Reducing sexual risk behavior among steady heterosexual serodiscordant couples in a testing and counseling program.

Objective: To assess the effect of an HIV counseling and testing program targeting steady heterosexual serodiscordant couples.

Methods: We studied 564 couples who attended a sexually transmitted infections/HIV clinic in Madrid in the period 1989 to 2007 and participated in couples counseling and testing. Sociodemographic, epidemiologic, clinical, and behavioral information of both partners was obtained before testing the nonindex partner. Sexual practices reported in the first (preintervention) and second visit were compared, as well those reported in 4 additional visits.

Results: Among the 399 couples who returned for a second visit (71%), the median number of sexual risk practices in the previous 6 months decreased (26.9-0; P <0.001) and the percentage of couples who had not engaged in sexual risk behavior increased (46.1-66.7; P <0.001). This reduction was maintained by the 143 couples who had 4 return visits. The diagnosis of HIV-infection in the index case previous to entering the program was associated with a lower frequency of sexual risk behavior. Independent predictors of postintervention risky sexual behavior included preintervention sexual risk behavior (odds ratio [OR]: 2.8, 95% confidence interval: 1.7-4.4), index case aged over 35 (OR: 2.0, 1.2-3.3), and a recent pregnancy (OR: 3.1, 1.6-6.3). The incidence of HIV seroconversion was 3.9 per 1000 couple-years (1.4-9.7).

Conclusion: The diagnosis of HIV-infection and counseling appears to provide complementary reductions in sexual risk behaviors among serodiscordant steady heterosexual couples at follow-up, but the risk of transmission was not totally eliminated.