“Get your eight hours” is some of the most repeated advice in the “stay young” genre. The research on sleep and lifespan is more interesting than that, and in one respect it points the other way: sleeping too much is linked to higher mortality risk as well. Newer data also suggest that when you sleep, and how consistently, may matter as much as how long.
This article covers what large studies actually show about sleep duration and how long people live, why the relationship is harder to interpret than headlines suggest, and what you can reasonably do with it. It’s part of our complete, evidence-based guide to staying young, which covers the habits with the strongest links to longevity.
The big picture: a U-shaped curve, not “more is better”
The most widely cited evidence comes from a 2010 systematic review and meta-analysis by Francesco Cappuccio and colleagues, published in the journal Sleep. They pooled 16 prospective studies covering 27 separate cohorts: 1,382,999 adults followed for between 4 and 25 years, with 112,566 deaths recorded.
Compared with people sleeping a “normal” amount (usually around 7 hours a night), they found that:
- Short sleepers had a 12% higher risk of death over the follow-up period (pooled relative risk 1.12).
- Long sleepers had a 30% higher risk (pooled relative risk 1.30).
That’s the U-shape: risk is lowest in the middle and rises on both sides. A later dose-response meta-analysis by Yin and colleagues in the Journal of the American Heart Association (2017) put more detail on the curve. The lowest risk was at about 7 hours a day. Below 7 hours, all-cause mortality risk rose about 6% for each hour less. Above 7 hours, it rose about 13% for each extra hour.
Two things stand out. First, the sweet spot is closer to 7 hours than 8 or 9. Second, the long-sleep side of the curve is steeper than the short-sleep side. That second finding is also the easiest to misread, as the next sections explain.
Why short sleep is probably a real risk factor
Mortality studies are observational, so they can’t prove on their own that short sleep causes earlier death. For short sleep, though, controlled experiments point in the same direction, and they show plausible ways it could do harm.
Metabolism and blood sugar
In a classic 1999 study in The Lancet, Karine Spiegel, Rachel Leproult and Eve Van Cauter limited 11 healthy young men to 4 hours in bed for 6 nights, then compared them with the same men after a recovery period of 12 hours in bed. During the sleep-debt week, glucose tolerance was significantly worse, and hormone measurements shifted too. It was a small, short, extreme experiment, but it showed that a few nights of heavy sleep restriction can measurably disrupt metabolism in healthy young people. Poor glucose control is one route to type 2 diabetes and cardiovascular disease.
Immune function
A 2015 study by Aric Prather, Sheldon Cohen and colleagues, also published in Sleep, measured sleep objectively with wrist actigraphy (a movement-tracking device) for a week in 164 healthy adults. They then deliberately exposed them to a cold virus under quarantine. People who had slept 6 hours or less a night were more than four times as likely to develop a clinical cold as those sleeping more than 7 hours. A cold isn’t a mortality risk, but the study is a clean example of short sleep weakening the body’s defenses in a controlled setting.
This fits the consensus position. In 2015, the American Academy of Sleep Medicine and the Sleep Research Society reviewed the evidence and jointly recommended that adults (18–60) sleep 7 or more hours per night on a regular basis. Below that, the panel found consistent links with weight gain and obesity, diabetes, high blood pressure, heart disease and stroke, depression, and impaired immune function.
Why long sleep is harder to interpret
The 30% higher risk for long sleepers tempts people to set an alarm and cut their sleep short. That’s probably the wrong lesson.
Most researchers, including Cappuccio’s team, think long sleep is more likely a marker of poor health than a cause of it. Many conditions make people sleep longer or spend more time in bed: depression, undiagnosed illness, chronic inflammation, heart failure, sleep apnea, low physical activity, and some medications. Anyone already on a path toward serious illness may be sleeping 9–10 hours because of that illness. In a cohort study, that looks like long sleep predicting death.
This is called reverse causation. Studies try to adjust for known health conditions, but hidden illness is hard to remove fully. Unlike short sleep, long sleep has no body of controlled experiments showing that sleeping more causes harm.
