20th International AIDS Conference - Melbourne, Australia


WEPDC0101 - Poster Discussion Abstract

A comparison between respondent-driven sampling and time-location sampling among men who have sex with men in Shenzhen, China

Presented by Rui Cai (Netherlands).

R. Cai1, J. Zhao2, L. Chen2, J.H. Richardus1, S. De Vlas1

1Erasmus MC, University Medical Center Rotterdam, Public Health, Rotterdam, Netherlands, 2Shenzhen Center for Disease Control and Prevention, Shenzhen, China

Background: The HIV epidemic is concentrated among high-risk populations, in particular men who have sex with men (MSM) in China. With a rapid increase in HIV prevalence, MSM has become a key population for HIV control in China. It is difficult to acquire representative samples of this hidden population. Respondent-driven sampling (RDS), based on peer referral, and time-location sampling (TLS) based on random selection of venue-day-time periods, are among the most commonly used sampling methods. The aim of this study was to identify differences in HIV-related characteristics of MSM recruited by these two methods to provide valuable information for better surveillance and intervention.
Methods: We recruited 598 MSM by 10 seeds in RDS and 493 MSM from gay bars, saunas, recreational centers and parks in TLS survey in Shenzhen, China in 2010. In both surveys, information about socio-demographics, risk behaviors, utilization of HIV-related services (i.e. HIV testing and counseling, condom promotion, peer education that are available at venues) were collected by a same questionnaire and all participants were tested for HIV. We compared these HIV-related characteristics between samples of RDS and TLS.
Results: The HIV prevalence was comparable in RDS and TLS MSM (10% vs. 8%, p = 0.35). TLS recruited larger proportions of hidden subgroups than RDS: TLS MSM were older (30 vs. 28 years, p < 0.001), less educated and more likely to be migrants (without Shenzhen hukou registration, 94% vs. 86%, p < 0.001), to be non-gay identified (45% vs. 29%, p < 0.001) and to engage in risky sexual behaviors such as having multiple anal partners (69% vs.54%, p < 0.001). On the other hand, TLS MSM were more likely to have been covered by HIV-related services (79% vs. 70%, p = 0.007).
Conclusions: In Shenzhen, TLS is more effective to reach the hidden segments of MSM. But because HIV-related services are already commonly available at gay venues in Shenzhen, RDS is more informative for further allocating prevention efforts than TLS. Furthermore, researchers and public health authorities should take into account the different the sample composition of RDS and TLS and apply sampling methods consistently when evaluating trends over time.

Download the e-Poster

Back to the Programme-at-a-Glance