The relationship between sleep and physical activity in an in-patient rehabilitation stroke setting: a cross-sectional study.

BACKGROUND Appropriate sleep and physical activity are known to be important for positive neuroplastic changes in the brain and therefore may affect stroke recovery. OBJECTIVE To investigate the relationship between sleep and participation in different intensity levels of physical activity; to investigate the convergent validity of a commercially available device (Fitbit Flex) in measuring sleep and physical activity in people with stroke. METHODS A cross-sectional observational study in people with stroke undergoing rehabilitation. Participants wore two accelerometers on their unaffected wrist for seven consecutive days and recorded a sleep log, the Pittsburgh Sleep Quality Index and Fatigue Assessment Scale. Any relationships between sleep and activity were assessed with linear regression. Pearson and intra-class correlation coefficients were used to assess the validity of the two accelerometers (Fitbit Flex against the validated GENEActiv). RESULTS Twenty-three patients with stroke were recruited. Twenty complete data sets were analyzed. Participants had approximately 9 hours of sleep and 13 hours of sedentary behavior per day, with 99 minutes spent on physical activity (16 min spent on moderate to vigorous activity). Time spent on sleep was significantly related to sedentary, light and moderate physical activity time (r = -.67, .22, .20). The ICC of the Fitbit Flex in measuring light physical activity was .884 p < .001 but was not correlated for other measures. CONCLUSIONS More sleep was related to less sedentary behavior and more time spent on physical activity. Fitbit Flex was only valid for measuring light physical activity.