SETTING THE MINIMAL METRICALLY DETECTABLE CHANGE ON DISABILITY RATING-SCALES

Objective: To determine the minimal metrically detectable change (MMDC ) on rating scales by comparing two methods: (1) using the reliability coefficient derived from an external study in the calculation of the standard error of measurement (psychometric method); and (2) examining the variability of scores in a stable subsample from a longitudinal s tudy (empirical method). Design: Longitudinal survey. Setting: General community. Participants: Population-based representative sample of co mmunity-dwelling people older than 75 (n = 572). Main Outcome Measure: Disability as measured by the Functional Autonomy Measuring System (S MAF). Results: Using the psychometric method, a change in score of 3.7 on the SMAF was obtained using a reliability coefficient derived from an external test-retest study, and 5.2 using the reliability coeffici ent measured in a stable subsample of the longitudinal study. With the empirical method, a change of 5 points was established as the MMDC. C onclusion: The setting of the MMDC on a disability scale could be usef ul for calculating sample size or interpreting results from clinical t rials because it helps to establish the minimal clinically important d ifference, which should be equal to or larger than the MMDC. (C) 1997 by the American Congress of Rehabilitation Medicine and the American A cademy of Physical Medicine and Rehabilitation.