PURPOSE: To establish: (i) item characteristics and item selection for the REPAS; (ii) internal consistency, inter-rater and test-retest reliability of the final REPAS version and its subtests; (iii) the association between the REPAS and selected other clinical scales of impairment and activity limitation. METHOD: Thirty-three neurological patients with central paresis. Two REPAS assessments with a one-week interval by two independent raters. Concurrent assessment of the Motricity Index, Box-and-Block test, Functional Ambulation Category, Timed walking, Barthel Index, Disability Rating Scale, Carer Burden Scale, and Hygiene Score. RESULTS: Twenty-six of 52 REPAS items fulfilled the item selection criteria. The final test version showed a high internal consistency, inter-rater and test-retest reliability (correlation coefficients: 0.87-0.97, no significant difference between raters or with test repetition). Reliability of the arm and leg subtests was substantial (correlation coefficients: arm subtest 0.63-0.98, leg subtest 0.56-0.96). REPAS scores were moderately associated with basic ADL competence and a carer's burden with arm or leg adductor spasticity. The REPAS, arm subtest scores, degree of arm paresis and gross manual dexterity showed a moderately high association. CONCLUSIONS: The Ashworth scale-based guidelines assured comparability of test administration and scoring. The REPAS is a reliable and valid summary rating scale for resistance to passive movement.
