Measurement tools assessing disability in MS patients are essential to evaluate the effect of the growing number of drug treatments and to measure rehabilitation outcome. However, the most widely used EDSS (Expanded Disability Status Scale) has been shown to be insensitive to minor changes. The assessment of disability with simple measures such as the Rivermead Mobility Index (RMI) and gait speed timed over 10 m has been suggested. Two hundred MS patients attending a rehabilitation centre were assessed at the beginning and at the end of a four-week period (median 28.2 days; interquartile range (IQR) 1.8 days) using these two measures as well as the EDSS and the Ambulation Index (AI). The RMI was the most sensitive scale and measured changes in 39% of the patients, whereas AI changed in 18.5% and EDSS in only 7.5% of the patients. The RMI assessed more precisely ambulatory patients needing walking aids, for which EDSS proposes only two options (EDSS 6.0/6.5). To measure the variability of the 10 m walking time, 25 of the 115 ambulant patients were asked to cover the distance at five different times on the same day chosen at random. Variability within one patient on test-retest was calculated by dividing the standard deviation of the five measures by the mean value of the measures. In spite of its great variability of 20 & p l u s m n ; 8%, gait speed timed over 10 m was as sensitive as the AI and measured changes in 16.5% of these ambulatory patients (AI in 15.6%). The RMI and gait speed timed over 10 m should be used more widely to assess disability and to measure the physiotherapeutic effort devoted to improving mobility of MS patients.