Incidence and risk factors for heterosexually acquired HIV in an inner-city cohort of women: temporal association with pregnancy.

Background: A growing proportion of AIDS cases in the United States are due to heterosexual transmission of HIV, particularly in women. The risk of heterosexually acquired HIV was prospectively studied in a cohort of inner-city women with no history of parenteral drug use.

Methods: Study participants were evaluated at 6-month intervals for the presence of HIV antibody, sexually transmitted diseases, self-reported sexual behavior, and drug use by self-report and urine screening.

Results: Of 449 initially HIV-negative women who were seen at least once in follow-up, 4 seroconverted to HIV, with a cumulative incidence of 2.4% at 30 months. Risk factors for HIV seroconversion included nonparenteral drug use (p < .02) and anal intercourse (p < .01). Sexually transmitted diseases were not associated with HIV, although the power to detect such an association was limited. In addition, 3 of 4 seroconverters became pregnant, yielding a rate of 55.5 pregnancies/100 person-years of follow-up compared with a rate of 11.1 pregnancies/100 person-years of follow-up in nonseroconverters (p < .03).

Conclusion: The incident rate of heterosexually acquired HIV in this inner-city U.S. cohort of women who were not using parenteral drugs is comparable with that reported in some developing countries where heterosexually acquired HIV is endemic. Most seroconversions appeared related to risk behavior seen in association with nonparenteral drug use. The previously unreported association of incident HIV infection with pregnancy in this cohort may be related to either behavioral or biologic factors.