Chronically short sleep is associated with a slightly shorter life and, above all, fewer years in good health. But no study can tell you how many days your 5-hour night costs you, and the measure that best predicts mortality may not be duration: it may be regularity.
On one side, the entrepreneur who “functions perfectly on 5 hours.” On the other, the video claiming that one sleepless night destroys your brain. The data tell a less spectacular, more useful story.
Why nobody can put a number on your loss
No trial has made people sleep 5 or 8 hours for thirty years to count the deaths. That would be impossible and unethical. Everything we know comes from cohorts: thousands of people are asked about their sleep, then researchers observe who becomes ill or dies in the years that follow.
These studies show robust associations, repeated across many countries. They never fully separate sleep from the factors that go with it. Someone sleeping 5 hours may work nights, smoke, drink more, live with financial stress or have an undiagnosed illness. Keep that filter in mind for every number that follows.
The numbers from large cohorts
Sleeping less than 7 hours: a modest but consistent risk
The meta-analysis by Cappuccio and colleagues brought together 16 studies and more than 1.3 million people followed for 4–25 years. Short sleep was associated with a 12% higher risk of death. A 2025 meta-analysis covering 79 cohorts found a similar magnitude: a 14% higher risk below 7 hours a night, compared with 7–8 hours.
For perspective: this is a real effect at the population level, much smaller than that of smoking. It is not an individual death sentence.
The “good sleeper” score: 4.7 years for men
A US cohort of 172,321 adults developed a score based on five criteria: sleeping 7–8 hours, usually falling asleep without difficulty, rarely waking during the night, not taking sleeping pills regularly and feeling rested on waking.
At age 30, men who met all five criteria had an estimated life expectancy 4.7 years longer than those who met none or just one. For women, the gap was 2.4 years. Median follow-up was only 4.3 years, and life expectancy was extrapolated from that follow-up.
This figure does not mean that sleeping well adds 4.7 years. It describes an entire profile: on average, good sleepers also have fewer illnesses and more favorable living conditions.
What disappears first: years in good health
An analysis of three European cohorts examined years lived in good health between ages 50 and 75. Those sleeping 7–8.5 hours could expect around 19 healthy years out of those 25 years. Those sleeping less than 7 hours or more than 9 hours lost 1–3 of them. Severe sleep problems cost even more.
This is the most important message: chronic sleep deprivation seems less about bringing death forward than bringing forward the years of fatigue, metabolic disease and medication.
Regularity: the lever nobody measures
An average of eight hours does not guarantee a stable body clock. Sleeping from 11 p.m. to 7 a.m. during the week, then from 3 a.m. to 11 a.m. at weekends, is like making a round trip across four time zones every week.
The study by Windred and colleagues measured regularity using wrist-worn sensors in 60,977 UK Biobank participants. The most regular sleepers had a 20–48% lower mortality risk than the most irregular, depending on the model, and regularity predicted mortality better than average duration.
The usual limitation applies: people with regular schedules also often work daytime hours and have more stable lives. But this is consistent with what we know about the body clock, and it is the cheapest lever available.
The benchmarks I take from the recommendations and these data, not from a trial:
| Variable | Benchmark |
|---|---|
| Wake-up time | Variation of less than 30 minutes from day to day |
| Weekend shift | Less than one hour |
| Outdoor light | Within an hour of waking |
| Duration | 7–9 hours for most adults |
Wake-up time is the anchor. Morning light · FR reinforces it, and the weekend shift · FR is often what breaks it.
Sleeping too much: more symptom than cause
Cohorts find an even higher risk among people sleeping more than 9 hours: 30% in Cappuccio’s analysis, 34% in the 2025 meta-analysis. Concluding that long sleep is toxic would be a mistake.
Long sleep often accompanies an existing problem:
- depression
- inflammatory disease or undiagnosed cancer
- sleep apnea that fragments sleep
- frailty, a sedentary lifestyle or medication effects
This is called reverse causation: illness increases time in bed, not the other way around. Sleeping 10 hours after the flu or a competition is nothing to worry about. Needing 10 hours every night and still feeling tired is. The guide “always tired despite 8 hours” · FR helps you look for the cause.
Invisible Debt and the catch-up myth
Accumulated sleep deprivation has a sneaky property: after a few days, you stop feeling it, even as your performance continues to decline. This is what I call Invisible Debt.
On the metabolic side, the effect is measurable in the lab. In a randomized trial in women, slightly shortening sleep for six weeks reduced insulin sensitivity, independently of weight change. Recovery nights restore some of it, not necessarily all.
Your debt is not a bank account you can pay twelve hours into on Sunday. An extreme lie-in also shifts your body clock, so you go to bed late on Sunday night and start the week in deficit again. What works better:
- Keep your wake-up time stable, even at weekends.
- Move bedtime earlier by 15–30 minutes over several nights.
- Add a short, 20-minute nap before 3 p.m. if needed.
- Ease off very intense workouts during the recovery period.
The sleep debt calculation · FR gives you a number-based method.
What really protects you, in order of priority
Not all sleep hacks are equal. A silk eye mask, a cooling mattress or a glycine supplement can improve comfort. None makes up for 5-hour nights, chaotic schedules or untreated sleep apnea.
| Priority | Lever | Why |
|---|---|---|
| 1 | Get screened for sleep apnea if you snore, wake with a dry mouth or feel sleepy during the day | Associated with a substantially higher cardiovascular risk, and treatable |
| 2 | Give yourself a 7–9-hour sleep window | The foundation of the recommendations |
| 3 | Stabilize your wake-up time | The best predictor in sensor-based data |
| 4 | Stop caffeine early and limit alcohol | Both fragment sleep |
| 5 | Treat chronic insomnia | Cognitive behavioral therapy comes before sleeping pills |
For sleep apnea, start with the guide to symptoms and testing · FR. See a doctor if you feel sleepy while driving, if someone notices pauses in your breathing during sleep, or if insomnia lasts more than three months despite stable schedules.
The takeaway
Science cannot tell us that each lost hour takes a precise amount of life away. It says something more actionable: chronically short, irregular, fragmented or apnea-disrupted sleep is associated with more illness and fewer healthy years. The lowest-risk range is around 7–8 hours, with stable schedules.
If you changed just one thing tomorrow, make it your wake-up time: the same seven days a week.
Main sources
The abstracts of the meta-analyses, cohorts and trials below were checked on PubMed:
