Heterogeneity Among Sex Workers in Overlapping HIV Risk Interactions With People Who Inject Drugs: A Cross-Sectional Study From 8 Major Cities in Pakistan.

Concerns remain regarding the heterogeneity in overlapping human immunodeficiency virus (HIV) risk behaviors among sex workers (SWs) in Pakistan; specifically, the degree to which SWs interact with people who inject drugs (PWID) through sex and/or needle sharing.Following an in-depth mapping performed in 2011 to determine the size and distribution of key populations at highest risk of HIV acquisition in Pakistan, a cross-sectional biological and behavioral survey was conducted among PWID, female (FSWs), male (MSWs), and hijra/transgender (HSWs) sex workers, and data from 8 major cities were used for analyses. Logistic regression was used to identify factors, including city of residence and mode of SW-client solicitation, contributing to the overlapping risks of drug injection and sexual interaction with PWID.The study comprised 8483 SWs (34.5% FSWs, 32.4% HSWs, and 33.1% MSWs). Among SWs who had sex with PWID, HSWs were 2.61 (95% confidence interval [CI], 1.19-5.74) and 1.99 (95% CI, 0.94-4.22) times more likely to inject drugs than MSWs and FSWs, respectively. There was up to a 3-fold difference in drug injecting probability, dependent on where and/or how the SW solicited clients. Compared with SWs in Larkana, the highest likelihood of drug injection use was among SWs in Multan (OR = 4.52; 95% CI: 3.27-6.26), followed by those in Lahore, Quetta, and Faisalabad.Heterogeneity exists in the overlapping patterns of HIV risk behaviors of SWs. The risk of drug injection among SWs also varies by city. Some means of sexual client solicitation may be along the pathway to overlapping HIV risk vulnerability due to increased likelihood of drug injection among SWs. There is a need to closely to monitor the mixing patterns between SWs and PWID and underlying structural factors, such as means of sexual client solicitation, that mediate HIV risk, and implement prevention programs customized to local subepidemics.