HIV viral suppression is critical to achieving benefits of antiretroviral therapy (ART) not only for individual health, but also for prevention of onward HIV transmission. Limited data on viral suppression derive from key populations in low-and-middle-income countries where these groups continue to experience disproportionately high rates of HIV transmission.
We recruited 12,022 men who have sex with men (MSM) and 14,481 people who inject drugs (PWID) across 27 cities in India (target=1000 per site) using respondent-driven sampling from 10/2012 – 9/2013. Participants had to be ≥18 years old and self-identify as male and report sex with a man in the prior year (MSM) or report injection drug use in the prior 2 years (PWID). 1,146 MSM and 2906 PWID were HIV positive. We characterize barriers and facilitators of viral suppression (HIV RNA<150 copies/ml) among ART initiators using multi-level logistic regression.
Of HIV positive MSM and PWID, 347 (25%) and 595 (24%) initiated ART, respectively. Median age was 35 years, 33% had primary school education or less and 44% of PWID were female. 67% of PWID reported active drug injection; 16% and 21% of MSM and PWID, respectively, had evidence of alcohol dependence. Of those who initiated ART, 268 (78%) of MSM and 443 (77%) of PWID were virally suppressed. Barriers to viral suppression among both groups included homelessness and harmful/hazardous alcohol use or alcohol dependence (Table). Among PWID, active drug injection was an additional barrier to suppression and women were significantly more likely to be suppressed than men. Among MSM, higher educational attainment was significantly associated with viral suppression. Neither needle exchange nor opioid substitution was associated with suppression among PWID. Adjustment for adherence (visual analogue scale) or premature discontinuation did not substantially attenuate these associations.
Levels of viral suppression among those on ART in this population were high, but are still short of new UNAIDS targets. Moreover, while viral suppression among those on ART was high, overall viral suppression (19% among MSM and 18% among PWID) was suboptimal such that population viral suppression will only be achieved through broad-based interventions that simultaneously support testing, linkage and adherence. Interventions need to be targeted towards those with active substance use, as they may also be most likely to transmit HIV to others.