Chen · Sleep 2016 · prospective cohort study · n=796

Inverse U-shaped Association between Sleep Duration and Semen Quality: Longitudinal Observational Study (MARHCS) in Chongqing, China.

Cited 100 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study with follow-up and independent replication dataset

PubMed 26350472 · doi:10.5665/sleep.5322 · record verified 2026-08-26

What was done

A prospective cohort study (MARHCS) evaluated 796 male college students recruited in 2013 in Chongqing, China, with 656 (82.4%) followed up in 2014. At both time points, semen and peripheral blood samples were collected to measure semen quality and reproductive hormone levels. Sleep duration was assessed using the revised Munich Chronotype Questionnaire, and sleep quality was evaluated using the Pittsburgh Sleep Quality Index (PSQI) in 2014. Results were also examined in a replication dataset of 1,346 males.

What was found

There was an inverse U-shaped association between sleep duration and both semen volume and total sperm number, peaking at 7.0–7.5 h/day (dose-response p = 0.002 and p = 0.001, respectively). Sleeping >9.0 h/day was associated with a 21.5% (95% CI 9.2, 32.2) reduction in semen volume and a 39.4% (95% CI 23.3, 52.1) reduction in total sperm number. Sleeping ≤6.5 h/day was associated with a 4.6% (95% CI -10.5, 22.3) reduction in volume and a 25.7% (95% CI -1.2, 60.1) reduction in total sperm number. Individuals shifting sleep duration toward 7.0–7.5 h/day between 2013 and 2014 showed increases in both parameters. The association was independent of PSQI. No association was found between sleep duration and reproductive hormones.

Why it matters

This study suggests that both short and prolonged sleep duration are linked to poorer semen quality in young men, identifying 7.0–7.5 hours per day as the optimal window.

Limits

Sleep duration was self-reported via questionnaires rather than objectively tracked via polysomnography or actigraphy. The primary cohort was restricted to young male college students in a single city in China, limiting generalizability. Estimates for the short-sleep group (≤6.5 h) had wide confidence intervals crossing zero.

Cited by