Effectiveness of respondent-driven sampling to recruit high risk heterosexual men who have multiple female sexual partners: differences in HIV prevalence and sexual risk behaviours measured at two time points.

Regular HIV bio-behavioural surveillance surveys (BBSS) among high risk heterosexual (HRH) men who have multiple female sexual partners is needed to monitor HIV prevalence and risk behaviour trends, and to improve the provision and assessment of HIV prevention strategies for this population. In 2006 and 2008 we used respondent-driven sampling to recruit HRH men and examine differences in HIV prevalence and risk behaviours between the two time points. In both surveys, the target population had little difficulty in recruiting others from their social networks that were able to sustain the chain-referral process. Key variables reached equilibrium within one to six recruitment waves and homophily indices showed neither tendencies to in-group nor out-group preferences. Between 2006 and 2008 there were significant differences in condom use with main sexual partners; numbers of sexual partners; and alcohol consumption. Further BBSS among this population are needed before more reliable trends can be inferred.