Characteristics of sex partners and sexual partnership correlates of inconsistent condom use among male injection drug users in India.

Previous studies have established the risky behaviors of IDUs in India, and that IDUs are sexually active; however, there is a need to better understand the nature of sexual partnerships of IDUs. A total of 783 (Delhi) and 766 (Imphal) male IDUs were recruited into the study through respondent-driven sampling. We examined characteristics of sex partners of male IDUs and individual and sexual partnership characteristics associated with unprotected sex in Delhi and Imphal. While 16.8% of sexual partnerships in Delhi were male-to-male, there were almost no male-to-male partnerships in Imphal. The majority of partners of male IDUs in Delhi (82.5%) and Imphal (92.3%) do not inject drugs, with the exception of male partners of male IDUs in Delhi. Commercial partners (females: 58.3%; males: 71.3%) were the most common type of sex partners of male IDUs in Delhi, while regular partners (65.2%) were the most common type of sex partners in Imphal. In Delhi, characteristics of sex partners significantly associated with unprotected sex were being male/transgender (AOR 2.2; 95% CI: 1.2-4.0), being a regular (AOR 5.1; 95% CI: 2.8-9.4) or non-regular partner (AOR 2.7; 95% CI: 1.7- 4.5), and sharing needles/syringes with the index IDU (AOR 2.8; 95% CI: 1.4-5.3). In Imphal, partner characteristics associated with unprotected sex were being a regular (AOR 10.1; 95% CI: 41-25.1) or non-regular partner (AOR 3.4; 95% CI: 1.5-7.6), and living outside of town or state (AOR 3.3; 95% CI: 1.2-9.6). Enhanced understanding of disassortative sexual mixing and context of unprotected sex within sexual partnerships may enhance sexual risk reduction interventions for IDUs.

High HIV incidence in a cohort of male injection drug users in Delhi, India.

BACKGROUND: India has an estimated 177,000 injection drug users (IDU) with a national HIV prevalence of 7.14%. Reliable estimates of HIV incidence are not available for this population. METHODS: We report HIV incidence in a cohort of male, HIV-negative IDUs recruited through peer-referral, targeted outreach and as walk-in clients in Delhi from May to October, 2011. Fourth-generation Antigen-Antibody tests were used to diagnose new infections and results were confirmed using Western blot tests. HIV incidence based on HIV seroconversion was calculated as number of events/person-years. Cox regression was used to identify significant (p<0.05) seroconversion predictors. RESULTS: A total of 2790 male HIV-negative IDUs were recruited at baseline; 67.4% (n=1880) returned for their first follow-up visit and 96% (n=1806) underwent HIV testing. Participants were followed for a median of 9.7 months. A total of 112 new HIV infections occurred over a cumulative 1398.5 person-years of follow-up resulting in an incidence rate of 8.01 new infections/100 person-years (95% CI: 6.65-9.64); 74% of these participants reported risky injection practices in the past month. In multivariate analysis, moderate-high risk injection behaviors (Adjusted Hazard Ratio [AHR] 2.59; 95% CI 1.45-4.62) were associated with a higher risk of new infections. CONCLUSIONS: Male IDUs in Delhi continue to practice unsafe injection practices leading to high sero-incidence despite the availability of HIV prevention services offered through targeted intervention programs.

Strategies for recruiting injection drug users for HIV prevention services in Delhi, India.

