upper limb activity measures for 5- to 16-year-old children with congenital hemiplegia: a systematic...

Upper limb activity measures for 5- to 16-year-old children withcongenital hemiplegia: a systematic review

ROSE GILMORE1 | LEANNE SAKZEWSKI1,2 | ROSLYN BOYD2

1 Brain Research Institute, Melbourne, Victoria, Australia. 2 Queensland Cerebral Palsy and Rehabilitation Research Centre, Royal Children's Hospital, University of Queensland,Brisbane, Australia.

Correspondence to Dr Roslyn Boyd, Queensland Cerebral Palsy and Rehabilitation Research Centre, Royal Children's Hospital, University of Queensland, Herston Road, Herston, Queensland 4029,Australia. E-mail: [email protected]

PUBLICATION DATA

Accepted for publication 2nd April 2009.Published online 7th October 2009.

LIST OF ABBREVIATIONSAHA Assisting Hand AssessmentMUUL Melbourne Assessment of Unilateral

Upper Limb FunctionQUEST Quality of Upper Extremity Skills TestSHUEE Shriners Hospital Upper Extremity

Evaluation

AIM This systematic review aimed to compare the validity, reliability, evaluative validity, and clinical

utility of upper limb activity measures for children aged 5 to 16 years with congenital hemiplegia.

METHOD Electronic databases were searched to identify assessments that measure upper limb

activity available for use and for which published validity and reliability data for the population are

obtainable. Assessment items were coded according to the International Classification of Functioning,

Disability and Health (ICF) categories to determine if at least 35% of the assessment items fell within

the activity component of the ICF. Assessments that met these criteria were included in the review.

RESULTS Thirty-eight measures were identified, and five met the inclusion criteria. The best mea-

sure of unimanual capacity was the Melbourne Assessment of Unilateral Upper Limb Function

(MUUL); however, the Shriners Hospital Upper Extremity Evaluation (SHUEE) and the Quality of

Upper Extremity Skills Test (QUEST) could also be considered, depending on the type of informa-

tion required. The performance-based measure of bimanual upper limb activity in children with

hemiplegia with the best psychometric properties was the Assisting Hand Assessment (AHA). The

ABILHAND-Kids is a parent-report, performance-based questionnaire with excellent clinical utility

and psychometric properties.

INTERPRETATION Clinicians may choose to use more than one of these measures to detect

changes in unimanual or bimanual upper limb activity.

The need to measure outcomes effectively in children with achronic health condition, such as cerebral palsy (CP), isimportant to guide and evaluate interventions, to monitor pro-gress, to provide families with objective information, and toguide health policy.1,2

The International Classification of Functioning, Disabilityand Health (ICF) provides a framework for measuring anddocumenting health outcomes at the body function and struc-ture level as well as in activities and participation.3 It also con-siders the influence of contextual factors, including personaland environmental factors, on a child’s health status.1 Out-come measures, therefore, need to address the multidimen-sional nature of the ICF not only by body structure andfunction, but also by activity and participation.

Increasingly, public bodies, such as the US Food and DrugAdministration, recommend the use of valid and reliable out-come measures capable of detecting clinically significantchange.4 Detecting change using an assessment tool that is notvalidated or proven reliable for the population, even if usedand reported in other clinical trials, does not provide meaning-ful information about the outcome of an intervention.

Children with hemiplegia usually have difficulty with grasping,reaching, releasing, and manipulating objects with the impaired

upper limb.5,6 Motor and sensory impairments can lead to chil-dren learning to perform tasks mainly with their unimpairedarm. This results in a secondary ‘developmental disregard’ orlearned non-use of the impaired upper limb.6,7 Therefore,intervention is often focused on improving both unimanual andbimanual function of the upper limbs during activity.

Assessments with excellent psychometric properties and clin-ical utility that specifically measure the use of a child’s impairedupper limb, as well as bimanual use, are necessary to ensureaccurate measurement of outcomes in clinical trials involvingchildren with congenital hemiplegia. While conducting a cur-rent upper limb rehabilitation trial, the authors recognized theneed for a systematic review of upper limb assessment measuresfor this population of children to guide selection of upper limbactivity measures. The age range of 5 to 16 years is specified inline with the inclusion criteria of the current clinical trial.

Recent systematic reviews of measurement tools for chil-dren with CP have focused on activity limitation,8 activity andparticipation,9 participation,2 and self-concept.10 A recentreview of measures of activity limitation focused on evaluativemeasures of mobility and gross motor function.8 In contrast,the current systematic review aims to compare the validity,reliability, evaluative validity, and clinical utility of upper limb

14 DOI: 10.1111/j.1469-8749.2009.03369.x ª The Authors. Journal compilation ª Mac Keith Press 2009

DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY SYSTEMATIC REVIEW

activity measures for children aged 5 to 16 years with congeni-tal hemiplegia. For the purpose of this review, activity wasdefined as the execution of a task or action by an individual.3

METHODSearch strategyA systematic search was undertaken of computerized databases,including Medline (1966–August 2008), CINAHL (1982–August 2008), EMBASE (1988–August 2008), AMED (1985–August 2008), PsychINFO (1806–August 2008), the CochraneLibrary (August 2008), PEDro (August 2008), and OT Seeker(August 2008). The search strategy used the medical subjectheading (MeSH) terms and text words for ‘cerebral palsy’ and‘hemiplegia’. These were combined with search terms to limitthe findings to the target age group. MeSH terms and textwords for ‘outcome assessment’, ‘activity,’ and ‘upper limb’were used to focus search results on activity measures of theupper limb. Targeted reference scanning and citation trackingof key articles were also used to minimize the chance of miss-ing key studies.

