If HIV prevention works, why are rates of high-risk sexual behavior increasing among MSM?

Systematic reviews of HIV prevention research provide clear evidence that behavioral interventions can influence the sexual behavior of men who have sex with men (MSM). However, if HIV prevention works, why are rates of high-risk sexual behavior increasing among MSM in major European, Australian, Canadian, and U.S. cities? The evidence generated by systematic reviews alone may not provide a clear answer to this question. This is because (a) it is uncertain whether experimental interventions shown to be effective in one setting, place, or moment in time can be repeated successfully in another; (b) we have limited understanding of the processes that underlie the interventions; (c) interventions shown to work in an experimental study may not necessarily be effective in everyday life. To answer the question, we need to be alert to the changing risk environment in which men have sex with other men. We also need to develop a new program of research addressing the transferability, sustainability, and effectiveness of sexual health promotion among MSM. Randomized controlled trials will remain one of the optimal means of evaluating behavioral interventions in such a program. By further strengthening the evidence base, we may identify opportunities for innovative as well as effective HIV prevention initiatives.