Randomized, controlled trials for HIV/AIDS prevention in Africa: learning from unexpected results

Evaluation of: Wawer MJ, Makumbi F, Kigozi G et al.: Circumcision in HIV-infected men and its effect on HIV transmission to female partners in Rakai, Uganda: a randomized controlled trial. Lancet 374, 229-237 (2009). A randomized, controlled trial (RCT) testing the hypothesis that circumcision in HIV-infected men would reduce HIV transmission to uninfected female sexual partners failed to confirm such findings from observational studies in Africa. Researchers noted no statistically significant differences in incident HIV to partners of circumcised or uncircumcised HIV-infected men. Surprisingly, by the time the trial was terminated because of futility, the trend was toward higher incidence in partners of circumcised men. The trial may have been stopped too early to adequately assess this unexpected trend. Since the investigators focused exclusively on sexual transmission, it is possible, if not likely, that unmeasured nonsexual (unsanitary puncturing) exposures caused some incident HIV infections, thus contaminating their protocol. The history of RCTs for AIDS prevent ion in Africa has been highlighted by unexpected results. RCTs have almost exclusively focused on sexual transmission, a framework that may be a leading culprit behind unexpected and disappointing outcomes.