Doctor Medicine logo

How Much Sleep Adults Actually Need, and What to Do When You Can't Get It

Seven hours is the floor for most adults. Here is what the research says, why you can't judge your own tiredness, and how to recover.

By Pravin Kafle
How Much Sleep Adults Actually Need, and What to Do When You Can't Get It

I used to treat sleep as the most negotiable item on my calendar. Meetings were fixed, deadlines were fixed, and sleep was whatever remained. When we started digging into the clinical evidence for this piece, that habit looked a lot less harmless. The research is consistent, and in a few places it is uncomfortable to read.

Below, I walk through nine things we learned about adult sleep need, what happens when the number falls short, and what actually helps when a full night is out of reach. Everything is tied to a source so you can check the claims yourself.

1. The benchmark: seven or more hours, on a regular basis

The most cited answer to "how much sleep do I need?" comes from a joint consensus statement by the American Academy of Sleep Medicine (AASM) and the Sleep Research Society (SRS). A panel of 15 sleep experts spent about a year reviewing the evidence, using a modified RAND Appropriateness Method, and concluded that adults aged 18 to 60 should sleep 7 or more hours per night on a regular basis to promote optimal health (AASM/SRS consensus statement via PMC).

The panel did not stop at a comfort number. It tied short sleep to specific outcomes. Sleeping fewer than 7 hours regularly is associated with weight gain, obesity, diabetes, hypertension, heart disease, stroke, depression, impaired immune function, increased pain, more errors, and a greater risk of accidents (MDedge summary of the guideline).

Dr. Nathaniel Watson, the panel's moderator and later AASM president, was blunt about the lower end. He said the group found six or fewer hours to be "inadequate to sustain health and safety in adults".

Here is how the AASM breaks down sleep by age group (Sleep Education):

Age group Recommended sleep per day
Children 1 to 2 years 11 to 14 hours
Children 3 to 5 years 10 to 13 hours
Children 6 to 12 years 9 to 12 hours
Teens 13 to 18 years 8 to 10 hours
Adults 18 and over 7 or more hours

One detail I like: seven is a minimum, not a target to optimize down to. Plenty of people do better with a bit more.

2. Your personal number lives inside a range

Seven is the floor, but the consensus statement is careful to say that individual sleep need varies. Genetic, behavioral, medical, and environmental factors all shape it, and the biological mechanisms behind sleep need are still being studied (PMC).

The panel also placed no hard ceiling on sleep. Regularly sleeping more than 9 hours may be appropriate for young adults, people paying back sleep debt, and people with illnesses. The statement adds that anyone worried about sleeping too much or too little should talk with a healthcare provider.

So how do we find our own number? The AASM suggests a practical approach: start with the age-based recommendation, then adjust based on how you feel during the day (Sleep Education). Two warning signs it flags are needing caffeine to stay alert and sleeping much later on weekends than on weekdays.

Here is the quick test I use, and I would suggest the same to you:

  • Take a week where your schedule is flexible, such as a vacation.
  • Go to bed when you feel sleepy and let yourself wake without an alarm.
  • Note the average after the first few days, once the initial catch-up sleep fades.

That average is a rough read on your baseline. It is not a clinical measurement, but it is more honest than guessing.

3. Most of us are short, and the numbers are large

If you feel like everyone around you is tired, you are not imagining it. The CDC reports that 36.8% of U.S. adults say they do not get enough sleep (Sleep Education). In a 2024 AASM survey, 54% of respondents said they sleep too little.

Public health officials have been raising this flag for years. When the CDC released a state-by-state analysis based on surveys of more than 444,000 adults, Wayne Giles, MD, who directed the CDC's Division of Population Health, put it plainly: "As a nation we are not getting enough sleep". The same coverage noted that people with steady jobs and regular hours tend to sleep more, while smoking and health problems can take sleep away.

The takeaway for me is that short sleep is a population-level pattern, not a personal failing. That matters, because the fixes work better when we treat sleep as a system (schedule, environment, health) rather than a willpower problem.

4. You are a poor judge of your own sleepiness

This is the finding that changed my behavior. In a landmark 2003 experiment, Hans Van Dongen and David Dinges at the University of Pennsylvania randomized healthy adults to 4, 6, or 8 hours in bed per night for 14 consecutive days. A separate group stayed awake for three days straight (NASA Technical Reports Server abstract).

The results, published in the journal Sleep, were stark:

  • Restricting sleep to 4 or 6 hours produced cumulative, dose-dependent deficits in cognitive performance across all tasks.
  • After two weeks, six-hour sleepers performed at a level the authors compared to up to two nights of total sleep deprivation.
  • The 8-hour group showed no reliable decline (Soon Works research summary).

Here is the part that should make all of us pause. The restricted sleepers reported feeling only slightly sleepy, even as their performance fell. Van Dongen put it this way: "You can't rely on people to judge their own performance capabilities" when they are sleep deprived. In a separate interview he said routine sleep under six hours causes cognitive deficits even if we feel we have adapted to it.

One caveat I want to be fair about: the study ran for 14 nights in a lab. It did not follow people who have slept short for years, so we should not over-read it as a measure of lifelong habits. Still, the subjective-versus-objective gap is the key lesson. Feeling fine at six hours is not evidence that six hours is fine.

5. Fix the schedule before you buy the gadgets

When we ask what helps most with sleep, the unglamorous answer tends to win. Dr. Giles of the CDC pointed to simple lifestyle changes: going to bed at the same time each night, rising at the same time each morning, and removing TVs, computers, and mobile devices from the bedroom (ABC7 New York, quoting the CDC).

