OBJECTIVES:Survivors of stroke often report poor physical function leading to significantly decreased quality of life. Poor sleep quality constitutes one of the most common difficulties faced by older adults. The objective of this secondary analysis was to investigate the effect of physical activity on sleep quality and several other sleep specific outcomes in older stroke survivors.

METHODS:A single blind three-arm randomized control trial investigating Tai Chi (TC), SilverSneakers (SS) and usual care (UC) groups. For this secondary analysis, we combined TC and SS as the physical activity intervention arm with UC serving as the control arm. The 19-item Pittsburgh Sleep Quality Index (PSQI) questionnaire was used to quantify overall sleep quality with sub-components of the PSQI to calculate the sleep specific outcomes in intervention and control arms. Our primary analysis was unadjusted linear mixed model for repeated measures. In addition, a sensitivity analysis was conducted to account for missing data.

RESULTS:Post-intervention at 12-weeks, the difference in mean PSQI scores between intervention and control arms was 0.4 points (95% CI: -1.0, 1.7). At the 24-week follow-up, the difference in mean PSQI scores between intervention and control arms was 0.4 points (95% CI: -1.1, 1.8). There were no statistically significant differences in the PSQI scores or sleep specific outcomes at any time point. The rate of change of PSQI from baseline to any time point was statistically significant in primary analysis, however, in the sensitivity analysis this was non-significant. [(0.33,2.68) ,(0.36,2.71) vs (-3.03,0.83),(-1.89,1.91)]

CONCLUSIONS:We found no evidence that physical activity improved sleep outcomes in older community-dwelling stroke survivors. Our results suggest that missing data impacts the interpretation and significance seen in primary analysis could be a type one error. Future studies where sleep quality is measured by more accurate objective measures are recommended to examine this relationship.