Does Sleep Restriction Lower Testosterone in Young Men?
Yes. In this small laboratory study of 10 healthy young men, daytime testosterone levels were 10% to 15% lower after one week of sleep restricted to 5-hour bedtimes than in a rested condition of 10-hour bedtimes. The comparison was not an 8-hour week against a 5-hour week: the men spent a week at home on 8-hour bedtimes as a run-in, then moved into the laboratory for 11 days, 3 nights with 10-hour bedtimes followed by 8 nights with 5-hour bedtimes. For context, the authors note that normal aging is associated with a testosterone decrease of 1% to 2% per year.
Dr. Kumar’s Take
This is a small study, 10 young men, and I read it that way. But the design is tight: each man served as his own control, blood was drawn around the clock, and the drop in daytime testosterone reached statistical significance despite the sample size. What makes it interesting to me clinically is that cortisol did not move. Stress hormone activation is the usual explanation offered when a hormone falls during sleep loss, and here that explanation does not hold. The men also reported steadily lower vigor across the restricted week. Testosterone matters for muscle mass and strength, adiposity, bone density, and a general sense of vigor and well-being, and the symptoms of androgen deficiency, low energy, reduced libido, poor concentration, and sleepiness, overlap almost completely with what sleep deprivation does on its own. For any man trying to build muscle, recover from training, or simply feel like himself, I treat sleep as part of the hormonal picture, not something separate from it.
Key Findings
Ten healthy men, mean (SD) age 24.3 (4.3) years and mean (SD) body mass index 23.5 (2.4), completed the protocol. Across waking hours common to both conditions, 8 AM to 10 PM, testosterone was 16.5 (2.8) nmol/L after sleep restriction versus 18.4 (3.8) nmol/L in the rested condition (P = .049). The effect was especially apparent between 2 PM and 10 PM: 15.5 (3.1) nmol/L versus 17.9 (4.0) nmol/L (P = .02). The authors summarized the change as a 10% to 15% decrease in daytime testosterone.
Daytime cortisol profiles were similar under both conditions, so the testosterone decline was not accompanied by a rise in this stress-responsive hormone. Vigor scores fell progressively across the restricted week, from a mean (SD) of 28 (5) after the first night to 19 (7) after the seventh night (P = .002).
Sleep architecture changed substantially. Total sleep time fell from 8 hours 55 minutes (35 min) to 4 hours 48 minutes (6 min) (P = .002). During each restricted night the men lost 2 hours 45 minutes (29 min) of stage-2 sleep and 1 hour 3 minutes (18 min) of REM sleep, while gaining 9 minutes (8 min) of stage 3 + 4 sleep.
Brief Summary
This laboratory study examined the effect of one week of sleep restriction on testosterone in young healthy men. After a week of 8-hour bedtimes at home, from 11 PM to 7 AM, participants spent 11 days in the laboratory: 3 nights of 10-hour bedtimes, from 10 PM to 8 AM, as the rested condition, then 8 nights of 5-hour bedtimes, from 12:30 AM to 5:30 AM, as sleep restriction. Blood was sampled every 15 to 30 minutes for 24 hours after the second 10-hour night and again after the seventh 5-hour night, and assayed for total testosterone and cortisol. Sleep was recorded and scored each night. Participants completed visual analog scales for global vigor and global affect at 2-hour intervals each day.
Study Design
This was a within-subjects intervention study run at the University of Chicago between January 2003 and September 2009, with each man serving as his own control. Written informed consent was obtained from 28 people who responded to campus flyers; 10 men passed all screening tests and completed the study. Exclusion criteria included a history of endocrine or psychiatric disorders, irregular bedtimes, and sleep complaints. Testosterone and cortisol were measured by immunochemiluminescent assay. Comparisons between conditions used 2-sided nonparametric Wilcoxon tests at a significance level of .05.
Results You Can Use
Daytime testosterone across 8 AM to 10 PM was lower after the restricted week than in the rested condition, 16.5 (2.8) nmol/L versus 18.4 (3.8) nmol/L (P = .049), a decrease the authors put at 10% to 15%. The gap was widest from 2 PM to 10 PM, 15.5 (3.1) nmol/L versus 17.9 (4.0) nmol/L (P = .02).
