Health risks and genetic architecture of objectively measured multidimensional sleep health.

Abstract

A multidimensional sleep health framework improves screening and treatment efficacy by simultaneously addressing multiple sleep domains. However, limited studies have used objective measures to evaluate the co-occurrence of diverse unhealthy sleep characteristics and their pleiotropic health effects. To represent real-world sleep patterns, we introduce the Unfavorable Sleep Profile (USP), an integrated multidimensional sleep health metric developed using accelerometer data in the UK Biobank (N = 85,233; aged 43-79 years). USP captures five domains: sleep timing, efficiency, duration, rhythmicity, and regularity. Phenome-wide association study found that USP was significantly associated with 76 out of 526 incident health outcomes over 7.9 years of follow-up. We identified several upstream environmental risk factors associated with USP, including low socioeconomic status. Whole-genome sequence analyses identified common variants in MEIS1 and rare coding variants in TTC1 associated with USP. We validated the USP framework in an independent cohort, the Multi-Ethnic Study of Atherosclerosis. Our findings underscore the importance of multidimensional sleep health assessment in predicting and potentially mitigating a wide array of health disorders and advance genetic insights into sleep health.