We used hypothetical entry criteria typical of those used in clinical therapeutic trials to determine the patients who would have been eligible among those followed in a clinic-based study of multiple sclerosis (MS) in London, Ontario, between 1972 and 1984. For these patients, we determined the observed frequency of deterioration by 1 point on the disability status scale (DSS) of Kurtzke, which is the most feasible and frequently used endpoint in clinical trials. We calculated the number of patients required for a randomized clinical trial to detect a significant result (alpha = 0.05) with 80% or 90% power based on the observed rate of deterioration. To assess the linearity of the DSS, we determined the frequency of progression and staying time at each level of the DSS. Overall the frequency of progression was lower and the staying times were longer at higher levels of disability. There was considerable intrapatient as well as interpatient variation in saying time. These data have major implications for the design and conduct of clinical therapeutic trials in MS.