Abstract
Background: While sleep duration's association with stroke is established, the combined influence of sleep onset time and duration on stroke subtypes remains inadequately explored. Since circadian biology links sleep onset timing to vascular risk via mechanisms operating independently of sleep duration, we quantified their joint contributions to the risk of stroke. Methods: In this population-based cross-sectional study, from 31 December 2021 to 31 March 2022, we recruited 8168 ischemic stroke cases, 3172 intracerebral hemorrhage cases, and 13,458 control participants across 152 survey centers in 28 counties in Hunan Province, China. Standardized computer-assisted interviews collected sleep parameters. Conjoint analysis identified protective sleep profiles. Results: Short sleep duration (<6 h) was consistently associated with elevated ischemic risk across all sleep onset times (p < 0.05 in all cases, i.e., sleep before 10 p.m. [odds ratio (95%CI): 1.886(1.606, 2.214)], 10-11 p.m. [1.740(1.336, 2.265)], 11 p.m.-12 a.m. [2.335(1.190, 4.581)], and after 12 a.m. [2.834(1.193, 6.728)]). A sleep duration of 6-8 h with a sleep onset time between 10 p.m. and 12 a.m. was associated with the lowest ischemic risk (p < 0.001 in all cases). Conversely, prolonged sleep (>8 h) with an early sleep onset time (<10 p.m.) increased ischemic risk (OR 1.194, 95% CI 1.090-1.308, p < 0.001), whereas a late sleep onset time (11 p.m.-12 a.m.) in long sleepers was protective (OR 0.580, 95% CI 0.352-0.956, p < 0.001). Similar trends were observed for ICH, though the effect sizes were attenuated. Conclusion: Sleep duration and onset time interact to influence stroke risk. Optimal cerebrovascular protection requires ≥6 h of sleep, ideally initiated between 10 p.m. and 11 p.m. These findings highlight sleep optimization as a potential modifiable target for high-risk populations.