Physical Activity and Cognitive Function in Postmenpausal Women: Depressive Symptoms and Sleep Disorder as Sequential Mediators

Abstract

Purpose This study aimed to examine the association between physical activity (PA) and cognitive function in postmenopausal women, and to investigate whether depressive symptoms and sleep disorder act as sequential mediators of this association. Methods We analyzed 5614 postmenopausal women in the China Health and Retirement Longitudinal Study (CHARLS, 2011–2018). PA was grouped into quartiles (Q1–Q4), with total cognitive score as the primary outcome. Mixed-effects models with restricted cubic splines characterized dose–response relationships. A multilevel serial mediation model with bias-corrected bootstrapping (1000 resamples) estimated indirect effects, and a cross-lagged panel model (CLPM) evaluated longitudinal directionality. Results PA showed an inverted U-shaped association with cognition and a U-shaped association with depressive symptoms. Relative to Q1, moderate PA (Q2) was associated with higher cognitive scores (β = 0.559; 95% CI, 0.232–0.887), whereas the highest stratum (Q4, reflecting heavy physical labor) was associated with lower scores (β = −0.354; 95% CI, −0.697 to −0.010). In Q2, significant indirect effects were observed through depressive symptoms (a1×b1 = 0.053), abnormal sleep duration (a2×b2= 0.078), and the serial depression-to-sleep pathway (a1×d×b2 = 0.006), totaling 0.137 (95% CI, 0.083–0.201). In the CLPM, moderate PA predicted fewer subsequent depressive symptoms (β = −0.056) and better cognition (β = 0.039), consistent with the mediation results. Conclusion In postmenopausal women, moderate PA is associated with better cognitive function, partly through fewer depressive symptoms and improved sleep, whereas high-intensity physical labor is associated with poorer cognition. Depressive symptoms and sleep may therefore be relevant considerations when planning strategies to preserve cognitive health in this population.