Long Sleep Duration Tracks With Higher Mortality in Adults Living With Heart Disease
Among US adults with heart disease, sleep duration carried a clear prognostic signal. In a nationally representative cohort of 3,287 participants from the 2004 National Health Interview Survey linked to the National Death Index through December 31, 2019, long sleep duration, defined as 9 hours or more, was associated with higher risks of death from every outcome measured when compared with the reference group sleeping 7 to 8 hours.
The signal was not subtle. Over 15 years of follow-up, 49.4% of participants died, 16.7% from heart-related causes and 32.6% from non-heart-related causes. Long sleep duration was associated with a 51% higher risk of all-cause mortality (HR 1.51, 95% CI 1.32–1.73), a 53% higher risk of heart-related mortality (HR 1.53, 95% CI 1.22–1.92), and a 50% higher risk of non-heart-related mortality (HR 1.50, 95% CI 1.27–1.77).
The lowest predicted mortality risk appeared at 6 to 7 hours of sleep. In a cohort already burdened by established cardiovascular disease, that nonlinearity matters because it argues against a simple “more sleep is better” assumption and instead points toward a narrower physiologic range in which sleep may be most compatible with long-term survival.
A Nonlinear Curve Suggests Sleep May Mark Residual Risk Rather Than Act Alone
The analysis used Cox proportional hazards models stratified by variables that violated the proportional hazards assumption, and restricted cubic spline modeling to examine dose-response patterns. Those methods are important because they allow the data to reveal whether risk rises in a straight line or bends at the extremes. Here, the curve was nonlinear, with the nadir around 6 to 7 hours rather than at the longest sleep durations.
Long sleep in patients with heart disease is rarely just a habit. In clinical practice, extended sleep time can reflect poor baseline functional status, occult inflammation, depression, sleep-disordered breathing, medication burden, frailty, or recovery from chronic illness. Any of those factors could help explain why longer sleep duration may behave as a marker of residual risk rather than a benign preference. The biology is plausible: disturbed sleep can influence autonomic tone, metabolic regulation, inflammatory signaling, and cardiovascular recovery, and prolonged sleep may signal a system already under physiologic strain.
The cohort findings remained robust across multiple sensitivity analyses and within age and sex subgroups. That consistency makes the association harder to dismiss as a chance finding or a narrow artifact of one demographic stratum. It also suggests that sleep duration may carry clinically relevant information across a broad spectrum of adults with heart disease, even though the study design cannot prove causality.
One finding stands out sharply: long sleep duration was tied not only to cardiovascular death but also to non-heart-related mortality. That broader association points to sleep as a general health marker, not simply a cardiovascular one.
What the Mortality Pattern Means for Cardiovascular Care
Heart disease remains a leading cause of death in the United States, and patients who survive an initial cardiac event or live with chronic heart disease often enter a prolonged period of elevated vulnerability. Standard secondary prevention focuses on blood pressure, lipids, glucose control, smoking, and medications, but those measures do not fully capture the heterogeneity of risk that persists in day-to-day life. Sleep duration is attractive in this setting because it is easy to ask about, inexpensive to track, and potentially modifiable.
A brief sleep history may therefore do more than document a symptom. It may identify patients whose elevated risk is not obvious from traditional cardiovascular markers alone. In a busy clinic, a patient consistently sleeping 9 hours or more may warrant a broader review of fatigue, heart failure symptoms, depression, frailty, medication effects, and possible sleep disorders. The study does not show that changing sleep duration will lower mortality. It does show that prolonged sleep is associated with worse outcomes in a population where every added risk marker has practical value.
The use of nationally representative survey data linked to mortality records gives the findings population relevance, but the design also leaves important gaps. Sleep duration was categorized into three groups, which can miss day-to-day variability and cannot distinguish voluntary long sleep from illness-related hypersomnia. Sleep quality, sleep apnea, daytime sleepiness, and changes in sleep over time were not measured in the abstracted data. Those omissions matter because duration alone is only one part of sleep health, and the clinical meaning of long sleep can vary widely across patients.
Still, the direction of association was consistent across all three mortality endpoints, and the pattern aligned with prior concern that both short and long sleep may signal physiologic stress. For adults with established heart disease, the practical message is not to prescribe a single universal sleep target without context. It is to treat unusually long sleep as a clinically meaningful clue, especially when it appears alongside reduced activity, worsening dyspnea, depression, or other signs of declining health.
This cohort study leaves one central issue unresolved: whether long sleep duration itself contributes to mortality risk or whether it is mostly a marker of underlying disease severity. The data support risk stratification, not intervention. The next step is not a broader sleep recommendation, but better phenotyping of why some patients with heart disease consistently sleep longer and whether that pattern can be separated from the illnesses that drive it.