Few questions come up more often in my practice than this one: how many hours of sleep do I actually need? For years, the honest answer has been a range and a shrug, somewhere around 7 or 8 hours, give or take. A large study published in Nature in May 2026 sharpens that answer considerably. Drawing on data from roughly 500,000 adults in the UK Biobank, aged 37 to 84, researchers mapped how nightly sleep duration correlates with health across the whole body and found a clear window in which the body fares best.
The headline is simple and worth stating up front. There is a sleep sweet spot; it sits in a fairly narrow band, and straying too far in either direction carries real consequences, including a higher risk of disease and death.
How the Study Measured the Body’s Response to Sleep
To see how sleep affects the body, the researchers did not rely on a single yardstick. They developed 23 distinct measures, each capturing the condition of a specific organ or system at a different level of biology. Seven came from MRI imaging of the brain, heart, liver, pancreas, kidney, spleen, and fat tissue. Eleven came from proteins circulating in the blood that reflect activity in the brain, lungs, liver, immune system, heart, kidneys, and skin. Five came from the small molecules of metabolism.
For each measure, the team calculated how far an organ’s estimated biological condition diverged from what would be expected for the person’s actual age. That gave them a sensitive, organ-by-organ readout of how the body was holding up, which they could then compare with each person’s sleep duration. In effect, it is a full-body panel for testing the effects of sleep.
The U-Shaped Curve: Why Both Too Little and Too Much Hurt
When the researchers plotted sleep duration against each organ measure, the same shape appeared again and again: a U. The body did worst at both ends of the curve. People sleeping too little fared poorly. People sleeping too much fared poorly. The best readings, the bottom of the U, sat in a narrow band in the middle.
Across the nine organ measures that showed this pattern most clearly, the sweet spot fell between 6.4 and 7.8 hours of sleep, varying slightly by organ and by sex. Most measures bottomed out between 6 and 8 hours, and the study’s overall best estimates landed in that range. Different tissues showed slightly different optima, but they all pointed to the same general neighborhood: somewhere around 6.5-8 hours a night.
This is the opposite of a “more is always better” mindset. Sleep behaves like a nutrient. Too little starves the system, too much signals that something may be wrong, and the body does best in the middle.
Short Sleep and Long Sleep Are Not the Same Problem
One of the most useful findings is that short sleep and long sleep are both harmful, but in different ways and through different routes. They are not two versions of the same problem.
Short sleep (under 6 hours): a broad, direct effect
Insufficient sleep showed wide, direct associations with disease across nearly every system. Combining genetic analyses with years of follow-up on actual diagnoses, short sleepers carried elevated risk of ischaemic heart disease, heart failure, and coronary atherosclerosis; type 2 diabetes and obesity; high blood pressure and abnormal heart rhythms; depression, anxiety, and insomnia; asthma and chronic obstructive pulmonary disease; and a cluster of digestive disorders including reflux. Short sleep, in other words, appears to press directly on the whole body.
Long sleep (over 8 hours): a narrower, more indirect signal
Excessive sleep told a more focused story, clustering around the brain and conditions such as depression, schizophrenia, and bipolar disorder. The harm from long sleep also seemed to operate more indirectly. In the study’s analysis of late-life depression, the link with long sleep ran largely through changes in the brain, which accounted for about 62% of the effect, with fat-tissue and liver measures adding roughly another 24%. This supports a point worth keeping in mind: in older adults, sleeping too much is often less a cause of illness than an early sign of illness that has not yet announced itself.
The Mortality Numbers
The associations were not abstract. Tracking deaths from any cause over years of follow-up, the study found that both extremes raised the risk of dying. Short sleep was associated with a 50% higher risk of death from any cause (hazard ratio 1.50, 95% confidence interval 1.44-1.55). Long sleep was associated with a 40% higher risk (hazard ratio 1.40, 95% confidence interval 1.36-1.44). These are large, consistent signals from one of the best-studied populations in the world.
Is Sleep the Cause, or Just a Marker?
This is the question that decides whether sleep is worth acting on, so it deserves a careful answer. Most observational research cannot prove direction. To probe it, the authors used Mendelian randomization, a genetic method that helps separate cause from consequence. The result was encouraging: the analysis found no strong evidence that disease drives abnormal sleep, supporting the view that sleep duration is a modifiable risk factor rather than simply a byproduct of being unwell. The authors are appropriately cautious and cannot fully rule out reverse causality, especially for long sleep. But on balance, the evidence points toward sleep as something worth changing, not just measuring.
That is genuinely good news. Unlike many risk factors, sleep is a habit. It responds to light, routine, and discipline. It is one of the few health levers you can begin adjusting tonight.
What This Means for You
I would distill the practical lesson into a single sentence: aim for the middle, and treat a drift toward either extreme as information. If you are routinely sleeping fewer than 6 hours, that is a target for active change, because the data link short sleep directly to cardiovascular, metabolic, and mental health risks. If you have started sleeping more than 8 or 9 hours and still feel unrefreshed, do not simply write it off; treat it as a prompt to look for an underlying issue, from depression to sleep apnea, and bring it to your physician.
For most healthy adults, the destination is 6.5-8 hours of consistent, quality sleep. Protect it the way you would protect a medication schedule. Keep a steady bedtime, get morning light, move your body during the day, and guard the hour before bed from screens and stimulation. None of this is glamorous, but the data suggest it is among the most broadly protective things you can do.
A Word on the Study’s Limits
Good science states its own boundaries. Sleep duration here was self-reported, which is less precise than a sleep-lab measurement and can introduce error. The design was largely cross-sectional, a snapshot rather than a film, which limits firm causal claims. The participants were predominantly of European ancestry, so the exact numbers may not translate perfectly to other populations. And a single blood draw can miss the normal day-to-day fluctuations in proteins and metabolites. None of this overturns the core finding, which held up consistently across organs and across three independent technologies, but it does mean the precise sweet-spot hours are best read as a well-supported guide rather than a rigid prescription.
The Bottom Line
This study gives the old advice a firmer footing. There really is a window where the body does best, it sits around 6.5-8 hours a night for most adults, and both too little and too much sleep push you off the bottom of the curve and toward higher disease and mortality risk. The encouraging part is that sleep is changeable. Honoring the rest your body needs may be one of the simplest, most far-reaching steps you can take for your health.
