BMC Public Health · Nov 18, 2026
Sleep duration, sarcopenia, and incident cognitive impairment among middle-aged and older chinese adults: a prospective cohort study
By Gao L, Chen Y, Ma K et al.
Background Sarcopenia and sleep duration have both been associated with cognitive impairment (CI), but their joint associations and potential interactions remain unclear. We examined their mutually adjusted and joint associations with incident CI and evaluated multiplicative and additive interactions. Method This prospective cohort study included 3,248 CHARLS participants aged ≥ 50 years without CI at baseline. Cox proportional hazards models were used to assess the mutually adjusted and joint associations of sarcopenia and sleep duration with incident CI. Restricted cubic splines were used to examine dose–response associations, and multiplicative and additive interactions were evaluated. Subgroup and sensitivity analyses assessed heterogeneity, residual confounding, and reverse causation. Results In fully adjusted models, sarcopenia (HR = 1.26, 95% CI: 1.02–1.55) and short sleep duration (HR = 1.18, 95% CI: 1.04–1.34) were independently associated with incident CI, whereas possible sarcopenia and long sleep were not significantly associated with CI. Restricted cubic spline analysis indicated a significant overall association between sleep duration and CI (P for overall = 0.043), but no statistically significant evidence of nonlinearity (P for nonlinearity = 0.071). In joint analyses, short sleep was associated with incident CI among participants without sarcopenia (HR = 1.18, 95% CI: 1.03–1.35), whereas no significant multiplicative or additive interaction was observed. Neither multiplicative nor additive interaction was statistically significant. Sensitivity analyses generally yielded similar patterns, although some estimates were attenuated. Conclusion Among adults aged ≥ 50 years, sarcopenia and short sleep duration were independently associated with incident CI. No statistically significant interaction between sleep duration and sarcopenia was detected, and estimates for specific joint exposure categories remained uncertain.