Feasibility and Outcomes of Device and Self-Reported Physical Activity Measurement after Stroke in the ENAbLE Randomized Pilot Trial

Introduction: Physical activity is recommended for secondary stroke prevention. Remote telehealth delivery of complex stroke interventions (e.g. exercise) offers potential to meet the challenges of accessible stroke care. However, the feasibility of remotely evaluating recommended outcomes, such as device-measured physical activity via wearable technology, is unknown. Furthermore, the effectiveness of physical activity interventions aimed at improving long-term physical activity for people with stroke is unclear. We aimed to evaluate the feasibility of remote measurement of physical activity via a research grade wearable device in the ENAbLE Pilot trial and report the effect of the physical activity intervention on device- and self-report measure physical activity. 

Methods: Analyses of secondary outcomes from a randomized trial (ENAbLE Pilot ACTRN12620000189921) involving adults 3 months to 10 years post-stroke or TIA who were able to walk independently (with or without aid). Physical activity was measured using the International Physical Activity Questionnaire (IPAQ; self-report measure) and activPAL physical activity device. Feasibility outcomes included proportion of the IPAQ collected, and proportion of activPAL devices returned and days of valid data. To assess the effect of the intervention on physical activity outcomes, we used descriptive statistics and linear mixed models. 

Results: Nearly all self-report (99%) and over three quarters (80%) of device-based measurements were available for analyses. No statistically significant differences in device measured physical activity were identified between participants who received the physical activity intervention and those who did not at the 3- or 6-month time points. Participants who undertook the physical activity intervention were more active (as measured by the activPAL) at 12 months than non-physical activity intervention participants (time spent in moderate to vigorous physical activity: 0.31, 95% CI [0.07–0.55] hours/day; light physical activity [LPA]: 0.22, 95% CI [0.05–0.39] hours/day, and daily step count 2, 321, 95% CI [578–4, 064] steps). No statistically significant differences between groups were identified in the type of physical activity undertaken (IPAQ data), except at 12 months, when walking activity was greater in physical activity intervention participants. 

Conclusion: Remote measurement of physical activity using a wearable device after stroke and via self-report is feasible. The piloted physical activity intervention shows potential to improve physical activity.