BACKGROUND: We utilized multiple recruitment approaches to recruit IDUs in a longitudinal cohort study to examine HIV incidence and behavior change pre- and post-introduction of comprehensive HIV prevention services. METHODS: IDUs were recruited through peer referral, targeted outreach by outreach workers (ORWs) and as walk-in clients at drop-in centers. Participants received monetary compensation for participation (USD 0.80). Participants were given recruitment coupons to recruit peers (regardless of recruitment method). For peer referral, participants received a food coupon, as secondary compensation, for each peer he/she successfully recruited. We report the profile of IDUs by recruitment method, based on the baseline behavioral survey and HIV test results. Cost per IDU recruited by recruitment method was also calculated. RESULTS: A total of 3,818 IDUs were recruited between May 2011 and October 2011. More than half of the study participants were recruited through targeted outreach (ORW: 53.6%; peer-referral: 26.3%; walk-ins: 20.1%). Of the participants who were given recruitment coupons, 92.7% recruited no peers. Those who successfully recruited at least one peer were significantly more likely to be in a stable living accommodation compared to those who did not recruit any peers (51.1% versus 42.7%; p < 0.05). Only 45.9% of the food coupons were claimed for successful recruitment of peers. Peer-referred IDUs were more likely to be living with family or relatives (50.7% versus ORW: 40.1% and walk-in: 39.8%; p < 0.001) rather than on the street or shared housings compared to the other two recruitment modes. Walk-ins were more likely than peer-referred and ORW-referred IDUs to be HIV-positive (walk-ins: 26.1%; peer-referred: 19.1%; ORW: 19.9%; p < 0.01) and have risky injection practices (walk-ins: 62.2%; ORW: 57.0%; peer-referred: 58.6%; p < 0.05). The cost per IDU recruited through ORW referral method was the most costly at USD 16.30, followed by peer-referral at USD 8.40 and walk-in at USD 7.50. CONCLUSION: When recruiting a large number of IDUs, using multiple recruitment modes is ideal with regard to diversification of IDU characteristics and risk profile. Although it was the most costly, ORW recruitment was more effective than the other two methods. Lack of monetary compensation for successful recruitment of peers may have hampered peer-referral.

High uptake of HIV testing in a cohort of male injection drug users in Delhi, India: prevalence and correlates of HIV infection.

We report baseline findings from a longitudinal cohort study to examine HIV incidence, high-risk injection and sexual behaviors of 3,792 male injection drug users (IDUs) in Delhi. The majority (95.4 %) accepted HIV testing; HIV prevalence was 21.9 %. In multivariate analysis, belonging to states adjacent to Delhi (AOR: 1.23; 95 % CI: 1.07-1.52), earning INR 500-1,500 (AOR: 2.38; 95 % CI: 1.43-3.96); duration of drug use 2-5 years (AOR: 2.02; 95 % CI: 1.09-3.73), 6-10 years (AOR: 2.81; 95 % CI: 1.55-5.11), ≥11 years (AOR: 3.35; 95 % CI: 1.84-6.11); prior HIV testing (AOR: 1.60; 95 % CI: 1.35-1.91), self-reported risky-injection behavior (AOR: 1.60; 95 % CI: 1.33-1.92), and utilization of harm-reduction services (AOR: 1.32; 95 % CI: 1.11-1.58) were positively associated with HIV infection. Alcohol use ≤2 times/week (AOR: 0.67; 95 % CI: 0.55-0.82) or ≥3 times/week (AOR: 0.74; 95 % CI: 0.54-1.01), unit increase in age (AOR: 0.99; 95 % CI: 0.98-1.00), ≥7 years of schooling (AOR: 0.82; 95 % CI: 0.66-1.02) and unsafe sex with any female partner (AOR: 0.69; 95 % CI: 0.55-0.86) were negatively associated with HIV infection. HIV prevalence remains high among male IDUs in Delhi. HIV prevention programs should include comprehensive package of services for IDUs.

Assessment of unsafe injection practices and sexual behaviors among male injecting drug users in two urban cities of India using respondent driven sampling.

Designing interventions to reduce HIV transmission among injecting drug users (IDU) requires reliable estimates of risk behaviors. We present population-based estimates for unsafe injection practices and sexual risk behaviors among male IDUs recruited through respondent driven sampling in India (Delhi: 783; Imphal: 766). IDUs in Delhi, mostly street-based (68%), reported injecting pharmaceutical agents and a greater frequency of injections/day. IDUs in Imphal, mostly home-based (98%), used heroin/opioids and injected less frequently. Needle sharing was common (Delhi: 33%; Imphal: 43%). Sixty-five percent of IDUs in Delhi and 55% in Imphal were sexually active during the previous year. Multiple sexual partners were more frequent in Delhi (49% vs 21%); IDUs in Imphal reported more regular sex partners (82% vs 44%). Consistent condom use with regular partners was extremely low (Delhi: 8%; Imphal: 19%). HIV testing was infrequent (Delhi: 37%; Imphal: 49%). IDUs are a heterogeneous group with different prevention needs requiring need-based tailored prevention interventions.