale consists of 24 items (ankle/foot – 6, knee – 5, hip – 8 and pelvis – 5) chosen pathological gait in children with CP. They target the trunk, pelvis, hip, knee, ankle
based on what is thought to be the most representative kinematic gait parameters and foot in the sagittal, coronal, and transverse planes. In order to assist the
of gait in children with CP. The OGS tool was not available as cited in the website examiners, each joint’s movement range was calculated from normal IGA and was
(http://www.forusers.com/escala) and the corresponding author of this tool shared used to determine the gait deviation. These values were specified to assist observers
their copy but no related manual or additional instructions were supplied. It is where relevant. The deviations were subdivided into two clinically relevant degrees
inferred that the OGS examines gait in both stance and swing phase. Further details of severity (moderate and marked) based on magnitude. This tool uses a three-point
including the scoring system, score range and their significance were not available. ordinal scale indicating normal, moderate and marked deviation, respectively. The
total score range 0–34 where 0 denotes normal and >0 denotes abnormal gait. It has
good intraobserver and interobserver reliability among experienced and inexperi-
3.1.2. Salford Gait Tool enced raters and it correlates well with IGA results [14,16,23]. Available literature
The Salford Gait Tool (SF-GT) was an ordinal scale developed by Toro et al. to showed that scoring of the ankle and knee joint angles is easier and more reliable
examine the gait of children with CP in the sagittal plan [3]. SF-GT assesses the hip, than the trunk, hip and pelvis. Furthermore there is better reliability for stance
knee, and ankle joint positions at 6 gait events (initial contact, end double support, phase items than for swing phase items [14,23]. There was notable difference in
midstance, start double support, toe off and mid swing) in stance and swing phases. scoring subscales especially for the proximal segment and swing phase items.
The authors created a 5 point scale (2, 1, 0, 1, 2) to represent variations of joint
ranges both in categories and degrees to enable the examiners to qualitatively
describe the joints abnormality in these gait events. The output is a numeric 3.1.5. Physician’s Rating Scale
indicator for each of the 3 joints and the summed score provides an indication of The Physicians Rating Scale (PRS) was created by Koman et al. to examine the gait
gait pathology. The authors of this tool previously defined 13 gait styles [22] and of children with CP in the sagittal plane [13]. PRS was modified by other authors and
they adjusted the SF-GT’s upper and lower boundary scores, and the total scores quoted in the literature as modified PRS [24], observational gait scale [20],
reflect the 13 different gait styles using IGA. Toro et al. included a large cohort of Abbreviated PRS [25], Visual Gait Analysis Scale [26], and Visual Gait Analysis Scale
assessors with a wide range of examiners experience to establish SF-GT’s reliability [7]. Discussion of all the reported PRS’s variations in the literature will be beyond
and found that the knee and hip joint ranked high and low, respectively, in the level the scope of this review, therefore we included the recent version of PRS [4] for this
of agreement between and within observers. review but outlined the key differences between the other PRS variations in Table 3.
PRS has 6 parameters with its own defined categories based on degree ranges for
the hip and knee (score range 1–3 or 4) in the stance and swing phases of gait. It also
3.1.3. Observational gait analysis includes the initial foot contact, stance phase foot contact and the timing of the heel
Kawamura et al. reported using observational gait analysis (OGA) in children rise in its structure. These parameters are measured using a scoring system (range:
with CP [1]. OGA is an ordinal scale examining hip, knee, ankle and pelvis (10 items) 1–3 and 5) and the total score of PRS ranges from 1 to 26. The non-standardisation
in the sagittal, coronal and transverse plane both in stance and swing phases. OGA and lack of uniformity of the scoring system compromises the tool’s sensitivity.
estimated the individual items either as normal or decreased/increased and the The original version of PRS had no record of its development, reliability, or
outcome result was compared with the kinematics data of the subjects studied. validity and psychometric properties. Many authors have attempted to address the
They retrospectively examined the gait videos of 50 children with a diplegic above short falls and Brown et al. summarised and quoted as [4] ‘‘Corry et al.

