Associations of Accelerometer-measured Sleep Duration with Incident Cardiovascular Disease and Cardiovascular Mortality

Abstract Study Objectives This study aimed to determine the associations between accelerometer-measured sleep durations and the risks of incident cardiovascular disease (CVD) and CVD-related mortality. Methods A total of 92,261 participants (mean age: 62.4±7.8 years, 56.4% female) were included in UK Biobank between 2013 and 2015. Average daily sleep durations were measured using wrist-worn accelerometers over a seven-day period. Sleep durations were categorized as <7 hours/day, 7-9 hours/day (reference), and >9 hours/day. The incidence of CVD and CVD-related mortality were ascertained by hospital records and death registries. Results During a median follow-up period of 7.0 years, a total of 13,167 participants developed CVD, and 1,079 participants died of CVD. Compared with a sleep duration 7-9 hours/day, an accelerometer-measured sleep duration <7 hours/day but not >9 hours/day was associated with higher risks of incident CVD (HR 1.06, 95% CI: 1.02–1.10), CVD-related mortality (HR 1.29, 95% CI: 1.14–1.47), coronary heart disease (HR 1.11, 95% CI: 1.03–1.19), myocardial infarction (HR 1.14, 95% CI: 1.03–1.27), heart failure (HR 1.20, 95% CI: 1.08–1.34), and atrial fibrillation (HR 1.15, 95% CI: 1.07–1.24). A curvilinear dose‒response pattern was observed between accelerometer-measured sleep durations and incident CVD (Poverall<0.001), with L-shaped associations found for incident CVD and CVD-related mortality. Conclusions An accelerometer-measured sleep duration <7 hours/day but not >9 hours/day was associated with elevated risks of incident CVD and CVD-related mortality. Maintaining adequate sleep may help promote cardiovascular health.