Serum androgen levels in black, Hispanic, and white men.

Context: Racial/ethnic differences in androgen levels could account for differences in prostate cancer risk, body composition, and bone loss.

Objective: The objective of the study was to investigate racial/ethnic variations in testosterone, bioavailable testosterone, dihydrotestosterone (DHT), SHBG, and dehydroepiandrosterone sulfate (DHEAS) levels.

Design: The Boston Area Community Health (BACH) Survey was a multistage stratified cluster random sample, recruiting from 2002 to 2005.

Setting: The study was a community-based sample of Boston.

Participants: Participants included black, Hispanic, or white individuals, aged 30-79 yr, competent to sign informed consent and literate in English/Spanish. Of 2301 men recruited, 1899 provided blood samples (538 black, 651 Hispanic, 710 white).

Intervention: Intervention consisted of data obtained during in-person at-home interview, conducted by a bilingual phlebotomist/interviewer.

Main Outcome Measure(s): Testosterone, bioavailable testosterone, DHT, DHT to testosterone ratio, SHBG, and DHEAS were measured.

Results: With or without adjustment for covariates, there were no significant differences in testosterone, bioavailable testosterone, or SHBG levels by race/ethnicity. DHEAS levels differed by race/ethnicity before covariate adjustment; after adjustment this difference was attenuated. Before adjustment, DHT and DHT to testosterone ratios did not significantly differ by racial/ethnic group. After adjustment, there was evidence of racial/ethnic differences in DHT (P = 0.047) and DHT to testosterone (P = 0.038) levels. Black men had higher DHT levels and DHT to testosterone ratios than white and Hispanic men.

Conclusions: Because there are no racial/ethnic differences in testosterone levels, normative ranges need not be adjusted by race/ethnicity for androgen deficiency diagnosis for men aged 30-79 yr. Further investigation is needed to determine whether differences in DHT levels and DHT to testosterone ratio can help explain racial/ethnic variations in prostate cancer incidence, body composition, and bone mass.