Ghrelin rises about a seventh of a standard deviation#
Short sleep does raise ghrelin, the stomach hormone that signals hunger to the brain. Pooled across 21 studies and 2,250 people, short sleepers carry higher ghrelin than adequate sleepers by a standardized mean difference of 0.14 (95% CI 0.03 to 0.25)1. That is a real effect and a small one — roughly a seventh of a standard deviation, the kind of shift you can detect with 2,250 people and could never feel as a single person on a single morning.
Which sits awkwardly beside the number everyone quotes. The figure that made this literature famous is a 28% ghrelin rise, and the gap between "28%" and "SMD 0.14" isn't a rounding problem. It is the difference between one dramatic protocol and everything that came after it. The sleep and appetite hormones pillar audits the whole cascade; this article stays on ghrelin and asks the narrower question — why do careful trials of the same hormone, in the same kind of participant, keep landing in different places?
The headline number came from an all-nighter, not a short night#
The most precise dose-response data on this comes from a small crossover study that ran three conditions in the same nine men: a 7-hour night, a 4.5-hour night, and total sleep deprivation. Ghrelin was 22 ± 10% higher after the all-nighter than after 7 hours (0.85 ± 0.06 vs 0.72 ± 0.04 ng/mL, P = 0.048). After 4.5 hours of sleep, ghrelin landed in between — 0.77 ± 0.04 ng/mL — and did not separate from the rested night2. Leptin didn't move in any condition.
Read the middle column again, because it is the one that matters for you. The condition that resembles a bad week — four and a half hours, twice the sleep of the deprivation arm — produced a change the study could not distinguish from noise. The clean, quotable result required staying awake all night. Most people quoting the hunger-hormone story have never pulled an all-nighter in the week they're describing; they slept five or six hours and felt hungry, which is a real experience that this particular measurement doesn't explain.
The hormone result got stronger as the protocol got more extreme. That is the shape of a dose-response curve, and it means the version of sleep loss you actually live sits at the flat end of it.
The same lab, a milder dose, and nothing moved#
A year later, largely the same research group ran 15 healthy men through two consecutive nights of 4-hour sleep. Physical activity dropped measurably during the free-living daytime after the first restricted night (P = 0.008). But total energy intake, hunger, appetite, ghrelin and leptin were all unaffected3.
This is not a rival lab failing to replicate a competitor. It is overlapping authors, a similar population, and a null. Two things separate it from the 2008 result, and both are named in the papers themselves: the sleep dose was partial rather than total, and the eating was measured under time-constrained laboratory conditions rather than free-living ones. The behavioral finding — people moved less — survived; the hormonal finding did not.
That pattern repeats across the field. A 2026 narrative review cataloguing this literature concluded the evidence is "compelling but heterogeneous, with outcomes varying according to the specific intervention," and listed trials on both sides: increases in several men-only deprivation protocols, no significant change in a 4-night restriction study in 14 women and in the 15-man restriction study above6.
| Study | n | Sleep dose | Ghrelin result |
|---|---|---|---|
| Schmid 2008 | 9 men | Total deprivation | +22% vs 7 h |
| Schmid 2008 | 9 men | 4.5 h × 1 night | Intermediate, not significant |
| Schmid 2009 | 15 men | 4 h × 2 nights | No change |
| Broussard 2016 | 19 men | Restricted, crossover | Increased (P < 0.01) |
| Lin 2020 (pooled) | 2,250 | Mixed | SMD 0.14 (0.03–0.25) |
Most of the ghrelin ever measured was the inactive kind#
Here is the technical problem underneath the disagreement, and it is not a footnote. Ghrelin circulates in two forms. Only acylated ghrelin — the version carrying an octanoyl group — binds the receptor that drives appetite. Unacylated ghrelin is the bulk of what's in your blood and does not do that job.
Nearly all of this literature measured total ghrelin, which cannot distinguish the two and therefore mostly tracks the inactive form. As the 2026 review puts it, total-ghrelin assays "do not allow differentiation between the acylated and unacylated forms, potentially obscuring their distinct biological effects," and measuring the active form properly "requires strict pre-analytical control, including the addition of esterase and protease inhibitors to prevent rapid de-acylation"6. Blood that sits untreated quietly converts one into the other.
Worse for this specific question: the ratio of acylated to total ghrelin is not constant across the day. In a 24-hour profiling study, that ratio was lower during sleep than during wake5. So the exposure being studied — how much you slept — moves the very thing that determines how well total ghrelin stands in for active ghrelin. A confound that varies with the intervention is the hardest kind to reason around, and it plausibly contributes to why two competent labs get two answers.