Table 3
Physician Rating Scale and its modified versions quoted in the literature.

Author Scale name No. of items Scale section Score (range)

Knee Foot Other items

Koman et al. PRS [13] 6 Recurvatuma Equinus Crouch (involves hip, (0 to 14)b
Hindfoot position knee and ankles)a
Foot contact in stance Speed of gait
Corry et at PRS [24] 4 Recurvatuma Foot contact in stance Crouch (involves hip, ( 1 to 4)b
knee and ankles)a
Change in gait
Boyd and Graham Observational 8 Knee in midstancea Initial foot contact Timing of heel rise ( 1 to 3)b
Mackey et al. Gait Analysis [15,20] Foot contact in midstance Base of support
Hindfoot in midstance Gait assistive devices
Change in gait
Flett et al. Abbreviated PRS [25] 2 Foot contact Crouch (involves hip, (0 to 4)b
knee and ankles) a
Koman et al. PRS [21] 6 Knee in stance Foot contact in stance Crouch (involves hip, (0 to 14)b
Hindfoot position during knee and ankles)a
foot strike Speed of gait
Hind foot position
during stance
Ubhi et al. Video Gait Analysis 1 Initial foot contact Crouch (involves hip, (0 to 4)
Scoring [26] knee and ankles)
a
Wren et al. PRS [17] 5 Knee flexion in stance Foot contact Hip flexion in stance (0 to 4)b
Dorsiflexion in stancea
Maathuis et al. PRS [27] 4 Recurvatuma Foot contact Crouch (involves hip, ( 1 to 4)b
knee and ankles)a
Change in gait
Dickens and smith; Visual gait assessment 8 Knee peak extension Initial foot contact Hip in terminal stancea (1 to 5)b
Brown et al. scale [4,7] in terminal stancea Foot contact in stance Hip in midswinga
Knee peak flexion in swinga Timing of heel rise

PRS, Physician Rating Scale.

a
Variable score range for individual items.
b
Involves degree range.
 Author's personal copy

