# How sleep duration is linked to the risk of dying sooner > Large studies suggest a link between sleep duration and mortality risk, though the relationship may depend on how sleep is measured. By The Editors · Sleep · Published 2026-11-03 Canonical: You wake up, squinting at the ceiling. The clock says 5:30 AM. You wonder if those few hours of restless sleep will catch up to you later in life. \[1\] It is a question many people ask when the alarm goes off too early or when they find themselves still awake in the middle of the night. Does the amount of time you spend asleep actually connect to your chance of living longer? \[2\] Large studies of millions of people suggest there is a link, but the pattern is not a straight line. \[1\] ## The J-shaped pattern Most researchers look at a seven-hour window as the lowest-risk baseline. \[1\] When you sleep significantly less or significantly more than that, the risk of dying from any cause begins to climb. \[1\] This pattern forms a J-shape, where the risk rises steeply at the long end, once nightly sleep reaches 10 or 11 hours. \[1\] Every hour matters. \[2\] For every hour of sleep lost below seven hours, the risk of dying increases by about 6%. \[2\] On the other side, sleeping more than seven hours also carries a penalty, with risk rising by about 13% for every hour above that mark. \[2\] ## Long sleep as the stronger mortality signal Long sleep (nine hours or more per night) carries a larger mortality penalty than short sleep. \[4\] This finding, which notes that the editor-in-chief was also the author of the 2025 review, aligns with older independent findings. \[4\] This effect is more pronounced in women. \[4\] ## People and populations The relationship between sleep and death is not the same for everyone. \[3\] Some data suggests that short sleep, which is less than seven hours, is associated with higher mortality for women, but this association does not appear clearly in men. \[3\] However, recent reviews disagree on whether sex modifies the overall risk. \[2\]\[4\] In East Asian populations, including studies from Japan, China, and Korea, the connection for long sleep is stronger. \[6\] In these cohorts, sleeping ten hours or more was linked to a 34% to 48% increase in mortality for both men and women. \[6\] ## A complication in the measurement There is a catch in how we measure sleep. \[7\] Most of these large studies rely on people reporting their own sleep hours, which can be inaccurate. \[7\] When researchers use objective tools like actigraphy (wearable devices that track movement), the "optimal" amount of sleep often shifts lower. \[5\] Studies using these devices suggest the middle ground might actually be between five and six and a half hours. \[5\] All the evidence available is observational, so we cannot definitively say sleep duration causes these trends. There is also the possibility of reverse causation (an underlying, undiagnosed illness is what causes a person to sleep more, rather than the long sleep itself causing the health problem). \[7\] Because of this, some researchers argue that the link between sleep duration and mortality is not yet proven. \[7\] Additionally, the included studies involved cohorts with very different follow-up lengths, which can affect how we interpret the results. \[7\] ## What we know The evidence points toward a consistent pattern in large groups: deviating from a middle ground of sleep is linked to higher mortality. \[1\] Whether that middle ground is seven hours or closer to five depends on how you measure it. \[5\] Until studies can rule out the possibility that disease is driving the sleep, we can only see the pattern. \[7\] For now, the number of hours on your clock is just one more part of a much larger story. \[7\] ## Sources 1. Liu TZ et al., Sleep Medicine Reviews, 2017 2. Yin J et al., Journal of the American Heart Association, 2017 3. Liu TZ et al., Sleep Medicine Reviews, 2017 4. Ungvari Z et al., GeroScience, 2025 5. Kripke DF et al., Sleep Medicine, 2011 6. Svensson T et al., JAMA Network Open, 2021 7. Kurina LM et al., Annals of Epidemiology, 2013 --- From suber.health. This is general information, not medical advice. Quote with a link to the article.