Inclusion ⁄ exclusion criteriaTo be included, an assessment had to meet the following a pri-ori inclusion criteria: (1) it measures upper limb activity(including those which evaluate a combination of conceptsfalling within the activity and body function components ofthe ICF); (2) at least 35% of the assessment items should eval-uate upper limb activity according to the ICF definition; (3) itmust have published validity and reliability data for childrenaged 5 to 16 years with congenital hemiplegia; (4) it measureseither bimanual or unilateral use of upper limbs and (5) theassessment is available for use.

Assessment tools were excluded if they were (1) not pub-lished in English (as a result of lack of translation services); (2)primarily assessed participation, body structures, body func-tions, school functioning, self-care, or health-related quality oflife; (3) were a classification measure; or (4) were individual-ized, goal-setting tools. Classification measures discriminateand categorize and do not involve the same detailed explora-tion of upper limb activity which assessments allow11 andwere, therefore, excluded. Individualized, goal-setting toolswere not included as these assessments can measure compo-nents related to different body structures, function, daily livingskills, and participation, rather than necessarily focusing onupper limb activity.

The titles and abstracts of articles were screened by the firstauthor (RG). One key paper which had adequate detail todetermine inclusion, as well as the score form for each assess-ment, was then reviewed by the first author and the thirdauthor (RB). If an assessment measured any component ofupper limb activity and fulfilled the third inclusion criterionfor published validity and reliability data for this population,then each item in the assessment was linked to the most spe-cific ICF category, according to the updated linking rules pre-sented by Cieza et al.12 Conflicting viewpoints were discusseduntil consensus was reached. Once items were linked to anICF category, the percentage of items that measured upper

limb activity and fell within the mobility chapter of the activityand participation component of the ICF was calculated. If35% of items fell within this chapter, the assessment wasincluded.

To illustrate the method for coding items, two items fromthe Quality of Upper Extremity Skills Test (QUEST) assess-ment were selected. The first item was ‘shoulder flexion’. Thisitem was coded b7600 (third-level category: control of simplevoluntary movement function; second-level category: controlof voluntary movement function; chapter: neuromusculoskel-etal and movement-related functions; and component: bodyfunction). This item did not, therefore, contribute to the over-all percentage of items measuring upper limb activity. In con-trast, the item ‘grasp of 1-inch cube’ was coded d4401 (third-level category: grasping; second-level category: fine hand use;chapter: mobility; and component: activities and participa-tion). This item was included in the calculation for percentageof items measuring upper limb activity.

For some assessments, it was difficult to determine what anitem consisted of. It was decided to include all items that con-tributed to the overall score for the assessment. For example,all range of motion and spasticity measurements, history-basedactivities of daily living information, and patient ⁄ family goalsin the first part of the Shriners Hospital Upper ExtremityEvaluation (SHUEE) assessment were excluded as they didnot contribute to overall scores. This helped to distinguishbetween subjective or descriptive components of assessmentsand quantitative measures of upper limb function which couldbe subjected to psychometric testing. The title of potentialassessments was then used for a further search in the databasesto ensure that no article was missed. Relevant assessment man-uals were sourced.

Data extraction and quality assessmentA modified version of the CanChild Outcome Measures RatingForm13 was used to extract descriptive information and gradethe psychometric properties and clinical utility of the assess-ment tools. Recent reviews of participation outcome measuresin CP and neuromotor assessments for preterm infants utilizedthis form.2,14 In another recent review of gross motor activitylimitation measures for children with CP, the authors devisedtheir own quality scale.8 The Outcome Measures Rating Formincorporates the ICF framework and was designed to rate thequality and clinical utility of outcome measures and it was,therefore, deemed the most appropriate scale to evaluate theassessment tools.

The characteristics (for both clinical and research settings)of the assessment tools extracted included the target popula-tion, age range, purpose of tool, format, content of the assess-ment, number of scales ⁄ items, time to administer, cost andavailability of the manual, and training required.

An assessment was considered valid if it measured what it issupposed to measure.15 Information on validity included con-tent, construct, criterion, and evaluative validity or responsive-ness. Evaluative validity or responsiveness is the ability todetect minimal clinically important change over time.16 Reli-ability measures the degree to which a test is stable, measuring

Systematic Review 15

a construct in a reproducible and consistent way.13 Internalconsistency, interrater reliability, intrarater reliability and test–retest reliability were evaluated.

