Long Sleep Tied to 51% Higher Mortality Risk in Adults With Heart Disease
Nine or more hours of sleep carried a 51% higher risk of death over 15 years among US adults with heart disease, when researchers expected the usual seven-to-eight-hour range to hold steady as the safest.
The finding comes from a cohort study published in PLOS One, reported in the journal and covered by News-Medical.
Key Clinical Takeaways:
- Adults with self-reported heart disease who slept nine or more hours per day had roughly a 50% higher risk of death from all causes, heart-related causes, and non-heart-related causes than those sleeping seven to eight hours.
- The association persisted after ten sensitivity analyses, including excluding deaths in the first two years of follow-up.
- Short sleep of six hours or less showed no significant link to mortality after full adjustment.
- Researchers frame long sleep as a possible marker of hidden vulnerability — not a proven cause of death.
Why sleep duration drew researchers to heart patients
People living with heart disease continue to die at higher rates despite preventive efforts aimed at conventional risk factors: blood pressure, diabetes, cholesterol, and smoking. That gap has pushed investigators toward modifiable lifestyle factors, and sleep stands out among them because duration can be assessed easily, quickly, and at almost no cost, News-Medical reported.
Sleep regulates metabolism, inflammation, autonomic function, and cardiovascular recovery. Studies in the general American population have already shown a nonlinear relationship between sleep duration and death from all causes, heart-related causes, and non-heart-related causes. The question the PLOS One authors posed was whether the same pattern holds in a population that already carries a diagnosis of coronary heart disease, another heart condition, or a heart attack.
How the 15-year cohort was built
The study linked the 2004 National Health Interview Survey to the National Death Index through December 31, 2019. Of 94,597 survey participants, 3,287 adults had heart disease and provided complete data. The researchers stratified reported sleep per 24 hours into three bands: six hours or less, seven to eight hours (the reference group), and nine hours or more.
Cox proportional hazards models were adjusted in sequence for demographics, then health and lifestyle behaviors — body mass index, smoking, drinking, and physical activity — then comorbidities including hypertension, diabetes, stroke, and emphysema. Variables that violated the proportional hazards assumption were handled by stratification, and restricted cubic splines tested for nonlinear relationships.
To test the robustness of their results, the authors conducted ten sensitivity analyses, which included dropping deaths occurring in the initial two years, limiting the sample to particular cardiac conditions, using competing-risk models, removing participants with a cancer history, and incorporating survey weights.
The mortality numbers, side by side
Over roughly 15 years of follow-up, 49.4% of participants died. Of the full study population, 16.7% died from heart-related causes and 32.6% from non-heart-related causes. In the fully adjusted model, long sleep was associated with hazard ratios of 1.51 for all-cause mortality, 1.53 for heart-related mortality, and 1.50 for non-heart-related mortality compared with seven to eight hours of sleep.
| Cause of death | Hazard ratio, 9+ hours vs 7–8 hours | Relative increase |
|---|---|---|
| All causes | 1.51 | 51% |
| Heart-related | 1.53 | 53% |
| Non-heart-related | 1.50 | 50% |
Short sleep told a different story. Sleeping six hours or less showed no significant association with mortality after full adjustment. It was linked to non-heart-related death only in the least-adjusted model, and that association weakened as further adjustments were applied. The authors caution that this does not negate the known adverse cardiovascular effects of insufficient sleep.
Lowest mortality risk near 6.5 hours
Spline analysis showed a significant U-shaped pattern for all-cause and non-heart-related mortality. The lowest predicted risk for all-cause mortality fell at about 6.5 hours, and at 6 hours and 6.8 hours for heart-related and non-heart-related mortality respectively. For heart-related mortality, the nonlinearity itself was not significant.
The authors are explicit about what those numbers do not mean: the estimates do not establish an optimal sleep duration, and they do not indicate that reducing sleep would lower mortality risk.
The results remained broadly consistent when participants were stratified by age — under 65 versus 65 and over — and by sex, with interaction tests that were not significant. One subgroup exception: the heart-related association with long sleep was not significant among those under 65, possibly because fewer heart-related deaths occurred in that group. The authors treat these subgroup differences as hypothesis-generating.
Long sleep link persists across sensitivity analyses
Across the sensitivity analyses, long sleep was persistently associated with higher mortality, with reported all-cause hazard ratios generally ranging from 1.42 to 1.77. One analysis softened the picture: in the survey-weighted analysis, the all-cause increase was smaller, at 26%. The association also persisted after excluding deaths during the first two years of follow-up, which reduced — though did not eliminate — the concern that underlying illness explained the findings.
The authors suggest the 51% all-cause increase is larger than the roughly 30% increase reported in a previous general-population meta-analysis, and they suggest this may reflect a stronger prognostic value for long sleep duration in a population already carrying heart disease.
Prolonged sleep signals health vulnerability
The authors’ interpretation is that prolonged sleep may signal underlying health vulnerability in this high-risk population rather than directly cause death. Possible contributors they cite include depression, sleep-disordered breathing, inflammation, frailty, multimorbidity, and medication burden.
On that reading, asking about sleep duration is a simple, inexpensive, and rapid way to identify adults with heart disease who may need further evaluation — a conversation a patient could raise with a clinician reviewing residual cardiovascular risk.
Several limitations complicate interpretation, and the authors state them plainly. Sleep was self-reported and measured only once, which may not capture long-term patterns. Residual confounding is possible from depression, sleep apnea severity, medications, and heart disease severity.
Heart disease itself was defined by self-report, the study population mixed different heart conditions, and the findings apply only to the non-institutionalized population. Because the study is observational, causality cannot be inferred — long sleep may result from poor health rather than contribute to it.
What remains unresolved
For clinicians caring for adults who already have heart disease, the PLOS One authors point to a practical implication: regularly asking patients how long they sleep could serve as an added way to flag those at heightened risk who might otherwise be missed. Patients or clinicians weighing these findings against an individual’s sleep patterns could seek evaluation, particularly where excessive daytime sleepiness or suspected sleep-disordered breathing is present.
In the authors’ words, upcoming research should rely on objective and repeated assessments of sleep to clarify whether sleeping long in this group is a sign of underlying health vulnerability, a manifestation of sleep disorders already present, or a marker of worsening heart disease. That question — which of the three explanations the 51% figure actually represents — is the one the current data cannot yet answer.
Disclaimer: The information provided in this article is for educational and scientific communication purposes only and does not constitute medical advice. Always consult with a qualified healthcare provider regarding any medical condition, diagnosis, or treatment plan.