Insufficient sleep tracks more closely with shorter life expectancy across U.S. counties than diet quality, physical activity, or social isolation, according to a peer-reviewed analysis of CDC survey data spanning 2019 to 2025. Only smoking showed a stronger negative association. A separate UK Biobank cohort study of roughly 59,078 adults reached a similar finding: small, realistic improvements in sleep produced larger modeled reductions in all-cause mortality than equivalent gains in exercise or nutrition. Together, the two studies shift the weight of evidence in a direction that could reshape how public-health agencies prioritize their spending.
Sleep deficits now rival smoking as a longevity threat
Andrew McHill, the senior author of the U.S. county-level analysis at Oregon Health and Science University, put the finding bluntly: “Sleep stood out more than diet, more than exercise, more than loneliness, indeed, more than any other factor except smoking.” That conclusion rests on county-by-county life-expectancy figures matched to responses from the Behavioral Risk Factor Surveillance System, the largest ongoing telephone health survey in the country. The study, published in SLEEP Advances, covers the period from 2019 through 2025, a window that captures both pre-pandemic baselines and the steep mortality shifts that followed.
In practical terms, the analysis suggests that chronic sleep deprivation may be eroding life expectancy in a way that rivals long-established threats. Counties with the highest share of adults reporting fewer than seven hours of sleep per night tended to post markedly lower life expectancy, even after accounting for socioeconomic indicators and other health behaviors. While the study is observational and cannot prove causation on its own, the strength and consistency of the association elevate insufficient sleep from a quality-of-life issue to a central longevity concern.
The findings land in a policy environment where public-health departments have traditionally focused on smoking cessation, nutrition, and physical activity. McHill and colleagues argue that sleep should now be considered alongside those pillars, not as a secondary concern. If the relationship holds up under further scrutiny, failing to address widespread sleep deficits could mean leaving substantial gains in life expectancy on the table.
The practical question the data raises is whether sleep-focused interventions, such as later high-school start times, limits on mandatory overnight shift work, or subsidized screening for sleep apnea, could move life-expectancy curves faster than traditional programs built around produce prescriptions or gym-access subsidies. No randomized trial has tested that head-to-head at the county level. But the observational signal is strong enough that researchers are now calling for exactly that kind of experiment. Counties already investing in dietary and fitness programs have a natural comparison group if even a handful adopt sleep-centered policies over the next two years.
Two independent datasets point in the same direction
The U.S. county analysis and the UK Biobank study used different populations, different measurement tools, and different statistical approaches, yet both landed on sleep as the behavioral factor with the largest apparent effect on mortality risk. The UK Biobank paper, published in eClinicalMedicine, drew on a cohort of roughly 59,078 participants and modeled what would happen if individuals made modest, achievable changes to sleep, moderate-to-vigorous physical activity, and diet quality simultaneously. Within those simulations, improving sleep produced the largest estimated drop in all-cause mortality among the three behaviors.
Crucially, the modeled sleep changes were not extreme. The authors focused on shifts that most adults could plausibly make: going to bed a bit earlier, extending sleep by less than an hour, or reducing the frequency of very short nights. Even within that modest range, the projected mortality benefit outpaced what similar incremental gains in exercise or diet would deliver. That does not mean exercise and nutrition are unimportant; rather, it suggests that for many people, sleep may be the lowest-hanging fruit for extending healthy years of life.
A third line of evidence adds texture to the sleep finding. A prospective cohort study using UK Biobank accelerometry data introduced the Sleep Regularity Index and found that day-to-day consistency of sleep predicted mortality risk better than total hours of sleep alone. Participants whose sleep and wake times varied widely from one day to the next faced substantially higher risks of death over the follow-up period, even when their average nightly duration fell within the recommended range.
That distinction matters for policy design. Telling people to “sleep more” is a blunt message, and one that can be hard to act on for workers juggling multiple jobs or caregivers managing unpredictable schedules. By contrast, helping people stabilize their sleep-wake timing is a more specific, actionable target that workplaces and school districts can influence through scheduling decisions. Policies that reduce swing shifts, minimize very early school start times, or align public transit schedules with typical work hours could all improve sleep regularity without requiring individuals to carve out more total time in bed.
Taken together, the three studies form a chain. The U.S. county data establishes the population-level association between insufficient sleep and shorter life expectancy. The UK Biobank cohort quantifies the relative benefit of sleep gains versus diet and exercise gains at the individual level. And the accelerometry analysis identifies regularity, not just duration, as the sleep variable that carries the most predictive weight for mortality risk.
Gaps in the data and what to watch next
Several open questions limit how far these findings can be pushed. The SLEEP Advances paper by McHill and colleagues uses CDC survey data, but the exact county identifiers and sleep-question wording for each survey year have not been released alongside the publication. Without that detail, independent researchers cannot fully replicate the geographic analysis or check whether changes in survey methodology between 2019 and 2025 introduced measurement drift. Transparency on those technical points will be essential for confirming that the observed associations are robust rather than artifacts of shifting survey practice.
The UK Biobank cohort, while large and deeply characterized, is drawn from a predominantly white British population that volunteered for a research study, a group that tends to be healthier and wealthier than the general public. That “healthy volunteer” effect raises questions about how well the mortality estimates will generalize to more diverse or socioeconomically disadvantaged communities, including many U.S. counties with the lowest life expectancy. The accelerometry sub-cohort also lacks linked county-level life-expectancy records, so its findings on sleep regularity cannot be mapped directly onto the U.S. geographic framework used in the McHill analysis.
Methodologically, there is another limitation. No published table yet shows simultaneous regression coefficients for sleep, physical activity, and diet quality within the same U.S. county model. As a result, the claim that sleep “stood out” relative to those factors rests partly on separate analyses rather than a single head-to-head comparison with all variables entered together. Future work that places all major health behaviors into the same statistical model, while carefully addressing collinearity and confounding, will be needed to confirm whether sleep truly exerts the largest independent influence on life expectancy at the county level.
Despite those gaps, the convergence of evidence is already shaping how clinicians and public-health experts talk about sleep. For anyone weighing personal health choices, the practical takeaway is straightforward: stabilizing a consistent sleep schedule and avoiding chronically short nights may deliver a larger longevity benefit than adding a gym session or a serving of vegetables, though all three still matter and often reinforce one another. For clinicians, routinely screening for irregular or insufficient sleep and treating common disorders such as insomnia and sleep apnea could be framed not only as quality-of-life measures but as mortality-reduction strategies.
For policymakers, the next development to watch is whether any U.S. county or state launches a controlled pilot pairing later school start times, workplace schedule reforms, or targeted sleep-disorder treatment with systematic tracking of mortality and other hard outcomes. That kind of experiment would convert a strong observational signal into the causal evidence needed to justify redirecting public-health dollars at scale. Until then, sleep sits just below smoking on the short list of behavioral factors that appear to exert outsized influence on how long, and how well, people live-and it is likely to loom larger in debates over which interventions deserve priority in the years ahead.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
