Predicting Real-World Physical Activity in Multiple Sclerosis: An Integrated Approach Using Clinical, Sensor-Based, and Self-Reported Measures

Figures

• 

  Figure 1. Organizational outline summarizing results from three regression models predicting 3-month total step counts, evaluated using the Akaike Information Criterion (AIC) for goodness of fit. Model 1 included participant-reported outcomes alone, model 2 used sensor-derived clinical measures, and model 3 combined participant-reported outcomes with sensor-derived clinical measures. The best-fitting model (model 3) explained 43% of the variance in 3-month step counts by integrating both participant-reported and clinical measures. Abbreviations: MSWS-12, 12-Item Multiple Sclerosis Walking Scale; BW, backward walking; FW, forward walking; FES-I, Fall Efficacy Scale-International; MFIS, Modified Fatigue Impact Scale; PAR, Push and Release Reactive Balance Assessment; ECFT, eyes closed feet together sway condition; and AIC, Akaike Information Criterion.

• 

  Figure 2. Organizational outline summarizing results from three regression models predicting 3-month total physical activity, evaluated using the Akaike Information Criterion (AIC) for goodness of fit. Model 1 included participant-reported outcomes alone, model 2 used sensor-derived clinical measures, and model 3 combined participant-reported outcomes with sensor-derived clinical measures. The best-fitting model (model 3) explained 43% of the variance in 3-month total physical activity by integrating both participant-reported and clinical measures. Abbreviations: MSWS-12, 12-Item Multiple Sclerosis Walking Scale; BW, backward walking; FW, forward walking; FES-I, Fall Efficacy Scale-International; MFIS, Modified Fatigue Impact Scale; PAR, Push and Release Reactive Balance Assessment; ECFT, eyes closed feet together sway condition; and AIC, Akaike Information Criterion.