Jensen · Human reproduction (Oxford, England) 2014 · Cross-sectional study · n=8344

Alcohol and male reproductive health: a cross-sectional study of 8344 healthy men from Europe and the USA.

Cited 141 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional observational study

PubMed 24893607 · doi:10.1093/humrep/deu118 · record verified 2026-08-31

What was done

A coordinated international cross-sectional study evaluated 8,344 healthy men, comprising 1,872 fertile men (aged 18–45 years with pregnant partners from Europe and the USA) and 6,472 young men (aged 18–28 years from the general population across six European countries). Participants completed questionnaires detailing health, lifestyle, and alcohol consumption (beer, wine, liquor) during the prior week (median intake: 8 units/week). Standardized semen analysis evaluated volume, sperm concentration, motility, and morphology. Serum reproductive hormones (FSH, LH, total testosterone, SHBG, inhibin B, and free testosterone) were measured.

What was found

No consistent association was found between alcohol consumption (total or by alcohol type) and any semen quality parameter. However, a significant linear association was identified between total alcohol consumption and total or free testosterone in both cohorts. Compared to men consuming 1–10 units per week, those consuming >20 units per week had higher free testosterone levels: +24.6 pmol/L (95% CI: 16.3 to 32.9) among young men and +19.7 pmol/L (95% CI: 7.1 to 32.2) among fertile men. Alcohol intake was not significantly associated with serum inhibin B, FSH, or LH.

Why it matters

In the largest study to date on healthy men, low-to-moderate alcohol consumption was not associated with impaired semen quality, providing reassurance regarding moderate intake while highlighting an association with elevated testosterone that may reflect altered hepatic clearance.

Limits

The cross-sectional design cannot establish causality. The participation rate was low (20–30%), introducing potential selection bias. Alcohol intake was self-reported and assessed only for the previous week as a proxy for longer-term exposure, risking misclassification. Unmeasured confounders (diet, exercise, stress, occupation, risk-taking behavior) were not controlled for. Effects of heavy long-term drinking or binge drinking were not addressed.

Cited by