Links of Genetic Risk for Short Sleep Duration with Cognitive, Functional, and Biological Aging Outcomes in Older Adults

Embargo until
Journal Title
Journal ISSN
Volume Title
Johns Hopkins University
Background: Older adults are disproportionately susceptible to sleep problems. Nearly one-half of older adults live with insomnia and almost 7% experience excessive daytime sleepiness on a daily or almost daily basis. Sleep has been implicated in a number of health conditions (e.g., obesity, dementia) and is associated with several social and economic outcomes including workplace injuries and motor vehicle accidents. The U.S. population of older adults is expected to double from around 50 million to nearly 100 million individuals in the next 40 years, suggesting an increased prevalence of sleep disturbance and aging-related outcomes such as chronic disease, disability, and Alzheimer’s disease. Understanding the mechanisms and consequences of sleep disturbance can inform prevention and treatment strategies for sleep disturbances in older adults. One way to uncover mechanisms is using genetic markers of risk. However, little work has examined genetic risk for short sleep duration with measures of cognitive and physical performance or biological markers of aging. Method: Data for these analyses came from the Baltimore Longitudinal Study of Aging (BLSA), an ongoing, prospective cohort study of over 3,200 participants. Polygenic risk scores were calculated using genetic data ascertained between 1986 and 2017 on a subset of 848 participants. Cognitive and physical performance outcomes were gathered using data from each BLSA study visit, whereas biological markers of aging were calculated using the Horvath DNA methylation age calculator from epigenetic data, collected on 820 participants between 2008 and 2012. We investigated prospective links of polygenic risk for short sleep duration with cognitive performance (n=1,242) and physical performance (n=1,373-1,458) using linear mixed effects models, and cross-sectional associations with biological aging markers (n=467) using linear regression models. We also investigated associations of self-reported sleep duration with the biological aging markers (n=615). Across models, participants were predominantly white and male. All models were adjusted for age, sex, years of education, and medical comorbidities. Results: Our findings show no significant associations between polygenic risk score for short sleep duration and cognitive domain scores, but significant associations between polygenic risk score for short sleep duration and decline in SPPB score and some biological measures of aging (i.e., estimated granulocyte count and plasminogen activator inhibitor-1), though these results are attenuated in fully adjusted models. In models investigating phenotypic sleep duration as the primary predictor and biological measures of aging as the outcome, relative to individuals sleeping ≤6 hours, those sleeping >7 hours, showed faster Hannum age-acceleration, and greater PhenoAge, GrimAge, estimated granulocyte count, and PAI-1. We also found moderation of associations of self-reported sleep duration with some biological measures by age and sex. Implications: Higher genetic risk for short sleep duration is associated, to varying degrees, with faster decline in physical performance and worse levels of biological markers of accelerated aging among older adults. Findings suggest that shorter sleep duration may contribute to poor physical performance and accelerated aging. Prospective studies in larger samples are needed to examine whether these effects are robust to the inclusion of confounders, and whether effect estimates change across demographic subgroups.
Sleep: Aging