There is little consensus in the literature regarding accept-able standards for reliability coefficients, and terminology isoften used interchangeably. When interpreting the reliabilitycoefficients, Portney and Watkins15 suggested that valuesbelow 0.50 represent poor reliability, values of 0.50–0.75 rep-resent moderate reliability, and values above 0.75 representgood reliability. They present these values as a guideline onlyand emphasize that the degree of acceptable reliabilitydepends on how the results will be applied; for instance, theypropose a value of at least 0.90 when interpreting results foran individual. The above guidelines can also be applied tokappa statistics, which are appropriate for nominal data.15 TheCanChild Outcome Measures Rating Form describes reliabilitycoefficients above 0.8 as excellent, from 0.60 to 0.79 as ade-quate, and <0.60 as poor.13 Given the lack of consensus in theliterature, this systematic review used the guidelines outlinedin the CanChild Outcome Measures Rating Form, as well asconsidering the need for values to be above 0.90 in order tointerpret results for an individual.

RESULTSThirty-six assessments were identified by the search strategy,of which five met the predetermined inclusion criteria. Theincluded measures were the ABILHAND-Kids,17 the Assist-ing Hand Assessment (AHA),18 the Melbourne Assessment ofUnilateral Upper Limb Function (MUUL),19 the QUEST20

and the SHUEE.21 The characteristics of the included assess-ments are summarized in Table I. The combination of the ori-ginal yield for these five assessments and additional title-basedsearching resulted in 17 papers reporting on these assessments.Thirty-one assessments were excluded. Table SI (supportinginformation published online) summarizes reasons for exclu-sion.

The ABILHAND-Kids, the AHA, the MUUL, theQUEST, and the SHUEE were specifically designed forassessment of children with CP or a unilateral disability. TheABILHAND-Kids and the AHA measure performance, i.e.

what a child does. In contrast, the MUUL, the QUEST, andthe SHUEE measure a child’s capacity, or maximum capabil-ity. The AHA and the QUEST are suitable for children aged18 months and over, the MUUL for children 30 months andover, the SHUEE for children 3 years and over, and theABILHAND-Kids for children 6 years and over.

In the case of all of the selected tools, published studiesreporting data for children with congenital hemiplegia areavailable. In addition, studies reporting data for the AHAincorporate children with unimanual impairment, includingbrachial plexus palsy and congenital hemiplegia. The studydescribing data for the SHUEE included only children withhemiplegic CP. The validity and reliability of the MUUL,the ABILHAND-Kids, and the QUEST were investigated inchildren with various distribution and motor types of CP.

Validity and evaluative validityEvidence for the validity and evaluative validity of the includedassessments is reported in Table II. Most assessments had ade-quate content validity, although evidence was not reported forthe SHUEE. All assessments also showed adequate constructvalidity. The MUUL established criterion validity by compar-ison with expert clinical assessment. The QUEST and theSHUEE attempted to establish criterion validity with severalother functional assessments; however, no clear criterion stan-dard is accepted in the literature.

ReliabilityDetails of the reliability of assessments are reported inTable III. Internal consistency has been claimed for the ABIL-HAND-Kids, the AHA, and the MUUL. Strong inter- andintrarater reliability were reported for the AHA, the MUUL,and the SHUEE. The QUEST reported strong interraterreliability and variable intrarater reliability evidence. TheABILHAND-Kids reported evidence for test–retest reliability;however, it only reported Pearson’s r correlation. The intra-class correlation coefficient reflects both correlation andagreement and has, therefore, become the preferred statistic.15

Excellent test-retest reliability has been claimed for theMUUL, the QUEST, and the AHA.

Table I: Characteristics of included assessments

Measure Target populationType of activitymeasured

Percentage of items inactivity domain of ICF

Primarypurpose

Performanceor capacity

ABILHAND-Kids17 Children with CP aged 6–15y Manual ability 81 Discriminative PerformanceAHA18 Children with a unilateral

disability aged 18mo–12yBimanual play 82 Discriminative

EvaluativePerformance

MUUL40 Children with CP or neurologicalimpairment aged 30mo–15y40,a

Unimanual 46 DiscriminativeEvaluative

Capacity

QUEST20 Children with CP aged 18mo– 8y Unimanual (each armscored separately)

35 Evaluative Capacity

SHUEE21 Children with hemiplegic CP aged 3–18y Unimanual 81 DiscriminativeEvaluative

Capacity

aA revised version of the Melbourne Assessment of Unilateral Upper Limb Function (MUUL) capable of assessing children aged 30 months to15 years was presented at the Australasian Academy of Cerebral Palsy & Developmental Medicine conference in 2008. ICF, InternationalClassification of Functioning, Disability and Health; AHA, Assisting Hand Assessment; QUEST, Quality of Upper Extremity Skills Test; SHUEE,Shriners Hospital Upper Extremity Evaluation.

16 Developmental Medicine & Child Neurology 2010, 52: 14–21

Clinical utilityThe clinical utility of the included assessments is reported inTable IV. Administration time varied across the assessments,from 10 to 45 minutes. Assessor training is required for theAHA; for the other assessments training is either not requiredor not specified. The AHA’s method of administrationincludes therapist observation of a child’s performance inupper limb activity, while the QUEST, the SHUEE, and theMUUL involve a therapist systematically administering items.The ABILHAND-Kids is administered by parent-completedquestionnaire. The MUUL, the AHA, and the SHUEE usevideotaped assessments that are scored later.

