Underreported injection drug use and its potential contribution to reported increase in sexual transmission of HIV in Kazakhstan and Kyrgyzstan

BACKGROUND: We conducted a cross-sectional integrated bio-behavioral survey among sex partners of persons who inject drugs (PWID) to explore reasons for reported increase in reporting of heterosexually transmitted HIV in Kazakhstan and Kyrgyzstan. METHODS: Sexual partners of PWID were recruited through PWID. Behavioral data were collected through semi-structured interviews. Dried blood spots were obtained and tested for HIV and hepatitis C virus antibodies (HCVAb). Descriptive univariate and bivariate analyses, and multivariate analyses using logistic regression modeling were performed to identify factors associated with HIV and HCV infections. RESULTS: Among 1982 sex partners of PWID, overall HIV prevalence was 6.4%; 5.1% and 12.9% among those reported never and ever injecting drugs, respectively (p < 0.001). Overall, HCVAb prevalence was 21.3%; 15.0% and 53.9% among those reported never and ever injecting drugs, respectively (p < 0.001). Of HCV-positive participants, 58% and 34% (p < 0.001) reported prior history of injecting drug use among men and women, respectively. HIV prevalence was lower among HCV-negative (4.2%) compared to HCV-positive participants (14.4%) (p < 0.001). HIV prevalence was 3.5% (95%CI = 2.4–4.6) in a subset of female participants with no reported prior injecting drug use history and who were HCVAb-negative and did not report having an HIV-positive sex partner. Participant sex and number of sex partners as well as use of condoms in the past 12 months were not associated with HIV seropositivity. CONCLUSIONS: High prevalence of HCV among sex partners of PWID who denied ever injecting drugs suggests underreporting of injecting practices. The increased attribution of HIV infection to sexual transmission based on self-report may be partly explained by underreporting of injection drug use due to stigmatization of this behavior.