284 C. Rathinam et al. / Gait & Posture 40 (2014) 279–285

calculated the inter-observer reliability of a modified PRS but did not consider the to be evaluated and at what point in the gait cycle they should be
intra-observer reliability [24]; Flett et al. used a modified PRS in their study but this
assessed [17]. PRS contents were modified to suit the individual
was not tested for reliability or validity [26]; Koman et al. did not report validity or
reliability of the PRS [13]; Ubhi et al. used a modification of the foot contact subscale research need as an outcome measure. There is however very
of the PRS and assessed intra- and interobserver reliability but not validity [26]. limited validation for the modifications which were carried out
Mackey et al. examined the reliability and validity of the observational gait analysis and this compromises the reliability and validity [20]. Despite this
and found that the hip showed more variation than distal joint at initial foot contact factor and limited variations of subsection of the PRS, it has been
[15]. Wren et al. investigated the validity and reliability of the modified PRS and
found that inter-observer reliability was worst for the hip and ankle position and
acknowledged as an easy tool to administer and accommodates the
best for the crouch, foot contact and knee position [17]. Maathuis et al. reported observers’ variable level of experience [4,7,15,17].
‘excellent’ intra-observer reliability and poor inter-observer agreement but did not Bella et al. compared OGS, EVGS and a modified version of PRS
examine its validity [27]. Dickens and Smith investigated the reliability and validity in children with CP to find out the agreement level and how far
of the Visual Gait Assessment Scale (VGAS) [7] and found poor agreement of validity
these scales deviated from normal gait scores [28]. They reported
and reported further that the Intra- and inter-observer reliabilities had better
agreement for knee and foot than at the hip’’ [4]. the highest level of agreement index for PRS and EVGS, and
commented that they were easy to understand and administer
[28]. They concluded that these two tools were more appropriate
4. Discussion to assess the gait of children with a diplegic distribution of CP.
However, they were typically evaluated by experts and this could
This systematic review has identified 5 observational gait tools lead to a higher intra- and inter repeatability results. The outcome
for children with CP and none of the tools have accomplished the could be different if assessed by less experienced observers.
level of IGA’s consistency. In general they were easy to administer Many low cost advanced technologies are currently emerging.
for those with gait analysis experience, but are likely to be These include video gait assessment software, and also mobile
challenging for those with little or no experience. These tools are phone applications which can measure joint angles. These can be
valuable to determine the effect of clinical intervention on gait [8] utilised in the clinical setting and may be the way forward in
but they cannot be used for pre surgical planning or for diagnostic improving observational video gait analysis. This hypothesis needs
purpose due to their limited reliability and validity. In addition the to be tested through well-structured studies in different patient
total gait score is not a reliable indicator of gait deviation severity. groups.
OGS was reported to have very good inter rater reliability One of the limitations of this review was that some of the
however only the sagittal plane (ankle/foot and knee joints) items authors of the reviewed articles did not respond to our request to
scored maximum agreement. OGS was claimed to be an easy tool inform us of any further research on their observational gait
to use and took account of the raters clinical experiences for high assessment tools. This has limited our ability to confirm other
consistency of intra-rater data [12]. On the contrary, Bella et al. relevant unpublished and/or on-going research of these tools.
reported OGS as being a difficult scale to understand, to administer Although the appraisal tool used in this study rated the articles
and required a very high level of training for the raters to reviewed as good quality, it was not considered to reflect the true
standardise observations [28]. It was difficult to judge this point quality of the studies which were included. At present the checklist
due to the limited availability of OGS information. used in this study is the best available tool to assess reliability and
Toro et al.’s SF-FT’s inter- and intraobserver repeatability and validity related studies. Brink and Louw’s suggestion of reporting
intra observer agreement was reported to be better than EVGS and the association between the individual items and the methodo-
PRS. SF-GT’s mean agreement (58%) was slightly less than EVGS logical quality of the studies [18] needs to be considered carefully
(64%) but comparing these tools validity was limited due to the when appraising reliability and validity related studies.
different statistical methods employed in the studies. SF-GT’s In conclusion, our systematic review identified five video based
better outcome is confined to Toro et al.’s own classification of gait gait assessment tools to assess children with CP’s gait but they
styles [22]. SF-GT’s clinical use and user friendliness are not were not equal in their objectivity, reliability or validity to IGA. The
reported further. OGS, OGA and PRS lacked reference for the original source. OGA
Kawamura et al.’s OGA was reliable only for knee flexion at was poorly constructed and we are unable to endorse its use as an
midstance and pelvic obliquity; however it has higher inter rater outcome measure. We could not comment on the credibility of OGS
observer reliability. Kawamura et al. gave no justification for the and advocate that it should not be used until further studies
selected items for OGS but in comparison with kinematic values appraising its reliability, validity and clinical use are available. PRS
they concluded that visual observation was inadequate for gait has many variations with limited reliability and validity results,
analysis and favoured IGA. The OGA’s scoring method; the total and this scale is inadequate to describe the overall picture of gait.
score value; and the key differences between OGA and other SF-GT has good concurrent validity and reliability but examines
available scales at the time of construction were not documented. the sagittal plane gait deviation only and requires further
This was also true of the reliability and validity of the scale. These validation. EVGS includes gait data in all three planes and has
factors have led to question Kawamura et al.’s conclusion of good reliability and concurrent validity. EVGS is better than other
discarding video based visual observation in gait analysis. tools which can be used by examiners who possess a variable range
The EVGS is a more extensive tool to identify gait deviations in of experience but none of the tools are nearly equal to IGA. We
children with CP [28] and is sensitive enough to pick up changes suggest that the EVGS is the best scale currently available and it
following intervention [29]. EVGS provides an indication of the should be considered to assess the gait pattern of children with CP.
quality of gait, presenting good concurrent validity due to its Future work to investigate the use of low cost technology to
strong agreement with other evaluation methods [28,30]. Ong improve the observers’ accuracy of EVGS is recommended.
et al. investigated the reliability and validity of EVGS for
inexperienced observers and found that inexperienced observers
were reasonably reliable and ranked the scores similarly but less Acknowledgements
accurately than their experienced counterparts [16].
PRS has been widely quoted in the literature to assess the gait of The authors would like to thank Dr. Neil Messenger, Senior
children with CP [4,7,13,15,17,20,21,24,25,26] but it was only Lecturer, University of Leeds for his help with gait lab information;
validated for children who have CP with a hemiplegic distribution Dr. Paula Waddingham, Post Doctoral Fellow, Cambridgeshire
[7]. PRS does not provide any guidelines regarding specific angles Community Services NHS Trust and Dr. Lee Hooper, Senior Lecturer
 Author's personal copy

