Challenges of respondent driven sampling to assess sexual behaviour and estimate the prevalence of human immunodeficiency virus (HIV) and syphilis in men who have sex with men (MSM) in Singapore

ABSTRACT

There is a lack of representative samples to provide reliable and accurate seroprevalence of sexually transmitted infections (STIs) and human immunodeficiency virus (HIV) as well as behavioural information among men who have sex with men (MSM) in Singapore. We used respondent driven sampling (RDS) to recruit MSM. Participants completed a survey used by Asian Internet MSM Sex Survey (AIMSS) and were tested for HIV and syphilis. We compared the characteristics of the RDS participants with STI diagnosis against those who did not have any STI diagnosis in the past 6 months. We compared RDS participants with AIMSS participants. Of 72 MSM recruited, 1 was positive for HIV (1.3%) and 4 (5.5%) tested positive for syphilis. Median age was 30 years and majority was Chinese (69.4%). RDS participants who had any STI diagnosis reported to have more use of recreational drugs (P = 0.006), and lower condom use (P = 0.054). Comparing RDS participants (n = 72) with the AIMSS participants (n = 2075), RDS respondents had ≥1 male partner in the past 6 months (P = 0.003), more casual sex partners (P = 0.012) and more STI symptoms (P = 0.019). There was no difference in terms of HIV testing and recreational drug use. The HIV and syphilis seroprevalence rates from our study are similar to previous reports conducted in high-risk MSM. In contrast to other settings, RDS did not work well among MSM in Singapore. The public health implications of our study highlight the challenges in obtaining data for HIV surveillance in assessing prevalence and risk behaviours among MSM.