DISCUSSIONThis systematic review identified five assessments that measuredifferent components of upper limb activity and are suitablefor use with children with congenital hemiplegia. A combina-tion of assessments would be required to measure both uni-manual and bimanual upper limb activity in this group ofchildren. Researchers and clinicians searching for valid andreliable assessment tools to measure upper limb activity inchildren with congenital hemiplegia are faced with many mea-sures to choose from; however, in only a few cases are ade-quate psychometric properties reported in the literature.Assessments such as the Jebsen–Taylor Test of Hand Func-

Table II: Evidence of content, construct, criterion, and evaluative validity

Assessment Content Construct Criterion Evaluative

ABILHAND-Kids17 Based on existing scalesand expert advice17

Rasch model

Significant relationshipwith type of schooleducation, CP andGMFCS17

Invariance verified acrossdemographic and clinicalsubgroups of childrenwith CP41

NA Person separation reliabilityestimate: r=0.94 (n=36)17

AHA18 Developed by experts infield18

Rasch model(unidimensionalitydemonstrated for 91% ofresponses in 22-itemversion)36

Discriminates betweenchildren with differentlevels of hand function(separation value=6.16)and levels of impairmentnot related to age18,36

NA Person separation reliabilityestimate: r=0.97Ability to separate and distributeperson ability measures whichindicates sensitivity to change(18mo–12y, n=303, 409assessments)36

SDD was less than four rawscores (n=55, 18mo–12y)26

MUUL40 Literature review, review ofexisting assessments,workshops withexperienced occupationaltherapist40

Significant correlationswith MUUL and PEDI onself-care (Spearman’srho=0.939), mobility(0.783) and total functionalskills (0.718)69

Four experienced cliniciansassessed according totheir usual method. ICCbetween clinicians=0.92.Average correlationbetween clinician scoreand MUUL was 0.87(n=11)19

SDD=8.9% (5–8y, n=21)32

QUEST20 Literature review,discussion with cliniciansand experts

Correlations betweentherapist’s judgement ofchild’s hand function andQUEST scores for rightand left hands (n=71)20

Significant correlationswith PDMS-FM and MUUL(r=0.84 for total score ofQUEST and PDMS-FM;r=0.83 for total score ofQUEST and MUUL)32

Smallest difference of clinicalsignificance in casting trial was4.89 score units (p<0.03)43

SDD=7.1% (total test score, 5–8y,n=21)32

SHUEE21 NA Significant change in meandynamic positionalanalysis score for the wristpost-tendon transfer(n=18)21

Fair correlation betweenself-care scaled score ofPEDI and spontaneousfunctional analysis scoreof SHUEE (r=0.47)Inverse correlationbetween non-dominanttotal time of the JTT andspontaneous functionalanalysis section ofSHUEE21

NA

GMFCS, Gross Motor Function Classification System; ICC, intraclass correlation coefficient; NA, no data available; PEDI, Pediatric Evaluation ofDisability Inventory; PDMS-FM, Peabody Developmental Motor Scales – Fine Motor composite; SDD, smallest detectable difference; JTT, Jebson–Taylor Test of Hand Function; AHA, Assisting Hand Assessment; MUUL, Melbourne Assessment of Unilateral Upper Limb Function; QUEST,Quality of Upper Extremity Skills Test; SHUEE, Shriners Hospital Upper Extremity Evaluation.


tion22 and the Bruininks–Oseretsky Test of Motor Proficiency(2nd edn)23 are frequently used as outcome measures in clini-cal trials with children with congenital hemiplegia; however,there is no published study of validity and reliability in thispopulation. In the case of the Peabody Developmental MotorScales (2nd edn),24 reliability data are available for a smallsample of children with congenital hemiplegia. However,these scales were not developed specifically for children with

either CP or a unilateral impairment and, lacking validity data,failed to meet the third inclusion criterion for this review.Lack of adequate validity and reliability data for these assess-ments has implications for interpreting response to treatment,and further research is required to investigate these assess-ments in children with congenital hemiplegia.

When designing the second inclusion criterion for thisreview, we initially considered that included assessments

Table III: Evidence of reliability

Assessment

Internal consistency(Cronbach’s coefficientalpha) Interrater (ICCs) Intrarater (ICCs) Test–retest

ABILHAND-Kids17 0.94 (person separationreliability estimate; n=113,6–15y, variety of types ofCP)17

NA NA r=0.91 (n=36)17

AHA18 0.97 (person separationreliability estimate: 409assessments, 18mo–12y8mo, hemiplegic CP andOBPP)36

0.98 (two raters) (18mo–5y 11mo;16 children with congenitalhemiplegia and two with OBPP)25

0.97 (20 raters) (19mo–4y 2mo;six children with congenitalhemiplegia and two with OBPP)25

0.99 (20 raters)25 Small kids AHA ICC=0.99;school kids AHA ICC= 0.98;alternate forms ICC=0.99between small and schoolkids’ versions and 0.98between board games26

