Sexual Partner Referral for HIV Testing Through Social Networking Platforms: Cross-sectional Study

Background Men who have sex with men (MSM) who undergo voluntary HIV counseling and testing (VCT) often report condomless anal sexual intercourse, having many sexual partners, and being exposed to risky sexual networks. Limited research has discussed the application of motivational interviewing and convenience referral platforms to facilitate the referral of sexual partners for HIV testing among MSM. Objective This study aimed to evaluate the effects of VCT referral by sexual partners through social networking platforms and the test results after elicited interviews with MSM; compare the characteristics and risk behaviors among MSM tested without referral, index subjects, and referred sexual partners; and explore unknown sexual affiliations through visualizing and quantifying the social network graph. Methods This was a cross-sectional study. Purposeful sampling was used to recruit index subjects from a community HIV screening station frequented by MSM in Taipei City on Friday and Saturday nights. Respondent-driven sampling was used to recruit sexual partners. Partner-elicited interviews were conducted by trained staff before VCT to motivate MSM to become index subjects and refer sexual partners via the Line app, or to disclose the accounts and profiles of sexual partners on relevant social networking platforms. Referred sexual partners received rapid HIV testing, and the recruitment process was repeated until leads were exhausted. Results After the interviews, 28.2% (75/266) of MSM were successfully persuaded to become index subjects in the first wave, referring 127 sexual partners via the Line app for rapid HIV testing and disclosing 40 sexual partners. The index subjects and tested sexual partners had more sexual partners (F2=3.83, P=.02), more frequent anal intercourse (F2=10.10, P<.001), and higher percentages of those who had not previously received HIV testing (χ21=6.1, P=.047) compared with MSM tested without referrals. The new HIV-seropositivity rate among tested sexual partners was 2.4%, which was higher than the rate in the other 2 groups. The social network analysis revealed the following 4 types of sexual affiliation: chain, Y, star, and complicated. Among the HIV-negative sexual partners, 26.9% (43/160) had sexual affiliations with HIV-positive nodes, and 40% (10/25) were untested sexual partners with a direct sexual affiliation with an HIV-positive node. Four transmission bridges were found in the network graph. Conclusions Partner-elicited interviews can effectively promote referral for HIV testing and case identification via Line, and can clarify unknown sexual affiliations of MSM to facilitate the development of a tailored prevention program. Social network analysis is needed for an insightful understanding of the different network structures.