Background Antiretroviral therapy (ART) reduces viral load and has been suggested as a prevention measure to reduce transmission. However there are concerns that this may be counterbalanced to some extent by increased sexual activity associated with improved health and decreased condom use. The impact of ART on rates of HIV-transmission was investigated within HIV discordant couples in rural, Northern Malawi.

Methods The study was conducted as part of the Karonga Prevention Study. A demographic surveillance system was started in 2001 and annual home-based HIV sero-surveys added from 2007, with individual counselling and consent. From the first round of the sero-survey, 201 HIV discordant couples (147 male positive, female negative; 54 male negative, female positive) were identified for whom follow-up HIV data were available in the negative partner. Free ART is available at local health facilities and ART use was determined by linkage of data to a clinic based cohort and self-report. Patients attend ART clinics with a guardian who may be the spouse, and are counselled about the need for safer sex. For this analysis participant exposure to ART was defined as having started ART prior to study entry, and follow-up was censored at time of last observation, HIV seroconversion, last negative HIV test, death, departure or dissolution of marriage. Follow-up with re-testing is ongoing.

Results So far there are 249 and 99 person-years of follow-up in HIV negative women and men, respectively. 55 of the 201 positive partners had commenced ART at entry and within these couples no HIV transmission occurred over a 3-year follow-up period. Among 146 discordant couples without ART the rate of transmission to females was 5.2 / 100 person years (95% CI 2.7 to 10.5) and to males 1.3/100 person years (95% CI 0.18 to 9.4).

Conclusions In HIV discordant couples ART is associated with reduced rates of HIV transmission within partnerships.