Impact of improved treatment of sexually transmitted disease on HIV infection.

Grosskurth and colleagues report the outcomes of a randomized controlled trial to evaluate the impact of improved treatment of sexually transmitted diseases (STD) on HIV infection. The authors congratulate the research team on the high quality of the design and execution of their study. They are amazed by the reported 43% reduction in the incidence of HIV as a result of the medium-strength intervention. The effect upon other STDs is however less apparent. The researchers discussed possible effect modifications of different co-factors but concluded that bias is negligible or very limited. These authors do not however think that the researchers sufficiently consider the implications of the difference in baseline HIV prevalence between intervention and comparison communities. It is important to correct for the initial 14% difference in initial condition since the prevalent cases form the source of subsequent HIV infections. The authors therefore recalculated risk ratios using a correction factor for each of the six matched pair communities. In so doing the overall estimate of HIV reduction as a result of the intervention would be 33% probably a better estimate than the original 43%. The confidence interval for the effect was widened by the correction. A more sophisticated analysis should consider the possibility that frequencies changed during the research period since the HIV epidemic has most likely not yet reached its dynamic equilibrium. The authors stress in closing that a 33% reduction in the frequency of HIV is nonetheless remarkable for this type of intervention.