More than two thirds of our study participants attended an SOPV either to meet a sex partner or to have sex in the last 3 months, suggesting that SOPV attendance may be quite common among MSM in Lima. In addition, those who attended an SOPV were significantly more likely to report behaviors associated with elevated risk of HIV and STI transmission, including group sex, transactional sex, sex under the influence of alcohol, sex with a casual partner, and more overall partners. Notably, the majority of SOPV attendees indicated favorable attitudes toward theoretical SOPV-based sexual health interventions such as condom/lubricant distribution and HIV/STI testing.
Although SOPV attendance was associated with several sexual risk behaviors, one notable exception was condomless anal sex. One possible explanation is that we asked questions about sexual positioning and condom use with reference to only the last two partners, rather than asking about all recent partners (as was asked for other behaviors). However, similar results to ours were observed in a large online survey conducted in the mid-2000s of MSM in the US, which found that meeting a partner at a physical venue (e.g. bars/clubs, bathhouses, and public outdoor spaces) was associated with risk behaviors such as alcohol use – but not with condomless anal sex (24). In our exploratory analyses, sexual risk behavior also appeared to vary by SOPV category. For example, compared with other categories, the proportion of MSM reporting nearly every risk characteristic (except condomless anal sex) was higher for sex clubs (Table 4). In contrast, hourly hotels, which were by far the most common SOPV attended, appeared to be frequented by lower risk clientele. Taken together, these results suggest that more nuanced data collection may be necessary to characterize sexual risk behaviors associated with SOPV attendance in this setting. Such information, including the identification of specific SOPVs attended by the highest risk clientele, would be particularly germane to the development of SOPV-based outreach strategies to deliver HIV testing and other sexual health interventions.
Seventy-eight percent of our sample population reported meeting a sex partner online in the last 3 months, which is roughly double the percentage estimated by other recent surveys of MSM in Lima, including one conducted in 2013–2014 (in person) and another in 2012–2013 (online), which reported 37% and 44%, respectively (3,39). Our higher estimate is likely related to increased access to low-cost internet services and mobile devices over time in Perú; however, our use of an online survey, which selects for individuals with internet access, may have also played a small role. In addition, geosocial networking applications (e.g. Grindr), which have been associated with increased sexual risk behavior and STIs incidence in some settings (42–44), have had substantial growth in popularity since these earlier studies.
Our results also shed light on the relationship between online platform use and SOPV attendance. MSM who had sex at an SOPV, but not those who met a partner at an SOPV, were significantly more likely to report having a recent online partner. This mirrored what we observed for one’s living situation: living with family was associated having sex at an SOPV but not meeting a partner at one. One potential explanation for this pattern is that, although online platforms can facilitate the identification of sex partners, persistent stigma precludes those who live with family from having sex with these partners at home. SOPVs might fill this void by offering places outside of the home where MSM can have sex with online partners. We also found that, when stratified by SOPV category, the association between SOPV attendance and online platform use was significant only for hotels and bars/discos, suggesting that these venues may be preferred rendezvous sites for meeting online partners for sex, perhaps because identifying new sex partners in person may be easier at saunas, sex clubs, and porno theaters.
The primary limitation of this study relates to its generalizability. Our sampling method, which relied on recruitment through local social media networks, selected a population that is somewhat more educated and affluent than the general population of MSM in Lima. Therefore, our estimates of SOPV attendance and sexual behavior may not reflect the experiences of MSM in lower socioeconomic strata. Self-reported HIV prevalence was 26%, which is within the range reported by other studies of MSM in Lima (5,6). However, over 80% of MSM in this study reported having ever received an HIV test and over 90% of those who self-reported being HIV positive indicated they were taking ART. In both cases, these rates are somewhat higher than past population estimates of HIV testing and ART coverage, respectively, in the region (1,12), suggesting above average access to health services. If so, the estimates derived from our sample population may in fact under-represent the risk profile of MSM in Lima on average. Future studies of SOPV attendance and associated sexual risk behaviors in Lima should consider utilizing alternative sampling strategies to include MSM in less affluent communities, as well as TW – an extremely vulnerable population that our online recruitment strategy did not adequately reach.
Although, overall, participants reported favorable attitudes toward SOPV-based HIV/STI testing interventions (Fig. 1), interpretation of these data is limited by the small sample size for sex clubs, porno theaters, and bars/discos. The use of a survey to evaluate acceptability of hypothetical interventions may also overestimate uptake in real-world settings. Nevertheless, our results, which indicate that over two thirds of MSM would either “definitely” (45%) or “probably” (23%) accept SOPV-based HIV testing, are consistent with direct observations from the field, including a recent study that demonstrated 52% uptake of venue-based HIV testing when it was offered to MSM and TW at bars, clubs, and public parks in Lima (5).
Despite the limitations to its generalizability, the online administration of our survey allowed it to be conducted in a completely anonymous fashion. This significantly reduced the likelihood of social desirability bias and is an important strength of the study. While anonymous participation can also, in theory, enable a single person to complete more than one survey, there was no incentive to do so and completion of the survey was relatively time-intensive (~ 10–15 minutes on average). Therefore, duplicate survey responses or other spurious results related to external incentives are unlikely.
In summary, our results support the suggestion that SOPVs play an important role in MSM sexual networks in Perú. This may be particularly pronounced compared to settings where young adults tend to leave the family home at a younger age, or where stigma is a less prominent driver of sexual decision-making. Both SOPV attendance and the use of online platforms to meet sex partners appear to be highly prevalent behaviors among MSM in Lima. Furthermore, our findings indicate that MSM who attend SOPVs may be at particularly high risk for HIV/STI transmission, underscoring an opportunity to work with SOPVs to develop interventions that will expand access to testing, treatment, and prevention services. Although access to PrEP is relatively limited at present time in Perú, as it expands partnerships with SOPVs could be leveraged to promote linkage to existing providers or even facilitate future medication distribution. Importantly, our results suggest SOPV-based interventions would be accepted by most MSM in Lima who frequent these venues. Overall, these findings have important public health implications and suggest that SOPVs in Lima are suitable sites for targeted HIV testing and prevention interventions.