The authors studied the association between female and male alcohol intakes at the time of conception and the risk of spontaneous abortion, including early pregnancy loss detected by urinary human chorionic gonadotropin. After a nationwide mailing to about 50,000 members of four trade unions in Denmark in 1992-1994, 430 couples without previous pregnancy attempts were enrolled when birth control was discontinued, and they were followed until a clinically recognized pregnancy or for six menstrual cycles. Alcohol intake and potential confounding factors were reported in monthly questionnaires. Women collected morning urine for 10 days from the first day of vaginal bleeding in each cycle. The authors detected 186 pregnancies: 131 resulted in childbirth, and 55 resulted in spontaneous abortion (34 detected by urinary human chorionic gonadotropin). Depending on the intake in the cycle of conception and the adjustment factors, female alcohol intake was associated with 2-3 times the adjusted risk of spontaneous abortion compared with no intake, and male alcohol intake was associated with 2-5 times the adjusted risk. Only the adjusted relative risks for 10 or more drinks/week compared with no intake were statistically significant. Both male and female alcohol intakes during the week of conception increased the risk of early pregnancy loss.