Sarkola · Alcohol and alcoholism (Oxford, Oxfordshire) 2000 · randomized placebo-controlled trial · n=87

Acute effect of alcohol on androgens in premenopausal women.

Cited 72 times in the scientific literature.

Level 2 - randomized trial

Individual randomized controlled trial

PubMed 10684783 · doi:10.1093/alcalc/35.1.84 · record verified 2026-08-29

What was done

Eighty-seven premenopausal women in the mid-cycle phase of the menstrual cycle (47 using oral contraceptives [OC+] and 40 non-users [OC-]) were evaluated after consuming 0.5 g/kg alcohol versus placebo. Total testosterone, free testosterone fraction, androstenedione, dehydroepiandrosterone (DHEA), and dihydrotestosterone (DHT) were measured at 45 and 90 minutes post-drinking. A substudy of 10 OC+ subjects tested three alcohol doses (0.34, 0.68, and 1.02 g/kg) across 45, 90, and 150 minutes.

What was found

The abstract reports statistical directions without providing exact numerical concentrations or effect sizes. Total testosterone and free testosterone fractions were significantly higher after alcohol compared to placebo at 45 and 90 minutes in both OC- and OC+ subjects, with a more prominent testosterone effect in OC+ subjects. Androstenedione levels were significantly lowered and the testosterone:androstenedione ratio was significantly elevated by alcohol in both groups. No alcohol effects were observed on DHEA or DHT. In the 10-subject substudy, no significant dose or time effects on total testosterone were observed.

Why it matters

This study shows that acute alcohol intake alters androgen metabolism by increasing the conversion of androstenedione to testosterone in premenopausal women, an effect enhanced by oral contraceptive use. This mechanism may contribute to hyperandrogenism associated with heavy alcohol intake.

Limits

The abstract provides no exact numerical values, effect sizes, standard deviations, or exact p-values. Testing was restricted to the mid-cycle phase of premenopausal women, limiting generalizability to other menstrual phases. The dose-response substudy was small with only 10 participants.

Cited by