The practical takeaway: if you naturally wake up rested after 7–8 hours, you have nothing to fix. If you’ve noticed that you need much more sleep than you used to, and still feel unrefreshed, mention it to a doctor. It may be a signal worth checking.
The measurement problem most headlines skip
Nearly all the large mortality studies above relied on self-reported sleep: one survey question like “How many hours do you usually sleep?” That matters more than it sounds.
A 2008 study by Diane Lauderdale and colleagues in Epidemiology compared self-reports with wrist actigraphy in 669 middle-aged adults from the CARDIA study. On average, people overestimated their sleep by nearly an hour. So when a cohort study says “7 hours,” the people in that group may really have been sleeping closer to 6. The error also isn’t the same for everyone, which blurs where the true “sweet spot” lies.
None of this erases the U-shaped pattern, which has been replicated many times. It does mean the exact numbers (“7 hours, not 6.5”) are less precise than they look, and that “time in bed” and “time actually asleep” are different things.
Consistency may matter as much as duration
One of the most useful recent findings avoids the self-report problem altogether. In a 2024 study in Sleep, Daniel Windred and colleagues analyzed over 10 million hours of wrist accelerometer data from 60,977 UK Biobank participants (mean age about 63). They measured sleep regularity, meaning how similar each person’s sleep-wake pattern was from one day to the next, then tracked deaths over the following years.
Compared with the least regular sleepers, people in the more regular groups had a 20–48% lower risk of all-cause mortality. The pattern held for cancer and cardiometabolic deaths too. When the researchers compared models directly, sleep regularity predicted mortality more strongly than sleep duration did.
This is still observational, and irregular sleep can also reflect shift work, illness, or disrupted routines. But it’s a strong hint that a steady schedule may be at least as important as hitting a particular number of hours. It’s also something most people can actually change.
What to actually do with this
Here’s a sensible reading of the evidence, without overstating it:
- Aim for about 7 hours of actual sleep, which for most people means 7.5–8 hours in bed. The AASM/SRS guidance of “7 or more” is the best-supported target for adults.
- Keep your schedule consistent, weekends included. Going to bed and waking at similar times is the single change the regularity data most directly support.
- Don’t cut sleep to “avoid long sleep.” If long sleep is mostly a marker of illness, forcing yourself to sleep less won’t bring the risk down.
- Treat big changes in sleep need as information. A new need for much more sleep, loud snoring, or waking unrefreshed are worth raising with a clinician. Sleep apnea in particular is common and treatable.
- Stack it with the other habits. Sleep interacts with the rest of the longevity picture. Regular exercise tends to improve sleep quality, and good sleep supports recovery from training. See our deep dives on VO2 max and exercise intensity and grip strength and muscle mass for the fitness side of the same equation.
If you mostly feel tired even after a full night, our explainer on why sleep matters for feeling (and aging) well covers everyday sleep quality in more depth.
Frequently Asked Questions
Is 6 hours of sleep enough to live a long life?
For most adults, the evidence suggests 6 hours is on the short side. Meta-analyses put the lowest mortality risk at about 7 hours, and expert guidance recommends 7 or more hours a night for adults. A small minority of people may genuinely need less, but regularly getting 6 hours or fewer is linked to higher risk of several chronic diseases.
Is sleeping 9 or 10 hours bad for you?
Long sleep is linked to higher mortality in observational studies, but most researchers think it’s largely a marker of underlying illness, depression, or inactivity rather than a direct cause. If you’ve always slept long and feel well, there’s no evidence you should cut back. If your sleep need has recently gone up, talk to a doctor.
Does a regular sleep schedule really matter?
Increasingly, the data say yes. A 2024 UK Biobank study using wrist-worn activity trackers in about 61,000 adults found that more regular sleepers had 20–48% lower all-cause mortality risk than the least regular. Regularity predicted mortality more strongly than duration did.
How is this different from sleep and skin aging?
This article is about whole-body longevity and mortality risk. How sleep affects collagen, wrinkles, and skin appearance is a separate question, covered in our article on sleep and skin aging as part of the anti-aging guide.
This article is for informational purposes only and is not medical advice. If you have persistent sleep problems, excessive daytime sleepiness, or symptoms of sleep apnea, consult a healthcare provider.
