Sleep Duration and Multimorbidity as Joint Predictors of All-Cause Mortality in US Adults

Background Both short and long sleep durations have been linked to elevated mortality risk, but how these relationships differ across levels of multimorbidity remain unclear. This study examined the joint association of sleep duration and multimorbidity with all-cause mortality in a nationally representative sample of U.S. adults. Methods We analyzed data from 28,598 adults aged ≥18 years who participated in the 2004 National Health Interview Survey (NHIS) and were prospectively linked to mortality data through 2019 via the National Death Index. Self-reported sleep duration was categorized as 1–4, 5–6, 7–8 (reference), 9, or ≥10 hours. Multimorbidity burden was quantified using the Charlson Comorbidity Index (CCI) and grouped into tertiles. We also examined specific multimorbidity patterns (cardiovascular, cancer, respiratory, musculoskeletal, endocrine/metabolic). Cox proportional hazards models estimated hazard ratios (HRs) for all-cause mortality across sleep duration categories, stratified by multimorbidity burden and pattern, and adjusted for sociodemographic and behavioral covariates. Results A J-shaped association between sleep duration and mortality was observed. In unadjusted analyses, both short (1–4 hours) and long (≥10 hours) sleepers showed elevated mortality risk, particularly among participants with higher multimorbidity. After full adjustment, long sleep (≥9 hours) remained significantly associated with increased mortality across all multimorbidity tertiles and disease clusters. Hazard ratios increased progressively across comorbidity levels and sleep durations, suggesting a dose–response relationship. The sleep–multimorbidity interaction was statistically significant (p < 0.0001), with the strongest associations seen among individuals with respiratory, endocrine/metabolic, or musculoskeletal conditions. Conclusions Both short and long sleep durations were linked to increased mortality, with risk amplifying across higher levels of multimorbidity. Long sleep remained a strong independent predictor even after adjustment, underscoring the value of assessing sleep duration alongside chronic disease burden and the need for studies exploring mechanisms behind prolonged sleep and survival.