The hormonal change tracked with how the men felt. Vigor scores dropped steadily through the restricted week, from 28 (5) after the first night to 19 (7) after the seventh (P = .002). Cortisol, which can inhibit gonadal function when elevated, was unchanged, so the testosterone decline was associated with lower vigor but not with a stress hormone response.
The authors point out that 5 hours of sleep per night is a condition experienced by at least 15% of the US working population, which is what makes a one-week protocol relevant outside the laboratory.
Why This Matters For Health And Performance
Most of the daily testosterone release in men occurs during sleep. Beyond sexual behavior and reproduction, testosterone has important effects on muscle mass and strength, adiposity, bone density, and vigor and well-being. Sleep fragmentation and obstructive sleep apnea have been associated with reduced testosterone levels, and in older men, morning testosterone is partly predicted by total sleep time. The symptoms and signs of androgen deficiency, low energy, reduced libido, poor concentration, and increased sleepiness, are all effects that sleep deprivation can produce in healthy people, which is why the authors argue that sleep duration may deserve a place in the evaluation of androgen deficiency.
How to Apply These Findings in Daily Life
- Treat short sleep as a hormonal variable: One week at 5-hour bedtimes was enough to lower daytime testosterone in healthy men in their twenties
- Give yourself the time in bed, not just the intention: With 5-hour bedtimes, actual sleep averaged 4 hours 48 minutes, so time in bed sets the ceiling
- Watch the afternoon and evening: The testosterone gap was widest between 2 PM and 10 PM
- Track vigor: Falling energy across a week of short sleep tracked the hormonal change here
- Do not blame stress alone: Cortisol was unchanged, so a testosterone decline during short sleep does not require a stress response to explain it
- Address sleep fragmentation: Conditions such as obstructive sleep apnea have been associated with reduced testosterone
Limitations To Keep In Mind
The authors describe this as a small convenience sample: 10 young, healthy men recruited by campus flyer, so the results may not carry to older men or to those with existing health conditions. Men with a history of endocrine or psychiatric disorders, irregular bedtimes, or sleep complaints were excluded. The restricted condition ran for 8 nights, so this speaks to one week of curtailment rather than to years of it. The authors themselves call for additional investigation of the links between sleep and testosterone to determine whether sleep duration should be integrated into the evaluation of androgen deficiency.
Related Studies And Internal Links
- Sleep Debt Disrupts Metabolism and Hormone Function
- Sleep Curtailment Decreases Leptin and Increases Ghrelin
- One Week Sleep Restriction Reduces Insulin Sensitivity
- Prevalence and Geographic Patterns of Self-Reported Short Sleep Duration Among US Adults
- How to Sleep Better: Science Daily Playbook
FAQs
How much sleep did the men actually lose?
Total sleep time fell from 8 hours 55 minutes in the rested condition to 4 hours 48 minutes under restriction. The losses came mostly from stage-2 sleep, 2 hours 45 minutes per night, and REM sleep, 1 hour 3 minutes per night, while stage 3 + 4 sleep increased by 9 minutes.
Was the testosterone drop caused by stress?
Daytime cortisol profiles were similar in both conditions. Cortisol is a stress-responsive hormone that can inhibit gonadal function, and it did not rise here, so the testosterone decline was not explained by a cortisol response.
How does this compare with the effect of aging?
The authors compare the 10% to 15% daytime decrease with normal aging, which is associated with a testosterone decrease of 1% to 2% per year.
How common is sleeping 5 hours a night?
The authors describe it as a condition experienced by at least 15% of the US working population.
Conclusion
One week of sleep restricted to 5-hour bedtimes lowered daytime testosterone by 10% to 15% in 10 healthy young men, with the largest gap between 2 PM and 10 PM. The decline came with progressively lower vigor scores and no rise in cortisol. The authors call for further work on the links between sleep and testosterone, including whether sleep duration belongs in the evaluation of androgen deficiency.
