Continence Impairment and Depressive Symptoms Jointly Identify Older Adults at Increased Risk of Mortality: Parallel Analyses From NHANES and CHARLS.

BACKGROUND: Continence impairment and depressive symptoms frequently coexist in older adults, but whether their co-occurrence identifies a subgroup at elevated mortality risk across populations remains uncertain. METHODS: We conducted parallel analyses among adults aged ≥ 60 years in the 2005-2010 National Health and Nutrition Examination Survey (NHANES) linked mortality files and the 2011-2018 China Health and Retirement Longitudinal Study (CHARLS). Continence impairment was defined as urinary or fecal incontinence in NHANES and by a combined functional item on difficulty controlling urination and defecation in CHARLS; these measures were harmonized at the conceptual rather than item level. Participants were classified as having neither exposure, depressive symptoms only, continence impairment only, or both. NHANES used survey-weighted Cox models, whereas CHARLS used ordinary unweighted Cox models in a unified fixed 7-year cohort. Missing covariates were addressed by multiple imputation with 20 datasets. Additive interaction measures were estimated from fixed-horizon risk ratios, not hazard ratios. RESULTS: The analyses included 4851 NHANES participants (1052 deaths) and 4690 CHARLS participants (1044 deaths). Compared with neither exposure, the co-occurring group had higher mortality in NHANES (HR 1.60, 95% CI 1.20-2.13; p = 0.001) and CHARLS (HR 1.50, 95% CI 1.19-1.87; p < 0.001). Fixed-horizon risk-ratio analyses provided no statistical evidence of additive interaction. The CHARLS midpoint death-time analysis produced a similar co-occurring-group estimate (HR 1.54, 95% CI 1.23-1.93). CONCLUSIONS: Co-occurring continence impairment and depressive symptoms identified a higher-risk subgroup relative to participants with neither exposure in both cohorts, highlighting the value of considering these conditions together in mortality risk assessment.