Almost every positive trial was men only#
Scan the sample columns above: nine men, fifteen men, nineteen men. That is not an accident of this article's selection. Reviewing the field, one researcher noted that "studies enrolling exclusively males... report significant effects of sleep restriction on ghrelin concentrations whereas those including women do not," and proposed two separate mechanisms — appetite-side in men via ghrelin, satiety-side in women via reduced GLP-17. Increased food intake after short sleep showed up in both sexes; the hormone carrying the signal may not be the same one.
If that holds, then "sleep loss raises ghrelin" is a claim with a population attached to it, and half the people reading it are outside that population. It also means a woman whose ghrelin doesn't budge after a bad night has not disproved anything about her own extra eating — the eating is the robust finding here, and it is discussed in what a short night does to next-day calories.
What survives the whole audit#
Strip out the overstated effect sizes and something useful remains. In a randomized crossover of 19 men with 24-hour hormone profiling followed by an ad-libitum buffet, ghrelin rose under sleep restriction, and — this is the part worth keeping — evening ghrelin predicted what people then ate. Higher evening ghrelin correlated with more calories from sweet foods (r = 0.48, P = 0.04), and snack intake rose by 328 ± 140 kcal, mostly carbohydrate4.
A within-person correlation between the hormone and the behavior is a stronger claim than a group-mean shift, and it is the version of the ghrelin story that has held up. The mean rise is small; the linkage between an individual's rise and that individual's snacking is not trivial. Note also what the correlation attaches to — sweet, carbohydrate-dense snacks, not food in general, which is the same skew described in why sleep loss changes what you crave.
And the practical consequence is unglamorous. You cannot measure your ghrelin, you cannot supplement it downward, and if you could, ghrelin's actual job description suggests blocking it would do less than the marketing implies. What you can do is decide the composition of the evening in advance, because the extra intake reliably arrives as carbohydrate snacks rather than as a larger dinner — and protein's effect on satiety is one of the few levers that operates on exactly that gap.
FAQ#
Does a 5-hour night raise ghrelin, or does it take a full all-nighter?#
The dose matters more than the coverage suggests. In the study with three conditions in the same men, total sleep deprivation raised ghrelin 22%, while 4.5 hours of sleep produced an intermediate level that did not separate statistically from a 7-hour night2. Mild restriction sits at the shallow end of the curve.
Why do sleep-and-ghrelin studies disagree so much?#
Four differences do most of the work: total deprivation versus partial restriction, free-living versus time-constrained feeding, whether total or acylated ghrelin was assayed, and whether the sample included women. When trials differ on all four, opposite results are the expected outcome rather than a scandal.
Do women's ghrelin levels respond to sleep loss the same way men's do?#
Nobody has the data to say, because the trials that found effects were overwhelmingly male. One review of the field observed that studies including women generally did not find the ghrelin effect, and suggested reduced GLP-1 signaling as a separate satiety-side mechanism in women7. The increased eating appears in both sexes regardless.
Sources#
- Lin J, Jiang Y, Wang G, et al. Associations of short sleep duration with appetite-regulating hormones and adipokines: A systematic review and meta-analysis. Obesity Reviews. 2020;21(11):e13051.
- Schmid SM, Hallschmid M, Jauch-Chara K, Born J, Schultes B. A single night of sleep deprivation increases ghrelin levels and feelings of hunger in normal-weight healthy men. Journal of Sleep Research. 2008;17(3):331-4.
- Schmid SM, Hallschmid M, Jauch-Chara K, et al. Short-term sleep loss decreases physical activity under free-living conditions but does not increase food intake under time-deprived laboratory conditions in healthy men. American Journal of Clinical Nutrition. 2009;90(6):1476-82.
- Broussard JL, Kilkus JM, Delebecque F, et al. Elevated ghrelin predicts food intake during experimental sleep restriction. Obesity. 2016;24(1):132-8.
- Spiegel K, Tasali E, Leproult R, Scherberg N, Van Cauter E. Twenty-four-hour profiles of acylated and total ghrelin: relationship with glucose levels and impact of time of day and sleep. Journal of Clinical Endocrinology and Metabolism. 2011;96(2):486-93.
- Ditmer M, et al. Ghrelin's role in sleep and sleep deprivation: a narrative review. Frontiers in Psychiatry. 2026;17:1744781.
- St-Onge MP. Impact of sleep duration on food intake regulation: Different mechanisms by sex? Obesity. 2016;24(1):11.
