Sleep duration showed positive associations with all-cause and cause-specific mortality in both Chinese and UK cohorts, with 17 metabolic markers—including lipoprotein subclasses, cholesterol, and ketone bodies—potentially partially explaining the effect of sleep duration on mortality.
Key Findings
Results
Sleep duration was positively associated with all-cause and cause-specific mortality in both the China Kadoorie Biobank and UK Biobank cohorts after adjustment.
Data were drawn from two large population-based cohorts: the China Kadoorie Biobank (CKB) and the UK Biobank (UKB).
Sleep duration was self-reported by participants in both cohorts.
Cox proportional hazards models were used to assess the association between sleep duration and mortality outcomes.
Associations were observed for both all-cause and cause-specific mortality endpoints.
Results
A total of 17 metabolic markers were significantly associated with both sleep duration and mortality in the study cohorts.
The 17 markers encompassed lipoprotein subclasses in high-density lipoprotein cholesterol (HDL), relative lipoprotein lipid concentrations in very-low-density lipoproteins (VLDL), cholesterol, and ketone bodies.
Metabolic markers were quantified using nuclear magnetic resonance (NMR) spectroscopy.
Linear regression was used to assess the association between sleep duration and metabolic markers.
Cox models were used to further compare these associations with all-cause and cause-specific mortality.
Results
Lipoprotein subclasses and relative lipoprotein lipid concentrations showed similar directional patterns in their associations with both sleep duration and mortality across cohorts, while cholesterol and ketone bodies showed contrary patterns in the CKB.
In general, associations of sleep duration with lipoprotein subclasses and relative lipoprotein lipid concentrations, and of these markers with mortality, showed 'similar patterns.'
'Contrary patterns were observed for cholesterol and ketone bodies in the CKB.'
The UKB cohort showed 'metabolic markers presenting generally consistent association patterns with sleep duration and mortality.'
This cross-cohort difference suggests potential population-specific metabolic pathways linking sleep to mortality.
Results
Gender-specific analyses revealed sex differences in the associations between sleep duration, metabolic markers, and mortality.
In the CKB, significant associations were found only in females.
In the UKB, significant associations were observed in both genders.
The sex-specific difference was particularly pronounced in the Chinese cohort.
These findings suggest that the metabolic pathways linking sleep duration to mortality may differ by sex and by population.
Results
Sensitivity analyses confirmed the stability of the associations between sleep duration, metabolic markers, and mortality.
Sensitivity analyses were conducted to test robustness of findings.
The paper states that 'sensitivity analyses confirmed the stability of these associations.'
No specific sensitivity analysis methods or numerical results are described in the abstract.
Conclusions
The study concludes that lipid metabolism may partially explain the effect of sleep duration on mortality, suggesting that targeted metabolic interventions could potentially mitigate sleep-related mortality risks.
Authors suggest 'the potential of targeted metabolic interventions to mitigate the risks associated with abnormal sleep patterns, especially when modifying sleep duration itself is challenging.'
The projected health burden associated with sleep disorders 'could be reduced through metabolic-focused strategies.'
The authors note that further research is needed to determine whether these associations are causal.
Lifestyle or dietary modifications are mentioned as potential intervention approaches.
What This Means
This research used data from two very large studies—one in China (the China Kadoorie Biobank) and one in the UK (the UK Biobank)—to examine how sleep duration relates to specific substances in the blood (metabolic markers) and whether those substances are in turn linked to risk of death. Sleep duration was reported by participants themselves, and blood metabolic markers were measured using a precise laboratory technique called nuclear magnetic resonance spectroscopy. The researchers found that sleeping for longer or shorter durations was associated with differences in 17 specific metabolic markers, including types of cholesterol-carrying particles (lipoproteins) and ketone bodies (byproducts of fat metabolism). These same markers were also associated with the risk of dying from all causes and from specific causes.
The patterns were not identical across the two countries. For lipoprotein particles, the associations between sleep, the marker, and mortality generally pointed in the same direction in both cohorts. However, for cholesterol and ketone bodies, the direction of association with mortality was opposite in the Chinese cohort compared to the UK cohort. Additionally, among Chinese participants, the sleep-metabolism-mortality link was statistically significant only in women, while in the UK it was significant in both men and women. These differences may reflect variations in genetics, diet, lifestyle, or other factors between the two populations.
This research suggests that changes in lipid (fat) metabolism may be one pathway through which abnormal sleep duration affects health and longevity. This is potentially meaningful because, while it can be difficult to change how long someone sleeps, it may be more feasible to address metabolic risk factors through diet, exercise, or other lifestyle changes. The authors propose that metabolic monitoring or interventions could help reduce the health risks tied to poor sleep, though they caution that more research is needed to confirm whether these relationships are truly cause-and-effect.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Hua G, Zhang L, Li C, Wang Y, He X, Cheng C, et al.. (2026). Association of Sleep Duration, Metabolic Markers, and Risk of All-Cause and Cause-Specific Mortality in China and the UK.. Global heart. https://doi.org/10.5334/gh.1582