Purpose. To establish interrater and test–retest reliability of a clinical assessment of motor and sensory upper limb impairments in children with hemiplegic cerebral palsy aged 5–15 years. Method. The assessments included passive range of motion (PROM), Modified Ashworth Scale (MAS), manual muscle testing (MMT), grip strength, the House thumb and Zancolli classification and sensory function. Interrater reliability was investigated in 30 children, test–retest reliability in 23 children. Results. For PROM, interrater reliability varied from moderate to moderately high (correlation coefficients 0.48–0.73) and test–retest reliability was very high (>0.81). For the MAS and MMT, total score and subscores for shoulder, elbow, and wrist showed a moderately high to very high interrater reliability (0.60–0.91) and coefficients of >0.78 for test–retest reliability. The reliability for the individual muscles varied from moderate to high. The Jamar dynamometer was found to be highly reliable. The House thumb classification showed a substantial reliability and the Zancolli classification an almost perfect reliability. All sensory modalities had a good agreement. Conclusions. For all motor and sensory assessments, interrater and test–retest reliability was moderate to very high. Test–retest reliability was clearly higher than interrater reliability. To improve interrater reliability, it was recommended to strictly standardize the test procedure, refine the scoring criteria and provide intensive rater trainings.

Objective: To identify psychometrically sound and clinically feasible assessments of arm activities in children with hemiplegic cerebral palsy for implementation in research and clinical practice. Data sources: PubMed, CINAHL, Cochrane Library, Web of Science and reference lists of relevant articles were searched. Review methods: A systematic search was performed based on the following inclusion criteria: (1) evaluative tools at the activity level according to the International Classification of Functioning, Disability and Health; (2) previously used in studies including children with hemiplegic cerebral palsy aged 2-18 years; (3) at least one aspect of reliability and validity in children with cerebral palsy should be established. Descriptive information, psychometric properties and clinical utility were reviewed. Results: Eighteen assessments were identified of which 11 met the inclusion criteria: eight functional tests and three questionnaires. Five functional tests were condition-specific, three were generic. All functional tests measure different aspects of activity, including unimanual capacity and performance during bimanual tasks. The questionnaires obtain information about the child’s abilities at home or school. The reliability and validity have been established, though further use in clinical trials is necessary to determine the responsiveness. Conclusions: To obtain a complete view of what the child can do and what the child actually does, we advise a capacity-based test (Melbourne Assessment of Unilateral Upper Limb Function), a performance-based test (Assisting Hand Assessment) and a questionnaire (Abilhand-Kids). This will allow outcome differentiation and treatment guidance for the arm in children with cerebral palsy.

This study investigated interrater reliability and measurement error of the Melbourne Assessment of Unilateral Upper Limb Function (Melbourne Assessment) and the Quality of Upper Extremity Skills Test (QUEST), and assessed the relationship between both scales in 21 children (15 females, six males; mean age 6y 4mo [SD 1y 3mo], range 5–8y) with hemiplegic CP. Two raters scored the videotapes of the assessments independently in a randomized order. According to the House Classification, three participants were classified as level 1, one participant as level 3, eight as level 4, three as level 5, one participant as level 6, and five as level 7. The Melbourne Assessment and the QUEST showed high interrater reliability (intraclass correlation 0.97 for Melbourne Assessment; 0.96 for QUEST total score; 0.96 for QUEST hemiplegic side). The standard error of measurement and the smallest detectable difference was 3.2% and 8.9% for the Melbourne Assessment and 5.0% and 13.8% for the QUEST score on the hemiplegic side. Correlation analysis indicated that different dimensions of upper limb function are addressed in both scales.