Wife-Husband Differences in the Associations Between Sleep Durations and Global Cognition Among Middle-Aged and Older Married Couples in China: A Longitudinal Dyadic Analysis

BACKGROUND: Sleep duration has well-established U-shaped associations with cognitive health, but research has largely ignored spousal interdependence. This study examined actor and partner effects of sleep duration on global cognition in middle-aged and older Chinese couples, testing whether associations differed by sex. METHODS: Data were drawn from CHARLS waves 1-4 (2011-2018). The analytical sample comprised 4,260 couples (8,520 individuals) aged ≥45 years living a married, two-person, mixed-sex household. Global cognition was assessed with a composite battery adapted from the Telephone Interview for Cognitive Status and Chinese Mini-Mental State Examination; nocturnal sleep and napping durations were self-reported. Actor-Partner Interdependence Models with quadratic terms examined actor effects (own sleep predicting own cognition) and partner effects (own sleep predicting spouse's cognition), with false discovery rate (FDR)-adjusted q-values and sex-equality tests. RESULTS: Significant inverted U-shaped actor effects were found for wives (β=-0.010, 95% CI: -0.014, -0.005, p < 0.001) and husbands (β=-0.011, 95% CI: -0.016, -0.006, p < 0.001); both survived FDR correction (q = 0.002, q < 0.001). Husbands' sleep was nominally associated with wives' cognition similarly (β=-0.005, p = 0.025), while wives' sleep was not associated with husbands' cognition (p = 0.618). The husband-to-wife association did not survive FDR correction (q = 0.239), and sex-equality tests were non-significant (χ²(9)=9.35, p = 0.406). Napping showed limited, largely non-significant associations with cognition. CONCLUSIONS: Own nocturnal sleep duration shows a robust, symmetric inverted U-shaped association with cognitive function in both spouses. The possible husband-to-wife partner effect did not survive correction for multiple testing and warrants adequately powered investigation before informing couple-based clinical recommendations.