Individual and joint associations of physical activity and sleep duration with cognitive function in a Chinese nationwide cohort

Le résumé fourni par la source

Despite growing evidence linking sleep duration and physical activity (PA) to cognitive performance, their interactive effects remain understudied. This study aimed to investigate the individual and joint associations of sleep duration and PA with cognitive function in a Chinese nationwide cohort of mid-to-older-aged population. We performed a repeated-measurement analysis of adults aged 45 + years from 125 cities across 28 provinces, using four waves (2011 − 2018) of China Health and Retirement Longitudinal Study (CHARLS). Sleep duration was acquired by questionnaires and was categorized into three categories (short < 6 h, optimal 6–8 h, long > 8 h). PA was measured as MET-hours/week from self-reported intensity and frequency, and categorized into low (< 2160 MET-h/week), moderate (2160 − 7038 MET-h/week), and high (≥ 7038 MET-h/week) tertiles. Cognitive function was evaluated across domains by trained investigators in face-to-face computer assisted interviews. We used a multivariable-adjusted linear mixed-effects model to evaluate the change of cognitive score (β) and its 95% confidence interval (CI) associated with non-optimal sleep duration and PA intensity, and estimated the combined associations with optimal sleep and moderate PA as reference. Subgroup analyses were used to testify the robustness of the primary findings in the overall population. A total of 34,714 measurements from 11,828 subjects(age 59.8 ± 8.8 years, 54.8% male) were included. Non-optimal sleep duration showed significantly lower cognitive scores in global cognition compared to optimal sleep (β short = -0.170, 95% CI: -0.282, -0.057; β long = -0.760, 95% CI: -0.948, -0.573). Low and high PA levels were associated with poorer cognitive performance relative to moderate PA (β low = -0.200, 95% CI: -0.330, -0.070; β high = -0.338, 95% CI: -0.470, -0.207). The most pronounced adverse effect was observed for the combination of long sleep duration and low PA (β = -1.026, 95% CI: -1.353, -0.698, P < 0.001), while moderate PA tended to offset the cognitive impairment effect of short sleep (β = -0.128, 95% CI: -0.311, 0.054, P = 0.169). Subgroup and domain-specific analyses broadly aligned with overall population findings. This national study highlights independent and joint associations of sleep duration and PA with cognitive function, emphasizing collaborative lifestyle interventions for healthy aging.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.