Abstract
Multiple sclerosis (MS) is a chronic neurodegenerative disease characterized by mobility impairments that limit physical activity and reduce quality of life. While traditional clinical measures and participant-reported outcomes provide valuable insights, they often fall short of fully capturing the complexities of real-world mobility. This study evaluates the predictive value of combining sensor-derived clinical measures and participant-reported outcomes to better forecast prospective physical activity levels in individuals with MS. Forty-six participants with MS completed surveys assessing fatigue, concern about falling, and perceived walking ability (MSWS-12), alongside sensor-based assessments of gait and balance. Over three months, participants wore Fitbit devices to monitor physical activity, including step counts and total activity levels. Forward stepwise regression revealed that a combined model of participant-reported outcomes and sensor-derived measures explained the most variance in future physical activity, with MSWS-12 and backward walking velocity emerging as key predictors. These findings highlight the importance of integrating subjective and objective measures to provide a more comprehensive understanding of physical activity patterns in MS. This approach supports the development of personalized interventions aimed at improving mobility, increasing physical activity, and enhancing overall quality of life for individuals with MS.