How Much Sleep Do Adults Really Need?

In 2015 two professional bodies, the American Academy of Sleep Medicine and the Sleep Research Society, asked a panel of fifteen sleep scientists to answer a question that sounds simple: how much sleep does a healthy adult need? The panel reviewed the research and voted in structured rounds until it reached agreement. The answer fits on a single page.

Their recommendation was that adults should sleep “7 or more hours per night on a regular basis”. The same year, a separate panel of eighteen experts convened by the National Sleep Foundation landed on a similar range: seven to nine hours for young adults and adults, seven to eight for older adults, and eight to ten for teenagers.

So the headline is seven. But almost everything interesting about sleep sits in the fine print, and in a question the headline number ignores entirely: not how long you sleep, but when.

What the experts actually agreed on

The consensus statement makes three separate claims, and they deserve to be read separately.

• Regularly sleeping less than seven hours is linked to worse health. The panel listed weight gain and obesity, diabetes, high blood pressure, heart disease and stroke, depression and a higher risk of death, along with weaker immunity, more pain sensitivity, more mistakes and more accidents.
• More than nine hours is harder to read. Sleeping longer than that may be appropriate for young adults, for people recovering from sleep debt and for people who are ill. For everyone else, the panel said plainly that it is uncertain whether it carries any health risk.
• People differ. The statement notes that individual sleep need is shaped by genetic, behavioural, medical and environmental factors.

That last point is easy to skim past. A consensus number describes what works for most people in most studies. It cannot tell you whether you personally do well on seven hours or need closer to nine. Seven is a floor for a population, not a prescription for a person.

Why the numbers are fuzzier than they look

Most of what we know about sleep duration and long-term health comes from large observational studies. In 2010 Francesco Cappuccio and colleagues pooled sixteen of them, covering 27 separate cohorts, almost 1.4 million people and more than 112,000 deaths, with follow-up periods of between four and twenty-five years.

People who reported short sleep had a 12% higher risk of dying during follow-up than people in the reference group. People who reported long sleep had a 30% higher risk. Both ends of the scale, in other words, predicted death.

Before that alarms you, three caveats. First, these are relative risks, and modest ones: a 12% increase on a small yearly risk is a small change in absolute terms. Second, sleep was measured by asking people how long they usually slept, and a remembered average is an estimate, not a measurement. Third, and most important, this is correlation. Long sleep in particular may well be a consequence of illness, low mood or poor-quality sleep rather than a cause of anything. That is precisely why the consensus panel declined to call long sleep harmful.

The panel was confident enough about short sleep to make a firm recommendation. But the exact threshold of seven hours is a judgement made by experts looking at imperfect data, not a line found in nature.

When you sleep matters too

The consensus statement is about duration. It says nothing about timing, and that is where chronotype comes in. Your chronotype is your natural preference for when to sleep and wake, ranging from extreme early types to extreme late types, with most people somewhere in between. It is largely regulated by the body's internal clock and shaped by both genes and environment.

In 2006 Marc Wittmann, Jenny Dinich, Martha Merrow and Till Roenneberg gave the Munich ChronoType Questionnaire, together with questions on sleep quality, mood, and caffeine, nicotine and alcohol use, to 501 volunteers. Late chronotypes stood out. Work and school schedules pulled them out of bed early, they built up a sleep debt through the week, and they paid it back on free days.

The authors called this gap between biological time and social time social jetlag, because it resembles crossing a few time zones every weekend and crossing back on Monday. Links between chronotype, wellbeing and stimulant use were strongest in teenagers and young adults up to 25. The most striking was smoking, which was more common among late types at every age except retirement. The authors argued that these links were most probably driven by social jetlag rather than by chronotype as such.

A later and much larger study by Roenneberg and colleagues found that social jetlag was associated with a higher body mass index, over and above the effect of sleep duration.

Honesty about the design matters here too: these studies are correlational and based on questionnaires, and late types who smoke more may differ in other ways as well. But the idea is sharp. Two people can both average seven hours a night and have very different weeks, if one of them is fighting their own clock five days out of seven.

If you are not sure where you fall, our chronotype test estimates your natural timing, and the test on what kind of tired you are can help separate lost sleep from other kinds of fatigue.

What this means for you

None of this is medical advice, and no article can tell you how much sleep your body needs. The research does suggest a few useful ways to look at your own nights.

• Treat seven hours as a floor to test, not a target to hit. If you regularly sleep less, the consensus suggests taking that seriously even if you feel fine. If you sleep more and still feel tired, the number is not the whole story.
• Compare your work days with your free days. A large gap in when you fall asleep and wake up on the two kinds of day is what social jetlag looks like, and sleeping much longer at the weekend is often a sign of debt built up during the week.
• Remember that long sleep is ambiguous. Needing a lot of sleep can be normal, can be recovery, or can be a signal of something else. The evidence does not show that it is harmful in itself.
• Talk to a professional if it persists. The consensus panel itself recommends that people with concerns about how much they sleep consult a healthcare provider, and that is especially true when tiredness lasts for weeks or gets in the way of daily life.

Seven hours answers a question about populations. The more useful question for you is whether your sleep, taken at the time your body wants it, leaves you rested.