Temporal trends in HIV-1 incidence and risk behaviours in men who have sex with men in Bangkok, Thailand, 2006–13: an observational study

BACKGROUND:HIV-1 incidence in men who have sex with men (MSM) is often difficult to estimate. We therefore assessed temporal trends in HIV-1 incidence and behavioural risk factors in MSM in Bangkok, Thailand, from 2006 to 2013. METHODS:In this observational study, we used data for clients attending the Silom Community Clinic for voluntary counselling and testing (VCT) services and from the Bangkok MSM Cohort Study (BMCS) to investigate trends in HIV incidence per 100 person-years per quarter in both cohorts. During VCT, basic demographic data were gathered at registration. However, no behavioural risk data were gathered. In the BMCS, we gathered demographic and behavioural data at baseline and at regular study visits using audio computer-assisted self-interviewing. Questions were included about potential risk factors such as drug use, sexual practices, and how often condoms were used. We also analysed behavioural risk factors in the BMCS cohort, using a restricted cubic spline function for time. FINDINGS:From 2006 to 2013, 8176 MSM came for VCT; 1999 (24%) clients were initially seronegative and returned for another test. 235 (12%) individuals seroconverted. The overall HIV-1 incidence was 5.5 per 100 person-years (95% CI 4.8-6.3), with an increasing trend (adjusted p=0.02). In the BMCS, 1372 people were seronegative at baseline; 1259 (92%) had more than one follow-up test and 238 (17%) seroconverted. The overall HIV-1 incidence was 5.3 per 100 person-years (95% CI 4.7-6.1), with an increase and then a decline (inverted U-shaped curve, p=0.0001). Individuals aged 21 years and younger were at significantly higher risk of HIV infection than were those aged 30 years and older in the in the VCT (rate ratio 2.29, 95% CI 1.88-2.78, p<0.0001) and BMCS cohorts (1.99, 1.50-2.65, p<0.0001). Overall, drug use (p=0.03), drug use to enhance sex (p=0.0006), use of drugs for erectile dysfunction (p<0.0001), and 100% condom use (p<0.0001) increased over time, whereas the proportion of individuals reporting receptive anal intercourse decreased (p=0.004). INTERPRETATION:With a sustained high HIV-1 incidence and increasing drug use in MSM in Bangkok, we urgently need innovative and acceptable HIV prevention interventions, especially for young MSM. FUNDING:US Centers for Disease Control and Prevention.