C. Rathinam et al. / Gait & Posture 40 (2014) 279–285 285

in Research Synthesis and Nutrition, University of East Anglia for [13] Koman LA, Mooney 3rd JF, Smith BP, Goodman A, Mulvaney T. Management of
spasticity in cerebral palsy with botulinum-A toxin: report of a preliminary,
technical help, the staff of Hinchingbrooke Healthcare Library, randomized, double-blind trial. J Pediatr Orthop 1994;14(3):299–303.
Hinchingbrooke Health Care NHS Trust and Ms. Isla Kuhn, [14] Read HS, Hazlewood ME, Hillman SJ, Prescott RJ, Robb JE. Edinburgh visual gait
Cambridge University Medical Library for their help in training score for use in cerebral palsy. J Pediatr Orthop 2003;23(3):296–301.
[15] Mackey AH, Lobb GL, Walt SE, Stott NS. Reliability and validity of the Obser-
and collection of articles. vational Gait Scale in children with spastic diplegia. Dev Med Child Neurol
This article was written as part of fully funded East Midlands 2003;45(1):4–11.
Health, Innovation and Education Cluster internship programme [16] Ong AM, Hillman SJ, Robb JE. Reliability and validity of the Edinburgh Visual
Gait Score for cerebral palsy when used by inexperienced observers. Gait
(2012–2013). The study sponsors had no involvement in the study Posture 2008;28(2):323–6.
design, collection, analysis and interpretation of data, writing of [17] Wren TA, Rethlefsen SA, Healy BS, Do KP, Dennis SW, Kay RM. Reliability and
the manuscript, and in the decision to submit the manuscript for validity of visual assessments of gait using a modified physician rating scale
for crouch and foot contact. J Pediatr Orthop 2005;25(5):646–50.
publication.
[18] Brink Y, Louw QA. Clinical instruments: reliability and validity critical apprai-
sal. J Eval Clin Pract 2012;18(6):1126–32.
[19] Bellet RN, Adams L, Morris NR. The 6-minute walk test in outpatient cardiac
Conflict of interest statement rehabilitation: validity, reliability and responsiveness – a systematic review.
Physiotherapy 2012;98(4):277–86.
[20] Boyd RN, Graham HK. Objective measurement of clinical findings in the use of
None declared.
botulinum toxin type A for the management of children with cerebral palsy.
Eur J Neurol 1999;6:s23–35.
[21] Koman LA, Brashear A, Rosenfeld S, Chambers H, Russman B, Rang M, et al.
References Botulinum toxin type a neuromuscular blockade in the treatment of equinus
foot deformity in cerebral palsy: a multicenter, open-label clinical trial.
[1] Kawamura CM, de MCF, Barreto MM, de SKA, Juliano Y, Novo NF. Comparison Pediatrics 2001;108(5):1062–71.
between visual and three-dimensional gait analysis in patients with spastic [22] Toro B, Nester CJ, Farren PC. Cluster analysis for the extraction of sagittal
diplegic cerebral palsy. Gait Posture 2007;25(1):18–24. gait patterns in children with cerebral palsy. Gait Posture 2007;25(2):
[2] Martin K, Hoover D, Wagoner E, Wingler T, Evans T, O’Brien J, et al. Develop- 157–65.
ment and reliability of an observational gait analysis tool for children with [23] Viehweger E, Zurcher Pfund L, Helix M, Rohon MA, Jacquemier M, Scavarda D,
Down syndrome. Pediatr Phys Ther 2009;21(3):261–8. et al. Influence of clinical and gait analysis experience on reliability of
[3] Toro B, Nester CJ, Farren PC. The development and validity of the Salford Gait observational gait analysis (Edinburgh Gait Score Reliability). Ann Phys Reha-
