The surplus is real, it's a few hundred calories, and it arrives after 10pm#
Sleep loss makes you gain weight by making you eat more. Not by slowing anything down, not by trapping fat, not by wrecking a hormone — by putting several hundred extra calories a day into you while your energy expenditure sits roughly still. Pooling eleven controlled human experiments (n = 172), partial sleep deprivation raised energy intake by 385 kcal per day compared with the control condition (95% CI, 252 to 517; P < 0.00001), with no significant change in total energy expenditure or resting metabolic rate1. Intake up, burn flat. That is a surplus, and it is the whole mechanism. The wider case for sleep as a weight lever is the pillar; this article is the chain itself, link by link.
The more useful detail is when those calories land. In a laboratory study of 225 healthy adults, the sleep-restricted group ate 552.9 ± 265.8 additional calories between 22:00 and 03:59 — a window that simply does not exist for someone asleep at eleven3. Staying up doesn't just make you hungrier. It issues you extra hours, and those hours have a menu.
Sleep restriction doesn't only change your appetite. It hands you six extra waking hours between 10pm and 4am, and almost nobody spends them eating salad.
The composition of the extra intake is worse than the total#
The canonical single experiment is worth its numbers. Thirty normal-weight adults aged 30–49, who habitually slept 7–9 hours, were run through two inpatient conditions in random order: five nights at 4 hours in bed, five nights at 92. By day 5, they consumed 2,813.6 kcal on short sleep against 2,517.7 kcal on habitual sleep (P = 0.023) — a gap of just under 300 calories, arrived at by subtracting their two published figures. Fat intake rose specifically. Energy expenditure, measured in the same people, didn't move.
The pooled data says the same thing about what changes: alongside the extra 385 kcal came significantly higher fat intake and lower protein intake, with carbohydrate unaffected1. That is a genuinely unlucky combination. Fat carries 9 kcal per gram against protein's 4, so a tired day shifts intake toward the macro that costs the most per bite and away from the one that keeps you full and protects lean tissue. Spaeth's late-night eaters show the same signature: 33.0% of their 22:00–03:59 calories came from fat, against 28.2% during the daytime window.
Six weeks of losing 78 minutes a night#
Everything above is days long. The obvious objection — fine, but does a slightly short sleeper actually get heavier over a real stretch of time? — went largely unanswered for years, because the randomized literature on sleep and body weight was short, small, and inconsistent on weight itself.
It has an answer now. A pooled analysis of two randomized crossover trials put 95 adults through six weeks of adequate sleep and six weeks of mild sleep restriction, separated by a multiweek washout4. The dose was deliberately modest and realistic: sleep fell by 78.4 minutes a night (95% CI, −83.5 to −73.3). Over six weeks, that produced:
| Outcome | Effect of short sleep | 95% CI |
|---|---|---|
| Body weight | +0.45 kg | 0.33 to 0.57 |
| Waist circumference | +0.52 cm | 0.25 to 0.79 |
| Whole-body volume (MRI) | +0.56 L | 0.19 to 0.93 |
| Sedentary time | +17.2 min/day | 11.7 to 22.7 |
| Leptin | +2.03 ng/mL | 0.38 to 3.68 |
Half a kilogram is not a dramatic number, and the authors don't pretend otherwise — they list "effect sizes were modest" among their own limitations, along with a duration likely too short to shift body composition. But look at the confidence intervals: every one of them clears zero comfortably. This is a small effect measured precisely, which is a different thing from a big effect measured badly. Lose an hour and a quarter of sleep a night for six weeks and you can expect to be about half a kilo heavier. Do that for a year and the arithmetic gets your attention.
One row deserves a flag. Leptin — the satiety hormone that the popular account says falls with sleep loss — went up here. Whether that reflects the mild dose, the six-week timescale, or leptin resistance is exactly the sort of thing the appetite hormone story has to reckon with, and it's a good reason to treat the tidy leptin-down-ghrelin-up narrative as less settled than it sounds.
