Background Sleep duration plays a crucial role in cognitive health and is closely linked to cognitive decline. However, the relationship between sleep duration and cognitive function in the Chinese population remains poorly understood. Objective This study aims to evaluate the association between sleep duration and cognitive function among middle-aged and older adults in China. Methods Using data from 15,526 participants in the 2020 China Health and Retirement Longitudinal Study, we used 3 composite indicators, encompassing episodic memory, mental acuity, and overall cognitive function to evaluate cognitive performance. Sleep duration per night, self-reported through face-to-face interviews, was also obtained. Adjustments were made using multiple generalized linear regression models, accounting for demographic, lifestyle, and health-related covariates. Results Among the 15,526 respondents analyzed, 53.02% (8232/15,526) were female and 46.98% (7294/15,526) were male, with an average age of 61.5 (SD 9.27) years. Those reporting sleep durations of 4 hours or less (β=−1.85, 95% CI −2.07 to −1.62), 5 hours (β=−0.55, 95% CI −0.78 to −0.33; P<.001), 9 hours (β=−1.78, 95% CI −2.17 to −1.39), and 10 hours or more (β=−3.01, 95% CI −3.39 to −2.63) per night had a significant negative relationship with cognitive function. In the adjusted model, the negative impact of long sleep (≥10 hours) on overall cognitive function became more pronounced (β=−3.01, 95% CI −3.39 to −2.63; P<.001), followed closely by extremely short sleep (≤4 hours; β=−1.85, 95% CI −2.07 to −1.62; P<.001). Conclusions This study reveals an inverted U-shaped relationship between sleep duration and global cognitive decline, indicating that cognitive function should be closely monitored in individuals with both short and long sleep durations. Consequently, public health strategies should prioritize the promotion of moderate sleep to mitigate the cognitive risks associated with aging, particularly in culturally specific contexts.
Guo et al. (Mon,) studied this question.