Tool: an observation-based clinical gait assessment tool. Arch Phys Med bil Med 2010;53(9):535–46.
Rehabil 2007;88(3):321–7. [24] Corry IS, Cosgrove AP, Duffy CM, McNeill S, Taylor TC, Graham HK. Botulinum
[4] Brown CR, Hillman SJ, Richardson AM, Herman JL, Robb JE. Reliability and toxin A compared with stretching casts in the treatment of spastic equinus: a
validity of the Visual Gait Assessment Scale for children with hemiplegic randomised prospective trial. J Pediatr Orthop 1998;18(3):304–11.
cerebral palsy when used by experienced and inexperienced observers. Gait [25] Flett PJ, Stern LM, Waddy H, Connell TM, Seeger JD, Gibson SK. Botulinum toxin
Posture 2008;27(4):648–52. A versus fixed cast stretching for dynamic calf tightness in cerebral palsy.
[5] Grunt S, van PJ, van MMK, Brehm MA, Doorenbosch CAM, Becher JG. Repro- J Paediatr Child Health 1999;35(1):71–7.
ducibility and validity of video screen measurements of gait in children with [26] Ubhi T, Bhakta BB, Ives HL, Allgar V, Roussounis SH. Randomised double blind
spastic cerebral palsy. Gait Posture 2010;31(4):489–94. placebo controlled trial of the effect of botulinum toxin on walking in cerebral
[6] Narayanan UG. The role of gait analysis in the orthopaedic management of palsy. Arch Dis Child 2000;83(6):481–7.
ambulatory cerebral palsy. Curr Opin Pediatr 2007;19(1):38–43. [27] Maathuis KG, van CPC, van A, Rietman HS, Geertzen JH. Gait in children
[7] Dickens WE, Smith MF. Validation of a visual gait assessment scale for children with cerebral palsy: observer reliability of Physician Rating Scale and Edin-
with hemiplegic cerebral palsy. Gait Posture 2006;23(1):78–82. burgh Visual Gait Analysis Interval Testing scale. J Pediatr Orthop 2005;25(3):
[8] Harvey A, Gorter JW. Video gait analysis for ambulatory children with cerebral 268–72.
palsy: why, when, where and how! Gait Posture 2011;33(3):501–3. [28] Bella GP, Rodrigues NB, Valenciano PJ, Silva LM, Souza RC. Correlation among
[9] Toro B, Nester CJ, Farren PC. Inter- and intraobserver repeatability of the the Visual Gait Assessment Scale: Edinburgh Visual Gait Scale and Observa-
Salford Gait Tool: an observation-based clinical gait assessment tool. Arch tional Gait Scale in children with spastic diplegic cerebral palsy. Rev Bras
Phys Med Rehabil 2007;88(3):328–32. Fisioter 2012;16(2):134–40.
[10] Chang R, Guan L, Burne JA. An automated form of video image analysis applied [29] Gupta S, Raja K. Responsiveness of Edinburgh Visual Gait Score to orthopedic
to classification of movement disorders. Disabil Rehabil 2000;22(1/2):97–108. surgical intervention of the lower limbs in children with cerebral palsy. Am J
[11] Toro B, Nester C, Farren P. A review of observational gait assessment in clinical Phys Med Rehabil 2012;91(9):761–7.
practice. Physiother Theor Pract 2003;19(3):137–49. [30] Hillman SJ, Hazlewood ME, Schwartz MH, Van MLL, Robb JE. Correlation of the
[12] Araújo PA, Kirkwood RN, Figueiredo FM. Validity and intra- and inter-rater Edinburgh Gait Score with the Gillette Gait Index: the Gillette Functional
reliability of the Observational Gait Scale for children with spastic cerebral Assessment Questionnaire, and dimensionless speed. J Pediatr Orthop
palsy. Rev Bras Fisioter 2009;13(13):267–73. 2007;27(1):7–11.