Reducing HIV risk behavior of men who have sex with men through persuasive computing: Results of the Men's INTernet Study-II

Men who have sex with men (MSM) remain the largest population infected with HIV in the United States [1]. Although early (offline) prevention efforts were effective in reducing the spread of HIV [2], there has been a reversal in safer sex practices since the mid-1990s [3], leading researchers to conclude that HIV prevention efforts for MSM in the US have “faltered” [4]. Re-energizing HIV prevention for MSM remains an urgent priority in the fight against HIV/AIDS [5]. Increased sexual risk taking and rising HIV rates among MSM have coincided with the broad adoption of the Internet as a way for MSM to meet sexual partners [6]. The Internet is now the largest venue where MSM meet sexual partners [7]. A meta-analysis of 14 studies from 1999 to 2005 reported a weighted mean estimate of 40% of MSM meeting their sex partners online [8]. However, the analysis is now dated, the range of estimates was large (23% to 99%) and highly dependent upon the recruitment methods used [8]. Clinic studies have identified Men who use the Internet to seek sex with men (MISM) as at higher risk for HIV/STIs than other MSM [9]. Internet sex-seeking appears to increase risk through an increased number of partners [10] and therefore increased probability of sexual risk behavior [6, 11]. Tracing STI outbreaks [12] and HIV transmission [13] through Internet-mediated liaisons is well-documented. Given the global reach of the World Wide Web, its accessibility, affordability, anonymity, and popularity for sex-seeking among MSM [14], Internet-based interventions hold exceptional promise to address the global pandemic of HIV among MSM if they can be shown to reduce unsafe sexual behavior. The long-term objective of this research is to develop Internet-based interventions strong enough to lower sexual risk behavior among MISM. To date, three online interventions for adult MISM have been rigorously evaluated. These programs used tailored messaging to MISM entering relationships [15], visual stories to promote HIV testing and to reduce unsafe sex [16], and “gay” avatars on a virtual cruise [17]. However, two of the trials [15, 17] experienced attrition rates of 70–80%, preventing meaningful interpretation of results, while the third [16] did not attempt to measure behavior change. Thus, attrition in online interventions appears a major challenge.