In a 2008 online survey of men who have sex with men, those who reported engaging in group sex had a higher sexual risk profile and more psychosocial problems than those who said that they had only one-on-one encounters.1 Furthermore, both sexual risk and psychosocial problems were more frequent among men who had group encounters involving four or more partners than among those who did so with two or three partners. For instance, 12% of those reporting only one-on-one sexual activity in the past 60 days were HIV-positive, compared with 21% of those reporting sex with 2–3 partners and 35% of those reporting encounters with four or more; the proportions who had used three or more recreational drugs in the past 60 days were 22%, 29% and 41%, respectively.
Previous research on men who have sex with men has assessed overall levels of sexual risk among those who have group sex, but has not examined the "intersecting psychosocial conditions" that may contribute to risk or the behaviors associated with different types of group encounters. To explore these issues, investigators recruited users of a gay-oriented social networking website to participate in an online survey that assessed demographic characteristics, sexual behavior in the past 60 days and a range of psychosocial problems. The sample consisted of 7,158 men, of whom 5,038 reported only one-on-one sex, 1,320 reported group sexual encounters with 2–3 partners and 800 reported group encounters with four or more partners. The researchers used chi-square tests and logistic regression to assess differences among groups.
Respondents ranged in age from 18 to 81 (median, 39); the majority were white (82%), had at least a college education (55%) and lived in cities with populations of at least 250,000 (54%). Eighty-five percent characterized themselves as gay, 14% as bisexual and 1% as heterosexual. Some 16% were HIV-positive, and 15% had received an STD diagnosis in the previous 60 days; 27% had had anal sex with more than 100 men. Thirty-five percent of respondents reported one psychosocial problem, 14% reported two and 6% reported three or more. Some 25% said that they had used three or more recreational drugs in the past 60 days, and 12% had ever been in a drug or alcohol treatment program; 16% had symptoms of depression, 17% had ever been incarcerated and 25% had had their first drink at age 14 or younger.
Group differences were found on almost every demographic and psychosocial characteristic. Men reporting recent group sex involving two or three partners were more likely than those reporting only one-on-one sex to be HIV-positive (21% vs. 12%), to have had a recent STD diagnosis (16% vs. 13%), to have had more than 100 male anal sex partners (41% vs. 17%) and to have used three or more recreational drugs in the past 60 days (29% vs. 22%). However, the proportions were still higher among those reporting group sex with four or more partners. Some 35% of these men were HIV-positive, 23% had recently received an STD diagnosis, 72% had had anal sex with more than 100 men and 41% had recently used multiple recreational drugs. In addition, these men were more likely than others to have been incarcerated (24% vs. 16–17%) and to have had their first drink before age 14 (31% vs. 24–26%); they were more likely than men reporting only one-on-one sex to have recently been in a drug or alcohol treatment program (12% vs. 9%).
In the 60 days before taking the survey, all respondents had engaged in one-on-one sexual activity. Men reporting group sex with either 2–3 or more partners were more likely to have met one-on-one partners online (85% of each group) than were those having only one-on-one sex (79%); they also were more likely to have had anal sex with a one-on-one partner (96% vs. 68%). Those reporting group sex with four or more partners, however, were the most likely of any group to report unprotected anal sex (83% vs. 70–74%), −unprotected anal sex with a serodiscordant partner (41% vs. 25–29%) and drug use before sex (34% vs. 14–23%); they were the least likely to report sex with a main partner (32% vs. 36–44%). When behavior and psychosocial problems were assessed with regard to recent group sexual encounters, men who had participated in such activity with four or more partners again exhibited a riskier profile than those who had had encounters involving 2–3 partners.
Results of multinomial and multivariable logistic regression analyses confirmed the bivariate findings. Compared with men who had recently had only one-on-one sexual encounters, those who had had group sex with 2–3 partners were more likely to be HIV-positive (odds ratio, 2.9) and to have had multiple one-on-one encounters (2.5); they were less likely to have had one-on-one sex with a main partner (0.7). Men who had had encounters with four or more partners had elevated odds of being HIV-positive (5.6), having recently received an STD diagnosis (1.5), reporting 1–3 psychosocial problems (1.4–2.4), having had multiple one-on-one sexual encounters (3.8) and having had unprotected sex with a serodiscordant partner (1.5); they had reduced odds of reporting sex with a main partner (0.6). Results of comparisons between the two groups of men reporting group sex similarly showed those with the greater number of partners to be at higher risk.
As the researchers acknowledge, the generalizability of their findings is limited, the survey did not cover many areas of potential importance, and behaviors and attitudes may have changed since 2008, when the data were collected. Still, they conclude that among men who have sex with men, those who have group sex with four or more partners are "particularly high-risk" because they engage in a range of mutually reinforcing unhealthy behaviors. The investigators encourage further work aimed at determining strategies for helping these men to simultaneously avoid risky sexual behavior and address psychosocial problems.
—D. Hollander
REFERENCE
1. Hirshfield S et al., Drug use, sexual risk, and syndemic production among men who have sex with men who engage in group sexual encounters, American Journal of Public Health, 2015, 105(9):1849–1858.