How much sleep do adults need? What the evidence says, by age

Most adults do best on somewhere between 7 and 9 hours of sleep a night. That is the range set by the National Sleep Foundation, a US body whose expert panel reviewed the evidence and recommended 7 to 9 hours for adults and a slightly tighter 7 to 8 hours for older adults (National Sleep Foundation). The American Academy of Sleep Medicine and the Sleep Research Society, the two main professional bodies for sleep doctors and researchers in the US, put it as a floor rather than a range: adults should sleep 7 or more hours a night on a regular basis (AASM/SRS consensus).

Ireland has no official adult figure. The HSE (Health Service Executive, the public health service) publishes sleep guidance for babies and children but not an hours-per-night target for adults, so the numbers above come from American guidance. They are consensus positions from expert panels, and the evidence behind them is mostly large observational studies, which show a pattern but cannot prove that sleep itself is the cause. What follows is what those studies found, how much weight they bear, and what to do if you are outside the range.

Looking for younger age groups? See our child and teenager sleep guide for the recommended ranges from toddler years through age 18.

How many hours of sleep do adults actually need?

The best pooled evidence comes from prospective cohort studies. These follow large groups of healthy people for years and record who gets sick or dies. A 2017 meta-analysis (a study that combines the results of many earlier studies) found a U-shaped curve for death from any cause, with the lowest risk at about 7 hours a day. Each hour below 7 was linked to a 6% higher risk, and each hour above 7 to a 13% higher risk (Journal of the American Heart Association).

The same shape shows up for type 2 diabetes. A meta-analysis covering 482,502 people found the lowest risk at 7-8 hours a day, with a 9% higher risk for each hour shorter and a 14% higher risk for each hour longer (Diabetes Care). An umbrella review, which pools meta-analyses rather than individual studies, looked across 69 of them and found the same pattern for heart disease, stroke, depression and several other conditions (Sleep & Breathing).

Two cautions. Nearly all of these studies rely on people reporting their own sleep, which is unreliable. And they are observational, so they cannot separate sleep from the things that travel with it, such as illness, shift work or low income. The consensus statement itself says sleep need varies from person to person and “is influenced by genetic, behavioral, medical, and environmental factors” (AASM/SRS consensus). Read 7 to 9 as the band most people fall into, not a target every individual must hit.

Does it change with age — and do women need more than men?

The recommended range narrows a little with age. The National Sleep Foundation panel gives 7 to 8 hours for older adults rather than 7 to 9 (National Sleep Foundation). What older people actually get is a different question. A 2004 meta-analysis of 65 sleep-lab studies covering 3,577 healthy people aged 5 to 102 found that in adults, total sleep time, deep sleep and dreaming sleep all fell with age, while time taken to fall asleep and time awake during the night rose. After 60, the only measure that kept falling was sleep efficiency, meaning the share of time in bed actually spent asleep (Sleep). So broken, lighter sleep in later life is normal. It does not mean you need more hours in bed, and the guidance does not raise the target with age. That analysis is old and drew on healthy volunteers in sleep labs, so it describes what healthy people do sleep, not what anyone needs.

On women and men, the honest answer is that the evidence does not support a different target. The National Sleep Foundation issued one range for adults with no separate figure by sex (National Sleep Foundation), and the mortality curve in the 2017 meta-analysis “varied little by sex” (Journal of the American Heart Association). The idea that women need more sleep is unproven rather than disproven.

What happens if you regularly get too little?

The short-sleep signal is consistent. A 2017 review pooled 153 studies with 5,172,710 participants and found short sleepers had a 12% higher death rate, along with higher rates of diabetes, high blood pressure, heart disease and obesity. The extra mortality risk rose in a straight line once sleep dropped below six hours (Sleep Medicine). The AASM/SRS statement lists the same outcomes: weight gain, diabetes, high blood pressure, heart disease and stroke, depression, and a higher risk of death (AASM/SRS consensus).

The causal case is weaker than that list makes it sound, though not absent. A study of 451,420 people in UK Biobank (a long-running British research project that holds health records, lifestyle questionnaires and genetic data on a very large group of middle-aged and older volunteers) used Mendelian randomisation, a method that uses the genes people are born with to test whether a trait itself drives an outcome, and found no causal link between short sleep and mortality (Rejuvenation Research). A meta-analysis of 72 studies on inflammation gives a mixed picture. Shorter sleep was linked to slightly higher levels of C-reactive protein (CRP, a marker of inflammation in the blood) but not to interleukin-6, another inflammatory marker. The extreme of short sleep showed no link to either, and cutting people’s sleep in experiments did not raise any marker (Biological Psychiatry). Short sleep is a reliable marker of higher risk. How much of that risk the sleep itself causes is still unsettled.

Where trials exist, they do show that more sleep changes behaviour. In one randomised trial, 80 overweight adults aged 21 to 40 who habitually slept under 6.5 hours were coached to extend their time in bed to 8.5 hours. They slept about 1.2 hours longer a night and ate 270 kcal less a day, without being told to diet (JAMA Internal Medicine). That is one small, short trial in young adults, so it shows that short sleepers eat less when they sleep more, not that sleeping more makes the general population lose weight.

