Abstract
Although circadian rhythms are established as a fundamental mechanism in various biologic processes,
including metabolic and cardiovascular functioning, less is known regarding how disruption of circadian
rhythms may contribute to development of cardiometabolic disease in broader human populations. Prior
epidemiologic studies have predominantly focused on specific populations who experience extreme circadian
disruption, such as rotating night shift workers. In this application, we will consider irregular sleep schedules as
a ubiquitous marker of chronic circadian disruption and evaluate its role in cardiometabolic disease
development. First, we will leverage the wealth of data from the UK Biobank (UKB), which has measured
habitual sleep using accelerometer among 92,644 participants. We will characterize the dose-response
relationships of irregular sleep schedules with risk of hypertension, diabetes and cardiovascular disease and
identify potential threshold to define what level of sleep variability may be cardiometabolically unhealthy. We
will also evaluate whether the observed associations differed by sociodemographic factors (e.g., age, sex,
race/ethnicity) or other sleep traits (e.g., average sleep duration and insomnia symptoms). Further, given that
sleep regularity represents a highly modifiable risk factor, we will evaluate whether regular sleep schedules
may counteract genetic predisposition to cardiometabolic disease. Second, in the Nurses’ Health Study 3
(NHS3), we propose to measure habitual sleep using Fitbit and plasma metabolomics among 1,000 nurses,
encompassing a wide range of variations in sleep schedules (including a random subset with night shift work).
We hypothesize that irregular sleep schedules are associated with altered metabolites exhibiting circadian
rhythms, such as omega-3 fatty acids, linoleate, arginine and tyrosine. We further hypothesize that this
metabolic profile mediates the associations between irregular sleep and cardiometabolic traits including
obesity, blood pressure and heart rate variability. In addition, we will collect new data on several emerging risk
factors for irregular sleep that have not been examined in previous work, including mobile device use, late
eating, breakfast skipping, pet ownership and childbearing/rearing in women. To increase rigor, reproducibility
and generalizability, we will confirm our primary findings from UKB and NHS3 in the diverse Multi-Ethnic Study
of Atherosclerosis (n=2,156), which have existing data on objectively measured habitual sleep, genomics,
metabolomics and conventional epidemiologic risk factors. Overall, this project will provide larger-scale, more
diverse and more in-depth evidence for the cardiometabolic impact of irregular sleep schedules in US and
European populations, elucidate underlying biologic mechanisms, and ultimately foster development of
potential public health recommendations and interventions to reduce irregular sleep and improve
cardiometabolic health.
