A cross-sectional study was carried out at a programme to prevent mother-to-child transmission of HIV (MTCT) at a public antenatal clinic in Abidjan, Côte d'Ivoire. The objectives were to obtain information from women concerning their reactions to HIV test results received through the programme, their experiences with faithfulness to partners as a means of primary HIV prevention for themselves and their infants, their relationships with partners, their own and their partners' experiences with HIV testing, and their knowledge of their partners' HIV serostatus. The participants were a purposive sample of 87 women who had received HIV-1-positive test results and 30 women who had received HIV-1-negative test results through the clinic's programme. Eighty-five per cent of the HIV-positive women were surprised by their test result; 52% of those who tested HIV-negative anticipated that result. Nearly two-thirds of those who were surprised to be HIV-positive and a similar proportion of those who expected to be HIV-negative explained their reactions by referring to faithfulness to their partners. Only five of the 117 women interviewed expressed a belief that their partners were faithful to them; and only two, and none of those who received an HIV-positive test result, reported using condoms with partners. No more than one-fourth of either the HIV-positive or the HIV-negative groups of women had been previously tested for HIV; less than one-fourth of the women in each group reported having partners who had been tested for HIV, or knew their partners' serostatus. Relationship characteristics of some HIV-positive women may have increased their vulnerability to HIV infection. Although being faithful to partners can be effective for the primary prevention of HIV infection, the manner in which it was practiced by many of the women in our study may have further increased their risk of infection. Organisations that choose to fund HIV prevention programmes that promote faithfulness to partners, and the programmes that stress faithfulness, must ensure that women are informed about the conditions that can influence the effectiveness of faithfulness as a protective action. However, women need more than information. Prevention programmes, whether concerned primarily with prevention of MTCT or with HIV prevention more broadly, must promote and elicit cooperation from women's sexual partners to support women's efforts to be tested for HIV, to be tested for HIV themselves, to disclose their test results, to reciprocate women's faithfulness and, if HIV serodiscordant or unwilling to be faithful, to use condoms. These steps may increase the likelihood that women will be able to protect themselves and their infants from HIV infection by being faithful to their partners.