ABSTRACT
<p><b>OBJECTIVE</b>To understand the syphilis infection and its high risk factors among men who have sex with men (MSM) recruited from different channels.</p><p><b>METHODS</b>Supported by the China-Gates Foundation HIV program from July to December 2011, we cooperated with community based organizations to conduct syphilis testing intervention among MSM from 14 cities (Beijing, Shanghai, Tianjin, Chongqing, Harbin, Shenyang, Qingdao, Xi'an, Nanjing, Wuhan, Hangzhou, Changsha, Kunming, and Guangzhou) and one province (Hainan province). Participants were recruited from different channels by the staff of local CBOs, Demographic (e.g. age, marital status, and education) and behavioral (e.g. condom use and sexual partners) data were collected using anonymous questionnaires. Blood samples were also collected to test for syphilis. A total of 34 100 MSM participated in the survey. Participants were excluded from data analysis (1 399, 4.1%) if they did not receive syphilis tests or they completed less than 80.00% of the key questions in the survey. Chi-square tests were used to understand the socio-demographic and behavioral differences between each group. Results of syphilis tests were also compared. Logistic regression models were used to test the statistical significance of these differences.</p><p><b>RESULTS</b>A total of 32 701 MSM were enrolled and received syphilis testing. The average age of participates was 30.96 ± 9.57. And among them, 2 284 cases (7.0%) were recruited from gay bathhouses, 4 774 (14.6%) from gay bars, 6 266 (19.2%) from the internet, 1 997 (6.1%) from the parks/toilets and 17 380 (53.1%) from other channels. MSM recruited from the bathhouses had the highest syphilis infection rate than other 4 groups: gaybars (4.5%, 216/4 774), internet (6.7%, 422/6 266), parks/toilets (8.3%, 166/1 997), other channels (6.4%, 1 103/17 380) (χ² = 164.58, P < 0.001). The multivariate logistic regression analysis showed that being > 20 years of age (P < 0.001), having more than 2 homosexual partners in recent 3 months (8.0% (1 408/17 714), OR (95% CI) = 1.44 (1.04-1.98)), having no sex with females in past 3 months (6.8% (1 446/21 276), OR (95% CI) = 1.25 (1.07-1.46)), and not using condom at last anal sex (8.0% (769/9 668), OR (95% CI) = 1.13 (1.03-1.25)) were associated with a higher probability of being infected with syphilis. Whereas MSM married (7.2% (456/6 305), OR (95% CI) = 0.84 (0.73-0.98)), having a college or a higher education (5.3% (829/15 684), OR (95% CI) = 0.60 (0.53-0.67)), being local residents (6.5% (1 843/28 185), OR (95% CI) = 0.73 (0.61-0.87)) and living in the local province of project cities (6.6% (170/2 593), OR (95% CI) = 0.67 (0.53-0.85)) were protective factors.</p><p><b>CONCLUSION</b>MSM who were recruited from gay bathhouses have a higher rate of syphilis infection than those in channels relatively. They are older, with low education levels and high-risk sexual behaviors. Tailored interventions are required in the future, especially for MSM from gay bathhouses.</p>