Psychometric and Clinimetric Properties of the Melbourne Assessment 2 in Children With Cerebral Palsy

Abstract

ObjectiveTo examine the psychometric and clinimetric properties of the Melbourne Assessment 2 (MA2), an outcome measurement that is increasingly used in clinical studies. DesignPsychometric and clinimetric study. SettingCommunity. ParticipantsSeventeen children with cerebral palsy (CP) from 5 to 12 years were recruited for the estimation of the test-retest reliability and minimal detectable change (MDC). Thirty-five children with CP were recruited to receive an 8-week intensive neurorehabilitation intervention to estimate the validity, responsiveness, and minimal clinically important difference (MCID). InterventionsThirty-five children with CP received upper limb neurorehabilitation programs for 8 weeks. Main Outcome MeasuresThe MA2 and the criterion measures, including the Bruininks-Oseretsky Test of Motor Proficiency, 2nd edition (BOT-2), the Box and Blocks Test (BBT), and the Pediatric Motor Activity Log–Revised (PMAL-R), were evaluated at pretreatment and posttreatment. ResultsThe MA2 has 4 subscales: range of motion, fluency, accuracy, and dexterity. The test-retest reliability of the MA2 is high (intraclass correlation coefficient, .92–.98). The significant relationships between the MA2 and BBT, BOT-2, and PMAL-R support its validity. The significance of paired t test results (P<.001) and large magnitudes of the standardized response mean (1.70–2.00) confirm the responsiveness of the MA2. The MDC values of the 4 subscales of the MA2 are 2.85, 1.63, 1.97, and 1.84, respectively, and the suggested MCID values of these 4 subscales are 2.35, 3.20, 2.09, and 2.22, respectively, indicating the minimum scores of improvement to be interpreted as both statistically significant and clinically important. ConclusionsThe study findings indicate that the MA2 has sound psychometric and clinimetric properties and is thus an adequate measurement for research and clinical applications.