MUUL40 0.96 (n=20, 5–16y, varietyof types of CP)27

Trained and untrainedraters=0.9928

0.95 (15 raters, total test scores,n=20)27

LOA=14%27

LOA with training=5%29

Original 12-item test: kappa=0.65(first scoring, two raters) and 0.72(second scoring) (n=20, variety oftypes of CP)19

0.99 (12 trained raters)0.98 (12 untrained raters)(5y 5mo–12y)28

0.87 (first session, five raters,n=10)0.88 (second session, five raters,n=10)30

0.97 (20 raters, n=20, totaltest scores, variety oftypes of CP)27

LOA=12%27

LOA with training=3.5%29

Original 12-item test:kappa=0.82 for firsttherapist and 0.79 forsecond therapist (n=20,variety of types of CP)19

0.97–0.99 (five raters, 10children aged 10–17y)30

0.97 (two raters, 21children aged 5–8y)32

CCC=0.98 for first rater and0.97 for second rater27

QUEST20 NA 0.95 (two raters, total test scores,n=16)20

0.96 (two raters, total test scores,n=71, 39% hemiplegia)20

0.90 (three raters, total testscores, n=1720

0.72–0.90 (Spearman’scorrelation coefficient, usedWilcox signed-rank test toconfirm no systematicdifferences, three raters, total testscores, n=21, 29% hemiplegia)31

0.96 (two raters, total test scores,n=21, 100% hemiplegia32

0.91 (two raters, total test scores,n=26, 5 with hemiplegia)33

0.69: assessor 1 and 0.89:assessor 2 (n=26, 2–13y)33

0.95: assessor 1 and 0.63:assessor 2 (Spearman’scorrelation coefficient,n=21, 2y–4y6mo)31

0.95 (ICC for total testscore)20

0.85–0.94 (Spearman’scorrelation coefficient)31

SHUEE21 NA 0.90 (Spontaneous functionalanalysis)0.89 (Dynamic positionalanalysis)Weighted kappa=1.00 (Grasp &release)(four raters, n=11, 100%hemiplegia))21

r=0.99 (spontaneousfunctional analysis)r=0.98 (dynamic positionalanalysis)Student paired t-test –absolute value of meandifference between 2scoring sessionsstatistically significant butnot clinically important(three raters, n=11)21

NA

CCC, concordance correlation coefficient; ICC, intraclass correlation coefficient; LOA, limits of agreement; NA, no data available; OBPP, obstetricbrachial plexus palsy; r, Pearson’s product–moment correlation; AHA, Assisting Hand Assessment; QUEST, Quality of Upper Extremity Skills Test;MUUL, Melbourne Assessment of Unilateral Upper Limb Function; SHUEE, Shriners Hospital Upper Extremity Evaluation.


should logically measure more items falling within the activitycomponent of the ICF than in the body function component,i.e. a cut-off point of 50%. However, several unimanual mea-sures frequently used by clinicians to evaluate upper limbactivity in this population assess both upper limb activity andbody function according to the ICF and failed to meet the50% cut-off. To ensure that this review was useful for clini-cians and researchers measuring both unimanual and bimanualperformance, the cut-off point for the percentage of activityitems included was reduced to 35%.

Validity and reliability psychometric data for children withcongenital hemiplegia must have been published to meet theinclusion criteria for this review. This proved a necessary crite-rion to identify the assessments with sound psychometricproperties for this particular population. The ABILHAND-

Kids, the AHA, the QUEST, and the MUUL were identifiedas valid measures for children with congenital hemiplegia.With no criterion standard for measuring upper limb activityavailable, the MUUL was the only assessment to establish cri-terion validity. An experienced group of clinicians comparedtheir usual upper limb assessment with the MUUL.19 Thepaper reporting psychometric data for the SHUEE describedthe concurrent validity evidence of the SHUEE in comparisonwith the Pediatric Evaluation of Disability Inventory and theJebsen–Taylor Test of Hand Function. While presented inthe criterion validity column of Table III, these assessmentsare not universally accepted as criterion standard measures ofupper limb activity.

The AHA and the MUUL have established excellent inter-nal consistency, test–retest reliability, and inter- and intrarater

Table IV: Clinical utility of included upper limb activity measures

AssessmentFormat ofadministration

Administrationtime (min)

Assessortraining Manual ⁄ equipment Scales ⁄ items Scoring

ABILHAND-Kids17 Parent-completedquestionnaire

NS NR Free onlineNo equipmentrequired

21 activityquestions: rated asimpossible, difficultor easy

Online analysis.Calibrated in logits

AHA18 Therapist-led playsession. Videorecorded, scoredlater

10–15 Required Received duringtraining (AU$860)Kit (AU$550)

22 items assessingassisting handgeneral use, armuse, grasp-release,fine motoradjustment,coordination andpace

Raw scoresconverted intopercentage score.Calibrated in logits

MUUL40 Therapistadministers itemsin standardizedprocedure.Video recorded,scored later

30 Recommendedbut NR

Manual and kit(AU$720)

16 upper limbactivity items ratedon quality and taskachievement

Raw scoresconverted topercentages

QUEST20 Therapistadministers items,non-standardizedprocedure

30–45 NR, butrecommendedby otherpapers33,44

Free online 34 items, fourdomains:dissociatedmovements, grasp,protectiveextension, weightbearingthree subjectiverating scales notincluded in totalscore

