U shaped association between sleep duration and long term cognitive decline trajectories in a national cohort

Abstract

The existing evidence concerning the association between sleep duration and long-term cognitive trajectories remains inconclusive. This cohort study utilized data from the China Health and Retirement Longitudinal Study (CHARLS) spanning from 2011 to 2020, with biennial surveys conducted. Group-based trajectory modeling identified cognitive trajectories over a 9-year period. Multivariable logistic regression, incorporating multiple imputation techniques to address missing data, assessed the association between sleep duration and long-term cognitive function. The sample comprised 8668 participants with a mean age of 56.3 years, of whom 45.7% were female. Among them, 34.24% exhibited cognitive decline, while 65.76% maintained stable cognition. The odds ratios (OR) for incident cognitive decline were 1.17 (95% CI: 1.04–1.31) for individuals with short sleep duration (< 6 h) and 1.57 (95% CI: 1.31–1.89) for those with long sleep duration (> 8 h). Subsequent analysis revealed a U-shaped relationship between sleep duration and incident cognitive decline (nonlinear, p < 0.001). Specifically, when sleep duration was below 7.23 h, an increase in sleep duration significantly reduced the risk of cognitive decline (OR = 0.90, 95% CI: 0.86–0.95). Conversely, when sleep duration equaled or exceeded 7.23 h, an increase in sleep duration significantly elevated the risk of cognitive decline (OR = 1.31, 95% CI: 1.17–1.46). These findings suggest that sleep duration is independently associated with long-term cognitive decline in a U-shaped manner, with a nadir around 7.23 h. Screening and brief counseling for short (< 6 h) and long (> 8 h) sleep, alongside promotion of sleep hygiene and assessment for treatable sleep disorders, may offer scalable strategies to support healthy cognitive aging.