Can you sleep too much, or is long sleep a symptom rather than a cause?

On paper, long sleep looks worse than short sleep. In the mortality meta-analysis the risk rose 13% per extra hour over 7, against 6% per hour under (Journal of the American Heart Association). In UK Biobank, people reporting long sleep had a hazard ratio of 1.735 for death during follow-up, meaning their death rate was 1.735 times that of normal sleepers, compared with 1.246 for short sleep (Rejuvenation Research).

The same study then tested causation with Mendelian randomisation and found no causal association between long sleep and mortality. That is the pattern you would expect if long sleep is mostly a sign of something else, such as depression, heart failure, cancer or chronic pain, rather than a cause of harm in itself. The inflammation meta-analysis points the same way: long sleep was linked to higher levels of both inflammatory markers, which is what illness looks like in the blood (Biological Psychiatry). The professional bodies are careful here. Their statement says sleeping more than 9 hours may be appropriate for young adults, people recovering from lost sleep and people with illnesses, and that “for others, it is uncertain whether sleeping more than 9 hours per night is associated with health risk” (AASM/SRS consensus).

For the reader, that means two things. If you have always slept nine or ten hours and feel well, there is no good evidence you are harming yourself. If you have recently started needing much more sleep than before, or you sleep long and still wake exhausted, treat that as a reason to get checked rather than a habit to break. The only sleep habit in the UK Biobank analysis that showed a possible causal link to mortality was daytime napping. The direction of that result was repeated in FinnGen, a separate Finnish genetic dataset, so it has some independent support, but it rests on one genetic analysis and is a lead rather than a settled finding (Rejuvenation Research).

How do you tell if you’re getting enough?

The number on a sleep tracker matters less than how you function. There is also evidence that when you sleep may matter as much as how long. A UK Biobank study that tracked 60,977 people with wrist devices found that those with the most regular sleep and wake times had a 20%-48% lower risk of death than the least regular, and that regularity predicted mortality better than duration did (Sleep). That is one observational study built on a short stretch of device data, but it fits the guidance to keep consistent hours.

Signs you are probably getting enough:

• You fall asleep within a reasonable time and mostly stay asleep.
• You can get through the afternoon without fighting to stay awake.
• You do not sleep hours longer at the weekend to catch up.
• You are not relying on caffeine to feel normal by mid-morning.

If several of those are missing, you may be under-sleeping even if your hours look fine. The HSE defines insomnia as having difficulty sleeping 3 or more times a week for more than 3 months (HSE). It is common here. A survey of 1,110 Irish adults designed to be representative found 15.0% screened positive for insomnia disorder, the most common of the 12 mental health conditions measured (Epidemiology and Psychiatric Sciences). That is a self-report screening figure gathered during the COVID period, not a diagnosis rate, but it suggests a lot of people reading this are affected.

What actually helps if you’re not — and when should you see a GP?

The HSE’s first-line advice is behavioural: go to bed and wake up at the same time every day, be active during the day, relax for at least 1 hour before bed, keep the bedroom dark and quiet, and avoid screens right before bed, big meals within 2 hours of bedtime, and alcohol, tea or coffee within 4 hours of bedtime (HSE). The regularity study above gives a reason to take the fixed-times advice more seriously than it usually gets.

If that is not enough, the treatment with the strongest evidence is CBT for insomnia (CBTi: cognitive behavioural therapy for insomnia, a structured course that changes the habits and thoughts keeping you awake, including limiting time in bed to the time you actually sleep). A meta-analysis of 20 randomised trials with 1,162 adults found that after treatment people fell asleep 19 minutes faster, spent 26 minutes less awake during the night, and their sleep efficiency improved by 9.91 percentage points. Total sleep time rose by only 7.61 minutes, a change too small to be sure of (Annals of Internal Medicine). So CBTi makes sleep more solid rather than much longer, and those trials excluded people whose insomnia came with other medical or psychiatric conditions.

There is also evidence that fixing sleep helps mood. A randomised trial of 3,755 students at 26 UK universities compared a digital CBTi course with usual care. At 10 weeks the course cut insomnia substantially and produced small reductions in paranoia and hallucinations (Lancet Psychiatry). The mental health effects were small and the participants were students, so do not read it as a cure for anything, but it is a large trial showing that sleep treatment reaches beyond sleep.

In Ireland, the HSE says to talk to a GP or mental health professional if you have had sleep problems for a month and changing your habits has not helped, if you are finding it hard to cope, or if insomnia is affecting your day-to-day life. Your GP may recommend CBTi and can refer you to a free guided CBT service run by the HSE with a provider called Silvercloud, an online programme you work through at home with support from a therapist (HSE). Go sooner if you have started sleeping far more than usual, or if you sleep long and still feel drained, because the sleep is more likely to be a symptom than the problem.

The short version: aim for roughly seven to nine hours, keep your times steady, judge yourself by how you feel by mid-afternoon rather than by a number, and treat a big change in either direction as something to raise with a GP.