Raw scoresconverted topercentages

SHUEE21 Therapistadministers itemsin standardizedprocedure.Video recorded,scored later

15 NR Free online Two sections.First measuresactive and passiverange of motion inupper limbs,spasticity, ADLperformance andgoals.Second section hasthree components:spontaneousfunctional analysis,dynamic positionalanalysis and graspand releaseanalysis

Percentages of themaximum possiblescore for eachsection

NR, not required; NS, not specified; AHA, Assisting Hand Assessment; MUUL, Melbourne Assessment of Unilateral Upper Limb Function; QUEST,Quality of Upper Extremity Skills Test; SHUEE, Shriners Hospital Upper Extremity Evaluation.


reliability.18,19,25–30,32,36 The ABILHAND-Kids assessmentshows adequate internal consistency and test–retest reliabil-ity.17 The SHUEE and the QUEST demonstrate excellent in-terrater reliability.20,21,31–33 In the case of the QUEST,variable intrarater reliability and strong test–retest reliabilityhave been reported.20,31,33 The discrepancy between strong in-terrater reliability and variable intrarater reliability found inthe two studies related to the QUEST may be a result of smallnumbers of raters and depend on which round of assessmentscores was used to determine the interrater reliability. Data forthe SHUEE showed strong intrarater reliability using Pear-son’s correlation coefficient to demonstrate correlation andStudent’s paired t-tests to ensure that no systematic differencesoccurred.21

The AHA and ABILHAND-Kids assessments were bothdeveloped according to the Rasch measurement model, whichis becoming a preferred way to construct tests for use in reha-bilitation.34,35 Scales developed using a Rasch model are usefulwhen measuring children across the spectrum of severitybecause of the requirement that items form a unidimensionallinear scale capable of differentiating between children withvaried abilities. The AHA and the ABILHAND-KIDS reportgood validity and reliability with items forming a unidimen-sional construct, reliably separating and spreading the mea-sures of ability along the scale.17,18,25,26,36 The reportedperson separation reliability estimates also indicate sensitivityto change.36 An intervention study of modified constraint-induced movement therapy using the AHA showed a largeeffect size in children aged 18 months to 4 years.37 This isoutside the age range of the current review, but a promisingindicator for future studies. A further version of the AHA isbeing developed for adolescents but there is no publishedinformation available for this age group at present (Krumlin-de-Sundholm L, personal communication 2008).

Clinical utility needs to be considered as well as the psycho-metric properties of assessment tools. The ABILHAND-Kidshas excellent clinical utility, being available free online, com-plete with online analysis of data. It is quick to administer andis completed by parents, giving a different perspective on chil-dren’s manual ability in their everyday life. The MUUL man-ual specifies that it takes 30 minutes to administer and30 minutes to score. The AHA manual does not specify theamount of time needed to score, although, in the authors’experience, the AHA takes approximately 1 hour. TheQUEST can be administered and scored within 45 minutes.The SHUEE takes 15 minutes to administer and scoring timewas not reported in the manual. The AHA is the only assess-ment that requires formal training and so is more expensive.Both the AHA and the MUUL are accompanied by manualsdetailing specific procedures for administration, scoring, inter-pretation, and presenting evidence for the psychometric prop-erties of the test. Free online manuals are available for boththe SHUEE and the QUEST. The SHUEE manual providesclear instructions and is linked to the paper presenting psycho-metric data. The QUEST manual presents initial reliabilityand validity data but lacks some clarity regarding administra-tion and scoring of some items.

Clinicians may select an upper limb assessment tool to mea-sure a specific aspect of upper limb activity. For clinicians whowant to measure a change in unimanual function as a result ofspasticity management or following surgery, the MUUL, theQUEST, or the SHUEE may be useful. Of these assessments,the MUUL presented the most thoroughly tested psychomet-ric properties.

A clinician may also want to measure speed and dexterity,and there is no known test of speed and dexterity validated forchildren with CP. The Jebsen–Taylor Test of Hand Func-tion22 has frequently been used as a measure of speed and dex-terity; however, psychometric data reported for children withcongenital hemiplegia are not available. The AHA measureshow a child uses his or her impaired hand in bimanual func-tion in a naturalistic environment. Clinicians may want toinvestigate this skill after intensive training of the impairedlimb using therapies such as modified constraint-inducedmovement therapy or intensive bimanual training.

