Accelerometry is associated with walking mobility, not physical activity, in persons with multiple sclerosis

Accelerometers are seemingly a criterion standard of real-life walking mobility and this is supported by assumptions and empirical data. This application would be strengthened by including objective measures of walking mobility along with a matched control sample for verifying specificity versus generality in accelerometer output. We compared associations among accelerometer output, walking mobility, and physical activity between persons with multiple sclerosis (MS) and controls without a neurological disorder. Sixty-six persons (33 MS, 33 matched controls) completed a battery of questionnaires, performed the six-minute walk (6MW) and timed-up-and-go (TUG), and wore an accelerometer for a 7-day period. After this period, participants completed the Godin Leisure-Time Exercise Questionnaire (GLTEQ) and International Physical Activity Questionnaire (IPAQ). Accelerometer output was significantly correlated with only mobility measures (6MW, ρ=78; TUG, ρ=-68) in MS, whereas it correlated with both mobility (6MW, ρ=58; TUG, ρ=-49) and physical activity (GLTEQ, ρ=56; IPAQ, ρ=53) measures in controls. Regression analysis indicated that only 6MW explained variance in accelerometer output in MS (β=65, R²=43). These findings support the possibility that accelerometers primarily and specifically measure real-life walking mobility, not physical activity, in persons with MS.