A heterosexual partner study was carried out in Rio de Janeiro, Brazil, from August 1990 to December 1991. The main objectives were to determine the rate of male-to-female transmission of human immunodeficiency virus (HIV) and to determine risk factors. Male index cases were recruited according to the following criteria: 1) confirmed HIV positivity, 2) 18 years old or older; 3) heterosexual contact within the past year. Only female partners who reported not to have other risk factors but to have had sexual contact with the index case were invited to participate. Couples were interviewed for risk factors and had blood collected for laboratory studies. The overall prevalence of HIV infection was 45 percent among 204 female partners in the study. Using logistic regression, the authors found the following factors to be independently (p < 0.05) associated with HIV infection: 1) anal sex (odds ratio (OR) = 3.74, 95% confidence interval (CI) 1.87-7.45); 2) condom use during vaginal sex sometimes (OR = 1.45, 95% CI 1.07-1.96), rarely (OR = 2.10, 95% CI 1.15-3.83), or never (OR = 3.04, 95% CI 1.23-7.50) as compared with always; 3) frequency of sexual contacts in the year prior to interview (100+) (OR = 2.00, 95% CI 1.03-3.91); and 4) oral contraceptive use (OR = 2.04, 95% CI 0.97-4.29). In addition to a borderline significance of oral contraceptive use, there was a strong suggestion of an interaction with history of sexually transmitted diseases.