There are several potential limitations to this review. Arti-cles were included only if they were published in English and,therefore, some assessments may have been excluded. A num-ber of measures, such as the Assessment of Motor and ProcessSkills38 and the Wee-FIM,39 were excluded as their primarypurpose is to measure self-care. They may also measureaspects of upper limb activity, but this is not the primary pur-pose of these assessments. There were also several individual-ized, goal-setting tools which can be used to measure changein upper limb activity which were not included because thiswas not their primary purpose. Further information on theirpsychometric properties and clinical utility can be found else-where.2

CONCLUSIONThis systematic review identified five measures that assessupper limb activity in children with congenital hemiplegia.Published psychometric properties are available for all of thesemeasures in this population of children. When measuring uni-manual capacity, the MUUL had the most thoroughly testedvalidity and reliability evidence. For measurement of bimanualperformance, the AHA had the most thoroughly tested validityand reliability evidence. This assessment, which uses naturalis-tic observation, combined with the MUUL, could be useful tomeasure the relationship between changes in unimanualcapacity and bimanual performance. Both the MUUL and theAHA can be used to discriminate between individuals andevaluate change over time. The ABILHAND-Kids is a parentreport and shorter assessment tool designed with a discrimina-tive purpose, which provides a measure of overall manual abil-ity from a parent’s perspective. Further studies of theevaluative validity of all tools are required to ensure clinicallymeaningful change of activity limitations is being measured inresponse to intervention.

ACKNOWLEDGMENTSINCITE project grant: National Health Medical Research Council

Grant (368500); NHMRC Career Development Award (473840);

Smart State Fellowship from Queensland Government; Royal Chil-


dren’s Hospital Foundation, Queensland for RB; NHMRC Dora

Lush Fellowship (384488) for LS.

SUPPORTING INFORMATIONAdditional supporting information may be found in the online version

of this article:

Table SI: Excluded assessments

Please note: Wiley-Blackwell are not responsible for the content or

functionality of any supporting materials supplied by the authors. Any

queries (other than missing material) should be directed to the corre-

sponding author for the article.

REFERENCES

1. Majnemer A, Mazer B. New directions in the outcome evalu-

ation of children with cerebral palsy. Semin Pediatr Neurol

2004; 11: 11–7.

2. Sakzewski L, Boyd R, Ziviani J. Clinimetric properties of

participation measures for 5- to 13-year-old children with

cerebral palsy: a systematic review. Dev Med Child Neurol

2007; 49: 232–40.

3. World Health Organization. International Classification of

Functioning, Disability and Health. Geneva: World Health

Organization, 2001.

4. Cano S, Hobart J. Watch out, watch out, the FDA are about.

Dev Med Child Neurol 2008; 50: 408–9.

5. Eliasson A-C, Gordon AM, Forssberg H. Basic co-ordina-

tion of manipulative forces of children with cerebral palsy.


6. Hoare B, Imms C, Carey L, Wasiak J. Constraint-induced

movement therapy in the treatment of the upper limb in chil-

dren with hemiplegic cerebral palsy: a Cochrane systematic

review. Clin Rehabil 2007; 21: 675–85.

7. Charles J, Gordon AM. A critical review of constraint-

induced movement therapy and forced use in children with

hemiplegia. Neural Plast 2005; 12: 245–61.

8. Harvey A, Robin J, Morris ME, Graham HK, Baker R.

A systematic review of measures of activity limitation for

children with cerebral palsy. Dev Med Child Neurol 2008; 50:

190–8.

9. Morris C, Kurinczuk JJ, Fitzpatrick R. Child or family

assessed measures of activity performance and participation

for children with cerebral palsy: a structured review. Child

Care Health Dev 2005; 31: 397–407.

10. Dunn N, Shields N, Taylor NF, Dodd KJ. A systematic

review of the self-concept of children with cerebral palsy and

perceptions of parents and teachers. Phys Occup Ther Pediatr

2007; 27: 55–71.

11. Eliasson A-C, Krumlinde-Sundholm L, Rosblad B, et al.

Using the MACS to facilitate communication about manual

abilities of children with cerebral palsy [comment]. Dev Med

Child Neurol 2007; 49: 156–7.

12. Cieza A, Geyh S, Chatterji S, Kostanjsek N, Ustun B, Stucki

G. ICF linking rules: an update based on lessons learned.

J Rehabil Med 2005; 37: 212–8.

13. Law M. Outcome Measures Rating Form. 2004. Available

from: http://www.canchild.ca/en/canchildresources/resources/

measrate.pdf (accessed 14 January 2008).

14. Spittle AJ, Doyle LW, Boyd RN. A systematic review of the

clinimetric properties of neuromotor assessments for preterm

infants during the first year of life. Dev Med Child Neurol

2008; 50: 254–66.

15. Portney LG, Watkins MP. Foundations of Clinical

Research: Application to Practice, 2nd edn. Upper Saddle

River, NJ: Prentice-Hall, 2000.

16. Guyatt G, Walter S, Norman G. Measuring change over

time: assessing the usefulness of evaluative instruments.

J Chronic Dis 1987; 40: 171–8.

17. Arnould C, Penta M, Renders A, Thonnard J-L. ABIL-

HAND-Kids: a measure of manual ability in children with

cerebral palsy. Neurology 2004; 63: 1045–52.

18. Krumlinde-Sundholm L, Eliasson A-C. Development of the

assisting hand assessment: a Rasch-built measure intended

for children with unilateral upper limb impairments. Scand J

Occup Ther 2003; 10: 16–26.

19. Johnson LM, Randall MJ, Reddihough DS, Oke LE,

Byrt TA, Bach TM. Development of a clinical assessment of

quality of movement for unilateral upper-limb function.


