Trends in unprotected intercourse among heterosexual men before and after brothel ban in Siem Reap, Cambodia: a serial cross-sectional study (2003–2012)

BACKGROUND Following Cambodia's implementation of the 100% condom use program with enforcement of condom use and STI treatment services for sex workers in 2001, sexually transmitted infection and HIV declined markedly. In 2008, Cambodia implemented a law to ban brothel-based sex work. We reported trends in unprotected vaginal intercourse with sex workers among heterosexual men buying sex before (2003-2008) and after (2009-2012) the brothel ban in Cambodia. We also determined the association of brothel ban with these men's reports of unprotected intercourse with sex workers. METHODS In this serial cross-sectional study, we collected yearly behavioural data on random cross-sectional samples of heterosexual men buying sex who attended the only government health centre in Siem Reap for voluntary confidential counselling and testing (VCCT) between 2003 and 2012. We used multivariable Poisson regression analysis on the 10-year data of 976 men to obtain the adjusted prevalence ratio (aPR) of unprotected intercourse in the last 6 months by brothel closure. RESULTS Men buying sex from non-brothel-based sex workers increased almost 3-fold from 17% in 2007-2008 before brothel closure to 55% in 2011-2012 after brothel closure (p < 0.001). Unprotected intercourse with sex workers in the last week increased significantly from 37% (2003-2004) before brothel closure to 65% (2011-2012) after brothel closure. This increase corresponded closely with the increase in self-reported unprotected intercourse from 35% to 61% by the sex workers (n = 1805) attending the same clinic for VCCT. Brothel closure was associated with an increased risk (aPR: 1.65; 95% CI: 1.40-1.94) of unprotected intercourse with sex workers. HIV prevalence in the heterosexual men declined significantly from 26% in 2003-2004 to 4.8% in 2007-2008 and 0 case in 2009-2010 before increasing to 5.6% in 2011-2012. CONCLUSION Our findings suggest that the brothel ban had led to an increase in unprotected intercourse with all sex workers for men buying sex. This effect could be attributed to reduced condom access, a consequence of the lack of feasibility to implement the 100% condom use program following the brothel ban. The ban on brothels in Cambodia should be reviewed.