Why a five-day lab number dwarfs a six-week real-world one#
Spaeth's sleep-restricted subjects gained 0.97 ± 1.4 kg in five days. Zuraikat's gained 0.45 kg in six weeks. Twice the gain in a twelfth of the time — which looks like a contradiction and isn't one. The two studies did not test the same thing, and naming the differences is more useful than averaging them:
- The dose isn't comparable. Spaeth cut sleep to 4 hours in bed. Zuraikat trimmed 78 minutes off a normal night. Those are different interventions wearing the same label.
- The food environment isn't comparable. Spaeth's participants lived in a laboratory with food freely available around the clock — a setting engineered to let the appetite signal express itself. Zuraikat's went about their outpatient lives, where a 3am snack requires having bought it.
- Five-day "weight" is not five-day fat. Over that span, water, glycogen, and gut contents dominate the scale.
So the lab number is the effect's ceiling under ideal conditions, and the six-week number is what it looks like in a life. Both point the same way. Neither licenses the other's magnitude.
The part that isn't appetite#
One finding in the six-week data is easy to skim past: sedentary time rose by 17.2 minutes a day (95% CI, 11.7 to 22.7). Tired people sit down more. That is not a metabolic slowdown — it is a behavioral one, and it stacks on top of the intake effect rather than replacing it. It's also a reminder that the movement side of the ledger is more fragile and less generous than most people assume.
What is not happening is the thing most people blame. Across the pooled experiments, neither total energy expenditure nor resting metabolic rate changed with partial sleep deprivation1; St-Onge's inpatients showed the same flat expenditure alongside their extra 300 calories. If you want the full audit of that claim, including the trial that found expenditure going up, it's in does poor sleep slow your metabolism.
So the chain runs: fewer hours asleep → more waking hours, concentrated at night → more eating in those hours, tilted toward fat → a few hundred surplus calories → slow accumulation. Each link is measured. None of them is dramatic. The reason the direction of your appetite changes at all — why it's chips and not chicken — is a reward-circuit story told in why you crave junk food when you're tired.
The practical upshot is unglamorous. A surplus this size is invisible night to night and obvious over a season, which makes it precisely the kind of thing you cannot feel and can only count. And it means the fix for a badly slept week is not a punishing Saturday — it's noticing that the late-night window opened at all.
FAQ#
How many extra calories do you eat when you're sleep deprived?#
About 385 a day, pooled across eleven controlled experiments (95% CI, 252 to 517), with no matching rise in energy expenditure1. A single well-controlled trial found a similar gap — 2,814 vs 2,518 kcal on day 5 of 4-hour versus 9-hour nights2. The range matters more than the midpoint: this is a few hundred calories, not a thousand.
Can one bad night's sleep make you gain weight?#
Not meaningfully, and the scale will lie to you if you look. Body weight moves kilograms on water, glycogen, and gut contents alone, which is why five-day lab studies show larger "gains" (0.97 kg) than six-week trials do (0.45 kg) (Spaeth et al., 2013; Zuraikat et al., 2026). One bad night gives you a surplus of a few hundred calories. It's the repetition that accumulates.
Will short sleep make you gain weight even if your diet doesn't change?#
That's roughly the experiment that was run, and the answer is yes — but the diet does change, which is the point. Ninety-five adults given no dietary instruction, just 78 fewer minutes of sleep a night for six weeks, gained 0.45 kg (95% CI, 0.33 to 0.57) and 0.52 cm of waist4. Nobody told them to eat more. Being awake longer and more tired did that on its own.
Sources#
- Al Khatib HK, Harding SV, Darzi J, Pot GK. The effects of partial sleep deprivation on energy balance: a systematic review and meta-analysis. Eur J Clin Nutr. 2017;71(5):614-624.
- St-Onge MP, Roberts AL, Chen J, et al. Short sleep duration increases energy intakes but does not change energy expenditure in normal-weight individuals. Am J Clin Nutr. 2011;94(2):410-416.
- Spaeth AM, Dinges DF, Goel N. Effects of experimental sleep restriction on weight gain, caloric intake, and meal timing in healthy adults. Sleep. 2013;36(7):981-990.
- Zuraikat FM, Scaccia SE, Cochran JA, et al. Prolonged short sleep and its effect on body weight and composition: a pooled analysis of randomized trials. Ann Intern Med. 2026.
