Headlines promise dramatic hormone crashes after a bad night. The best available evidence is more nuanced, and more interesting.
Key takeaways
- Sleep loss does shift hormones, but not uniformly. Pooled analyses of acute sleep deprivation find no clear overall rise in cortisol, and mixed results for appetite hormones such as leptin and ghrelin.
- The clearest signal is for testosterone in young men after sustained restriction. One small, classic study found daytime testosterone roughly 10-15% lower after about a week of 5-hour nights, which is a modest, reversible-looking change in a very small sample.
- Single readings mislead. Timing, sex, BMI, assay and sleep protocol all change the picture, which is why trends over time are more informative than one blood draw.
01. Why this question matters
Sleep and hormones talk to each other all night. Cortisol follows a daily rhythm that dips in the evening and rises before waking. Growth hormone surges in deep sleep. Testosterone in men rises during sleep and peaks in the morning. Leptin and ghrelin, the “fullness” and “hunger” signals, are tied to when and how long we sleep.
So it is natural to ask: if you sleep badly, do these signals go wrong? Popular articles often say yes, loudly. The research is less tidy.
02. Cortisol: expected spike, unclear proof
Cortisol is the best-known stress hormone, so it is the usual suspect. A 2024 meta-analysis of 24 studies in Endocrine Journal pooled acute sleep deprivation (a night or less of lost sleep) and did not find a statistically significant overall increase. The pooled standardized effects were small and their confidence intervals included zero. A subgroup using serum measurements did reach significance, but that is a secondary analysis and should be read cautiously.
The authors also flag why the answer is blurry: different assays, mostly male participants, and only three randomized trials among the 24 studies.
03. Appetite hormones: inconsistent by design and by sex
A 2025 meta-analysis in Obesities pooled five to six randomized crossover trials (about 100 participants) and found no significant change in leptin or ghrelin after sleep deprivation. Ghrelin results varied widely between studies. The paper also contains some internal inconsistencies in reported numbers, which is a reason for extra caution.
A 2025 systematic review in Clocks & Sleep looked specifically at whether sex changes the response. Across eight experimental studies (302 participants), leptin, ghrelin and insulin results were inconsistent, and only one study was designed to test sex differences. Sex alone does not explain the variation; BMI, age and energy balance matter too.
In short: sleep loss does change appetite, but a simple “hunger hormone goes up” story is not what the controlled data show.
04. Testosterone: the clearest, smallest signal
A 2011 JAMA research letter by Leproult and Van Cauter studied healthy young men, mean age about 24, who had restricted sleep (5-hour bedtimes) after rested nights. Daytime testosterone was about 10-15% lower during restriction, and self-rated vigor fell. Only 10 men completed the protocol, so this is hypothesis-generating rather than definitive. It does not show that poor sleep causes clinical low testosterone in the wider population.
05. What’s new in regulation
United States. On January 6, 2026, the FDA issued a revised General Wellness: Policy for Low Risk Devices guidance, replacing the 2019 version. As summarized by law firms, non-invasive wearables that estimate physiologic parameters may qualify as general wellness products if they stay strictly wellness-focused. They must not claim clinical equivalence or medical-grade accuracy, reference diseases or diagnostic thresholds, give treatment guidance, or prompt clinical action. Products that cross those lines remain medical devices.
European Union. In late 2025 the European Commission proposed changes (a Digital Omnibus on AI on 19 November 2025 and a proposal to amend MDR/IVDR on 16 December 2025) that would change how the AI Act applies to AI-enabled medical devices. These are proposals; I could not confirm their final status, so check current EU sources before relying on them.
06. What this means for patients
- Treat sleep as a hormone-health habit, not a cure-all. Regular, sufficient sleep is a low-risk way to support normal rhythms.
- Don’t diagnose yourself from one test or one tracker reading. If you have persistent fatigue, low libido, mood change, weight change or irregular periods, talk to a physician. Hormone testing is time-of-day sensitive, and your doctor will interpret results in context.
