Use of multiple recreational drugs is associated with new HIV infections among men who have sex with men in China: a multicenter cross-sectional survey

Background There is limited information about the types of recreational drugs used by men who have sex with men (MSM) in China or the consequent impact on sexual health and human immunodeficiency virus (HIV) acquisition. Methods We recruited MSM from seven cities in China between 2012 and 2013 using multiple approaches including advertisements on gay websites, collaborating with local MSM community-based organizations, peer referrals, and venues such as gay bars and bathrooms visited by MSM. We divided participants into four subgroups based on the number of recreational drugs (RDs) used in the previous 6 months. We defined use of multiple RDs as use of ≥2 types of RDs. Demographics and HIV-related high-risk behaviors were collected, and blood samples were tested for recent HIV infection by the HIV-1 subtypes B, E, and D immunoglobulin G capture enzyme immunoassay (BED-CEIA). We used multivariable logistic regression adjusted for sociodemographics to determine the adjusted odds ratios (aORs) and associated 95% confidence intervals (CIs) of the subgroups of RD use for recent or established HIV infection. Results A total of 4496 Chinese MSM participated; 28.4% used RDs, and 5% used multiple types of RDs. The prevalence of each RD use was as follows: poppers (25.9%), ecstasy (2.4%), ketamine (1.2%), amphetamine (0.6%), tramadol (0.4%), methamphetamine (3.8%), and codeine (1.9%). Users of multiple RDs commonly used poppers combined with one or more other types of RDs. Multiple RD users were likely to be aged 26–30 years (vs. 18–25 and > 30 years), live in non-local cities (vs. local cities), never married (vs. married), have a high monthly income (vs. no income and 1–599 USD), use versatile positions during anal intercourse (vs. top or bottom), and have inadequate HIV-related prevention knowledge (vs. adequate). As the number of RDs used in the previous 6 months increased, the prevalence of HIV-related high-risk behaviors increased (P < 0.05 for all). The odds of recent HIV infection were higher among those who used one type (aOR = 2.2, 95% CI: 1.5–3.0) or two types of RD (aOR=2.3, 95% CI: 1.0-5.2) in the previous 6 months compared to the odds among those who did not use RDs. Conclusion The level and pattern of multiple RD use among Chinese MSM were different from high-income countries. MSM who used more RDs are more likely to engage in high-risk sexual behaviors, and these behaviors may be associated with increases in new HIV infections.