Currently, there is no recommended confirmatory procedure for detecting doping with the anabolic steroid 5 alpha-dihydrotestosterone (DHT) in sportsmen. To develop a method, we determined ratios of hormone concentrations in urine from 120 healthy men and used these to set discrimination limits. These limits were then applied to results for urine specimens from 10 volunteers given DHT percutaneously (125 mg twice daily for 4 days). The ratio of DHT to epitestosterone (EpiT) was chosen as the primary marker of DHT administration, and ratios of 5 alpha-androstane-3 alpha, 17 beta-diol (5 alpha-ADIOL) to EpiT, 5 alpha-ADIOL to luteinizing hormone, and 5 alpha-ADIOL to 5 beta-androstane-3 alpha, 17 beta-diol were proposed as secondary markers. To evaluate method suitability, we analyzed 194 samples from sports competitors; results for 193 samples were negative, but the ratios in 1 sample greatly exceeded all the chosen limits. In conclusion, we propose that a test scheme based on our recommendations be considered for implementation in all Olympic drug-testing laboratories.