Abstract TP347: Sleep Regularity Index in Post-Stroke Recovery

Introduction: Sleep–wake disruption is common after stroke and may impede neurological recovery, functional gains, and quality of life. Beyond sleep duration or efficiency, the day-to-day regularity of the rest–activity cycle is a salient feature of recovery. The Sleep Regularity Index (SRI) quantifies the probability that a person is in the same sleep or wake state at the same clock time on consecutive days. This study evaluates SRI during post-stroke recovery over time, estimating population-averaged associations with informative covariates, and characterizing individual transitions between predefined SRI states over time. Methods: We analyzed 76 subjects that were part of a larger on-going study exploring the prevalence and impact of non-OSA sleep disorders post stroke. Demographic and clinical variables were collected at 10, 60, and 90 days post stroke. Participants wore an activity monitor (AM) for 7-days at each time point. SRI was calculated from the raw epochs files based on matched daily sleep patterns. Covariates included presence of a non-OSA sleep disorder, time, stroke severity (NIHSS), body mass index (BMI), and selected functional (BBS, SIS) and mood measures (PHQ-9). Population-level associations between SRI and covariates were estimated using generalized estimating equations (GEE) clustering on participant to account for temporally correlated data structures. We evaluated pooled transitions across SRI states (low <65, medium 65-80, high >80) over time by calculating the proportion of individuals who transitioned between categories across time. Results: Population-level associations between SRI and covariates showed significant associations with PHQ9 and SIS total score. NIHSS was marginally associated. No other associations reached significance (Figure 1). Individual transitions between SRI states showed among those starting in the low SRI, about half remained low, while nearly half improved, split roughly evenly to mid or to high by the next follow-up period (Figure 2). Conclusion: Greater depression, lower self-reported quality of life, and greater initial stroke severity were associated with worse SRI. Transition analyses indicated about half of participants remained in the same SRI state across time, with greater persistence among those better sleep regularity. These findings support SRI as a feasible longitudinal metric for monitoring sleep-wake patterns after stroke and suggest links between functional recovery and sleep regularity.