Sexual Networks of Pregnant Women With and Without HIV Infection

Heterosexual transmission accounts for a growing proportion of new cases of HIV infection in the Western hemisphere, but the epidemiology of such transmission in Latin America remains poorly defined. Among pregnant women in Lima, Peru, where injection drug use is uncommon, nearly all HIV infections among women are thought to be heterosexually acquired. The authors hypothesized that the most important risk factors for HIV infection among women in Lima would be the behaviors of their partners, which might include unprotected sex with female sex workers (FSW) or with other men.

A 1996-1997 cross-sectional study examined 12,436 pregnant women at the Instituto Materno-Perinatal, the largest maternity hospital in Lima. For more detailed information on risk behaviors and sexual networks, the authors re-interviewed these HIV-seropositive women, HIV-seropositive women identified during 10 additional months of screening, and two groups of seronegative controls; the authors interviewed the most recent male partner of each woman, when available, about his own risk behaviors and sexual contacts.

All women seeking prenatal care, delivery, or care for miscarriage at the Instituto Materno-Perinatal between February 1996 and August 1997 were offered testing for HIV infection and syphilis. Peru's National AIDS Program offered free zidovudine to HIV-seropositive women to decrease vertical transmission. Throughout the case-control study, the authors sought to promptly assess each woman and her most recent male partner with one reactive HIV ELISA test; 90 of these women were subsequently HIV-seropositive by Western blot. Of these, 75 were available for interview, all of whom agreed to participate. Only 12 of these 75 reported knowing they had HIV infection before the study. The authors interviewed 25 during a prenatal visit and 50 postpartum.

In April-May 1997, the authors selected a random sample of 79 ELISA-negative postpartum women by bed assignment to serve as the primary control group (control group 1). The authors created another control group (control group 2) to control for potential reporting and interviewer bias; this group consisted of 58 women interviewed after an initial reactive ELISA but who subsequently had a negative follow-up ELISA, or a negative or indeterminate Western blot. With permission of the female case or control, the authors invited her most recent male partner to give written informed consent, undergo interview and free testing for HIV infection, and return for lab results and post-test counseling. The authors interviewed the most recent male partner of 41 (55 percent) HIV-seropositive women, 40 (51 percent) control group 1 women, and 30 (52 percent) control group 2 women.

A nurse, psychologist, or social worker (all female) performed a confidential, standardized, face-to-face interview with each woman, and in a separate interview, with her male partner. Questions addressed socio-demographic characteristics, history of STDs, risk behaviors, and perceived risk behaviors of their last three partners. Sexual network questions included number of partners in the past year, 5 years and lifetime; number of contacts with FSW or MSM; and use of condoms with FSW.

The authors concluded that men who have sexual contact with FSW, or who have sexual contact with other men (though self-identifying as heterosexuals) serve as bridges between these groups at high risk for HIV infection and heterosexual women. Interventions aimed at decreasing heterosexual acquisition of HIV infection by women in Peru should not focus primarily upon reducing the number of partners of women; the majority of the HIV-infected women in the study were monogamous. Education of boys and girls should begin before onset of sexual activity, and should address abstinence, monogamy, choice of a partner at low risk for HIV infection or other STD, avoidance of risky practices, and the protective effect of condoms.

Back to other CDC news for April 1, 2003

Previous Updates