- Look for patterns, not spikes. Sleep duration, regularity and symptoms over weeks tell you more than a single night.
For clinicians
| Study | Design | N | Key result | Main limitations |
|---|---|---|---|---|
| Chen et al., Endocr J 2024 | SR/MA; 21 crossover + 3 RCTs | 24 studies | Cortisol SMD 0.18 (95% CI -0.11 to 0.45) crossover; 0.26 (-0.22 to 0.73) RCTs; serum subgroup 0.46 (0.11 to 0.81) | I² 76.6% (crossover); mostly male samples (18 male-only studies); assay variation; English only |
| Gresser et al., Obesities 2025 | SR/MA of randomized crossover trials | 5-6 trials, 105 analyzed (abstract states 141) | Ghrelin SMD -0.27 (-1.00 to 0.46), I² 83.8%; leptin 0.10 (-0.22 to 0.42), I² 21.9% | Few, small trials; heterogeneous protocols; internal inconsistencies in reported N and I² |
| Alfikany et al., Clocks Sleep 2025 | Narrative SR, no pooling | 8 studies, 302 participants | Inconsistent leptin, ghrelin, insulin findings by sex | Only one study powered for sex differences; heterogeneous protocols; energy intake not consistently controlled |
| Leproult & Van Cauter, JAMA 2011 | Small experimental crossover (research letter) | 10 completers (of 28 consented), men ~24 y | Daytime T 16.5 vs 18.4 nmol/L (P=.049), about 10% lower; 2-10 PM about 13% lower | Convenience sample; short protocol; no clinical endpoints; applicability to older men or women unknown |
Practical points: acute sleep deprivation has not been shown to produce consistent cortisol elevation in pooled data; diurnal sampling time, assay type and sex should be reported; sleep history is a reasonable covariate when interpreting borderline morning testosterone in younger men, though the evidence here is thin. Recency note: the three 2024-2025 reviews are 12-28 months old; I did not find a newer systematic review in this run’s searches.
Disclaimer
This article is educational content, not medical advice. It does not diagnose or treat any condition. Consult a physician about your own health.
6Letters is a Korean digital health company working on non-invasive, wearable hormone monitoring. Learn more at 6letters.co.kr.
References
- Chen Y, Xu W, Chen Y, et al. The effect of acute sleep deprivation on cortisol level: a systematic review and meta-analysis. Endocrine Journal. 2024;71(8):753-765. doi:10.1507/endocrj.EJ23-0714
- Gresser D, McLimans K, Lee S, Morgan-Bathke M. The Impact of Sleep Deprivation on Hunger-Related Hormones: A Meta-Analysis and Systematic Review. Obesities. 2025;5(2):48. doi:10.3390/obesities5020048
- Alfikany M, Sakhr K, Kremers S, et al. The Role of Sex in the Impact of Sleep Restriction on Appetite- and Weight-Regulating Hormones in Healthy Adults: A Systematic Review of Human Studies. Clocks & Sleep. 2025;7(3):39. doi:10.3390/clockssleep7030039 (PMID 40843663)
- Leproult R, Van Cauter E. Effect of 1 week of sleep restriction on testosterone levels in young healthy men. JAMA. 2011;305(21):2173-2174. https://stacks.cdc.gov/view/cdc/212252
- U.S. FDA. General Wellness: Policy for Low Risk Devices (revised guidance, January 6, 2026). https://www.fda.gov/media/90652/download
- Covington & Burling. FDA Issues Revised Guidance on General Wellness Products (January 2026). https://www.cov.com/en/news-and-insights/insights/2026/01/fda-issues-revised-guidance-on-general-wellness-products
- Osborne Clarke. Simplification of the AI Act for medical devices (January 12, 2026). https://osborneclarke.com/insights/simplification-ai-medical-devices