20. DeMatteo C, Law M, Russell D, Pollock N, Rosenbaum P,

Walter S. Quality of Upper Extremity Skills Test. Hamilton:

McMaster University, Neurodevelopmental Clinical

Research Unit, 1992.

21. Davids JR, Peace LC, Wagner LV, Gidewall MA, Blackhurst

DW, Roberson WM. Validation of the Shriners hospital for

children upper extremity evaluation (SHUEE) for children

with hemiplegic cerebral palsy. J Bone Joint Surg Am 2006;

88: 326–33.

22. Taylor N, Sand P, Jebsen R. Evaluation of hand function in

children. Arch Phys Med Rehabil 1973; 54: 129–35.

23. Bruininks RH, Bruininks BD. Bruininks–Oseretsky Test of

Motor Proficiency, 2nd edn. Circle Pines: AGS Publishing,

2005.

24. Folio R, Fewell RR. Peabody Developmental Motor Scales,

2nd edn. Austin: Pro-ed, 2000.

25. Holmefur M, Krumlinde-Sundholm L, Eliasson A-C. Interr-

ater and intrarater reliability of the assisting hand assessment.

Am J Occup Ther 2007; 61: 79–84.

26. Holmefur M, Aarts P, Hoare B, Krumlinde-Sundholm L.

Retest and alternate forms reliability of the assisting hand

assessment. J Rehabil Med (forthcoming)

27. Randall M, Carlin JB, Chondros P, Reddihough D.

Reliability of the Melbourne assessment of unilateral upper

limb function. Dev Med Child Neurol 2001; 43: 761–67.

28. Cusick A, Vasquez M, Knowles L, Wallen M. Effect of rater

training on reliability of Melbourne assessment of unilateral

upper limb function scores. Dev Med Child Neurol 2005; 47:

39–45.

29. Corn K, Imms C, Timewell G, et al. Impact of second skin

Lycra splinting on the quality of upper limb movement in

children. Br J Occup Ther 2003; 66: 464–72.

30. Mackey AH, Miller F, Kilgour G, et al. Reliability of the

Melbourne assessment in children with hemiplegia. Dev Med

Child Neurol 2006; 8:(Suppl. 106) 23–24.

31. Haga N, van der Heijden-Maessen HC, van Hoorn JF,

Boonstra AM, Hadders-Algra M. Test–retest and inter- and

intra-reliability of the quality of the upper-extremity skills

test in preschool-age children with cerebral palsy. Arch Phys

Med Rehabil 2007; 88: 1686–9.

32. Klingels K, De Cock P, Desloovere K, et al. Comparison of

the Melbourne assessment of unilateral upper limb function

and the quality of upper extremity skills test in hemiplegic

CP. Dev Med Child Neurol 2008; 50: 1–6.

33. Sorsdahl AB, Moe-Nilssen R, Strand LI. Observer reliability

of the gross motor performance measure and the quality of

upper extremity skills test, based on video recordings. Dev

Med Child Neurol 2008; 50: 146–51.

34. Davies P, Gavin W. Measurement issues in treatment

effectiveness studies. Am J Occup Ther 1999; 49: 645–

52.

35. Velozo C, Kielhofner G, Lai J. The use of Rasch Analysis to

produce scale-free measurement of functional ability. Am J

Occup Ther 1998; 53: 83–90.

36. Krumlinde-Sundholm L, Holmefur M, Kottorp A, Eliasson

A. The Assisting Hand Assessment: current evidence of valid-

ity, reliability, and responsiveness to change. Dev Med Child

Neurol 2007; 49: 259–64.

37. Eliasson A-C, Krumlinde-Sundholm L, Shaw K, Wang C.

Effects of constraint-induced movement therapy in young

children with hemiplegic cerebral palsy: an adapted model.


38. Fisher AG. Assessment of Motor and Process Skills.

Vol. 1: Development, Standardization, and Administration

Manual, 5th edn. Fort Collins, CO: Three Star Press, Inc.,

2003.

39. Chen C, Bode R, Granger C, Heinemann A. Psychometric

properties and developmental differences in children’s

ADL item hierarchy. Am J Phys Med Rehabil 2005; 84:

671–9.

40. Randall M, Johnson L, Reddihough D. The Melbourne

Assessment of Unilateral Upper Limb Function: Test

Administration Manual. Melbourne: Royal Children’s Hospi-

tal, 1999.

41. Arnould C. Hand functioning in children with cerebral

palsy (Dissertation). Universite catholique de Louvain,

2006.

42. Bourke-Taylor H. Melbourne assessment of unilateral upper

limb function: construct validity and correlation with the

pediatric evaluation of disability inventory. Dev Med Child

Neurol 2003; 45: 92–6.

43. Law M, Cadman D, Rosenbaum P, Walter S, Russell D,

DeMatteo C. Neurodevelopmental therapy and upper-

extremity inhibitive casting for children with cerebral palsy.


44. Hickey A, Ziviani J. A review of the quality of upper extremi-

ties skills test (QUEST) for children with cerebral palsy. Phys

Occup Ther Pediatr 1998; 18: 123–35.


upper limb activity measures for 5- to 16-year-old children with congenital hemiplegia: a systematic...

Documents