Multiple sclerosis (MS) is a demyelinating disease of the central nervous system that often affects the motor system. We tested the hypothesis that physical activity was lower in a group of 17 MS patients (mean +/- SD; age = 46 +/- 6 yr, 11 females, 6 males) compared with 15 healthy sedentary control subjects (age = 44 +/- 7 yr, 9 females, 6 males). Physical activity was measured with a three-dimensional accelerometer and with an activity questionnaire for 7 d. Vector magnitude values from the accelerometer for the MS and sedentary control subjects were 121,027 +/- 59,336 and 185,892 +/- 60,566 arbitrary units/day, respectively (P = 0.01). Estimated energy expenditure values derived from the questionnaire were 35.9 +/- 3.0 and 36.2 +/- 4.1 Kcal.kg-1.d-1 (NS), respectively. Thus, when measured directly with an accelerometer, activity was lower in MS compared with sedentary control subjects. The data also suggest that the accelerometer was more sensitive than the questionnaire for detecting differences in activity between two relatively sedentary populations, including one with neurologic disease.