Here is how I translate that into a plan:

  1. Pick a fixed wake time. It anchors your body clock more than a bedtime does, because you can control when you get up even if you cannot control when you fall asleep.
  2. Back-calculate a bedtime. If you want 7.5 hours and need 30 minutes to fall asleep, count back eight hours from your alarm. The AASM offers a free bedtime calculator for this (Sleep Education).
  3. Move screens out of the bedroom. Charge your phone across the room. A cheap alarm clock solves the "I need it for my alarm" excuse.
  4. Protect the wind-down. Treat the last 30 to 60 minutes before bed as a boundary, not a spare work window.

None of this is exciting. All of it is backed by public health guidance, and it costs nothing.

6. Recover sleep debt, but don't count on the weekend alone

Almost all of us have tried to "catch up" on Saturday. Does it work? The evidence is mixed in a useful way.

On the positive side, the AASM points to a study finding that adults who regularly sleep under six hours on weekdays may lower their risk of heart disease by adding extra sleep on weekends. Van Dongen has also said that sleeping in on the weekend is an effective way to recover from sleep debt (The Daily Pennsylvanian).

On the cautionary side, the AASM warns that leaning too heavily on weekend sleep can throw off your internal clock (Sleep Education). And recovery is not instant. One review of the Van Dongen data notes that groups on restricted sleep had not returned to baseline performance after three recovery days (Soon Works).

A rule I follow: keep the weekend wake-up within about an hour or two of weekday timing, and spread recovery across several nights. If I need an extra 90 minutes, I try to take it as an earlier bedtime on a few evenings, not a long lie-in once a week. That approach is my own practical adaptation, not a clinical rule, so adjust it to your life.

7. Use caffeine as a tool, not a crutch

Caffeine is the most common countermeasure, and the research supports using it with restraint. Van Dongen described caffeine on a limited, occasional basis as a good countermeasure for sleep loss, while stressing that the real fix is more sleep.

The AASM lists "you need caffeine to stay alert" as a sign that you may not be sleeping enough (Sleep Education). I read that as a useful self-check. If my third coffee is doing the work my pillow should have done, the problem is upstream.

Practical guardrails I use:

  • Stop caffeine at least six to eight hours before bed. This is a common clinical suggestion, and your sensitivity may differ.
  • Use caffeine for specific high-stakes moments, such as a long drive after a bad night, not as a daily baseline.
  • Treat a short nap as another option when you can fit one in. Keep it brief and early enough in the day that it does not push your bedtime later.

If you are drowsy while driving, caffeine is not a safe plan. Pull over. Short sleep is tied to a greater risk of accidents in the AASM guideline, and that risk is not theoretical.

8. When it is insomnia, ask for CBT-I before pills

Sometimes the problem is not a lack of time. You lie in bed, tired, and sleep will not come. An estimated 6 to 10 percent of American adults meet diagnostic criteria for insomnia disorder, according to the American College of Physicians (ACP) guideline as reported by Psychiatric News.

The ACP recommends cognitive behavioral therapy for insomnia (CBT-I) as the first-line treatment for chronic insomnia, ahead of medication (ACP newsroom). CBT-I combines cognitive therapy about sleep with behavioral tools such as sleep restriction and stimulus control, plus sleep education.

Why behavior first? The ACP's president at the time noted that CBT-I "is likely to have fewer harms" than drugs. There is also a practical numbers argument. A typical person with chronic insomnia takes 60 to 70 minutes to fall asleep. Medications in the reviewed short-term studies cut that by roughly 10 to 20 minutes, so people still waited 40 to 60 minutes (The Hospitalist summary of the ACP guideline).

For CBT-I results, one summary pooling 20 randomized trials with 1,162 adults reports that it cut time to fall asleep by about 19 minutes and time awake overnight by about 26 minutes (RISE Science). That source is a company blog summarizing the trials, so I would treat the figures as a helpful illustration and confirm them with a clinician. Also notice that total sleep time barely changed in that summary, which is a reminder that CBT-I often improves how efficiently you sleep rather than adding hours.

If CBT-I alone does not work, the ACP advises a shared decision with your clinician about adding short-term medication. The organization notes that medications should ideally be used for no longer than four to five weeks while CBT-I skills handle the longer term (Sleep Review).

9. Know when to stop self-managing

I like practical hacks, but there are limits. The AASM says that if you are worried you sleep too little or too much, you should talk with a healthcare provider (PMC). It also notes that an AASM-accredited sleep center can diagnose and treat sleep disorders that may be hurting your rest (Sleep Education).

Signs that I would not try to fix on my own:

  • You sleep a full seven or eight hours and still wake exhausted.
  • A partner notices loud snoring, gasping, or pauses in your breathing.
  • You have trouble sleeping most nights for several weeks.
  • Daytime sleepiness is affecting your driving or your work.

None of these automatically means a disorder, but each is a reason to ask someone with training.

The short version: what I would do this week

If you only take one page from this article, here is the plan I would follow, based on the evidence above:

  1. Treat 7 hours as a minimum, and aim for 7 to 9 depending on how you feel.
  2. Don't trust "I feel fine" after weeks of short sleep. The Penn data shows the feeling lags the damage.
  3. Set a fixed wake time and a bedtime that matches your target hours.
  4. Keep screens out of the bedroom.
  5. Recover debt across several nights, not one big weekend sleep.
  6. Use caffeine sparingly, and never as a fix for drowsy driving.
  7. If sleep will not come, ask about CBT-I before medication.
  8. Book a clinician visit if sleepiness or breathing problems persist.

We are all going to have short nights. A newborn, a deadline, or a bad week will happen. The goal is not perfection. It is to make short sleep the exception, notice the early signs, and recover deliberately.

This article is general information, not medical advice. If you have concerns about your sleep or health, please speak with a qualified healthcare provider.