Excessive total daily sleep duration (≥9 h) or insufficient sleep duration (<4 h) was significantly associated with increased risks of all-cause mortality and premature death among middle-aged and older adults, with a U-shaped association observed for both total and nighttime sleep duration.
Key Findings
Results
Both total daily sleep duration and nighttime sleep duration showed significant U-shaped associations with risks of all-cause mortality and premature death.
Restricted cubic spline analysis revealed a significant U-shaped association between total sleep duration and the risks of all-cause mortality and premature death (P < 0.05).
A similar U-shaped association was found between nighttime sleep duration and these risks (P < 0.05).
Kaplan-Meier curves demonstrated that both total daily sleep duration and nighttime sleep duration were significantly associated with the risks of all-cause mortality and premature death (P < 0.05).
The reference category for comparisons was 7–8 hours of sleep.
Results
Total daily sleep duration of ≥9 hours or <4 hours was significantly associated with elevated risk of all-cause mortality compared to 7–8 hours.
Total daily sleep duration ≥9 h was associated with HR = 1.904 (95% CI: 1.181–3.070) for all-cause mortality.
Total daily sleep duration <4 h was associated with HR = 2.537 (95% CI: 1.495–4.308) for all-cause mortality.
Both associations were statistically significant (P < 0.05).
Results were derived from multivariable Cox proportional hazards regression models.
Results
Nighttime sleep duration of ≥9 hours or <4 hours was significantly associated with elevated risk of all-cause mortality.
Nighttime sleep duration ≥9 h was associated with HR = 1.824 (95% CI: 1.100–3.024) for all-cause mortality.
Nighttime sleep duration <4 h was associated with HR = 1.637 (95% CI: 1.007–2.662) for all-cause mortality.
Both associations were statistically significant (P < 0.05).
Analyses used Cox proportional hazards regression with multivariable adjustment.
Results
Extreme total daily sleep durations (≥9 h or <4 h) were significantly associated with increased risk of premature death.
Total sleep duration ≥9 h was associated with HR = 2.047 (95% CI: 1.175–3.566) for premature death.
Total sleep duration <4 h was associated with HR = 2.445 (95% CI: 1.260–4.743) for premature death.
Both associations were statistically significant (P < 0.05).
Results
Extreme nighttime sleep durations (≥9 h or <4 h) were significantly associated with increased risk of premature death.
Nighttime sleep duration ≥9 h was associated with HR = 2.320 (95% CI: 1.190–4.521) for premature death.
Nighttime sleep duration <4 h was associated with HR = 1.943 (95% CI: 1.003–3.762) for premature death.
Both associations were statistically significant (P < 0.05).
Results
Daytime napping and sleep disturbance were not significantly associated with mortality outcomes.
Neither daytime napping duration nor sleep disturbance showed a statistically significant association with all-cause mortality or premature death.
Sleep status variables examined included sleep disturbance, total daily sleep duration, nighttime sleep duration, and napping duration.
Sensitivity analyses yielded largely consistent results across all sleep variables.
Methods
The study included 16,375 middle-aged and older adults from a nationally representative Chinese cohort.
Data were drawn from the China Health and Retirement Longitudinal Study (CHARLS).
The dataset is described as nationally representative.
Statistical methods included Kaplan-Meier curves, restricted cubic spline analyses, Cox proportional hazards regression models, and sensitivity analyses.
All-cause mortality and premature death were the dependent variables.
What This Means
This research suggests that both sleeping too little and sleeping too much are associated with higher risks of death and premature death among middle-aged and older adults in China. Using data from over 16,000 participants in a nationally representative study, researchers found that people who slept fewer than 4 hours or 9 or more hours per day had significantly higher risks of dying compared to those who slept 7–8 hours. This pattern held for both total daily sleep (including naps) and nighttime sleep alone, forming a U-shaped curve where the lowest risk was in the middle range.
The increased risk associated with very short sleep (<4 hours total daily) was especially pronounced, with participants more than 2.5 times as likely to die from any cause compared to those sleeping 7–8 hours. Similarly, sleeping 9 or more hours was associated with roughly double the risk of premature death. Notably, daytime napping habits and self-reported sleep disturbances were not significantly linked to mortality risk, suggesting that the total amount of sleep — rather than when it occurs or whether it is disrupted — may be the more critical factor.
This research suggests that maintaining a moderate sleep duration of 7–8 hours may be important for longevity in older populations. The findings highlight that both ends of the sleep duration spectrum — not just short sleep — may carry health consequences. Because the study is observational and based on self-reported sleep data, it cannot prove that sleep duration directly causes changes in mortality risk, but the consistency of results across sensitivity analyses strengthens confidence in the association.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Xu F, Albadry M, Dirsch O, Dahmen U. (2026). The effect of sleep status on all-cause mortality and premature death among middle-aged and older adults: A national cohort study.. Archives of gerontology and geriatrics. https://doi.org/10.1016/j.archger.2026.106385