A Dyad-Based Intervention to Improve Adherence to Antiretroviral Therapy among People with HIV who Inject Drugs in Kazakhstan: Results of a Randomized Controlled Trial.

Kazakhstan has one of the fastest-growing HIV epidemics in the world, concentrated in people who inject drugs (PWID) with suboptimal antiretroviral therapy (ART) adherence. We compared the efficacy of a dyad-based social support intervention to standard of care on increasing ART adherence among PWID in Kazakhstan. PWID with HIV and on ART for ≥ 3 months ('index cases') were recruited from Kazakhstan's largest HIV clinic. PWID invited a treatment support partner (i.e., spouse, family member, or friend) to participate. Dyads were randomized 1:1 to a standard of care or intervention arm, the latter involving three intervention sessions in four weeks. Assessments were completed at baseline (pre-intervention), one-month (immediately post-intervention), and three-month and six-month follow-up. We used multilevel mixed models to examine changes in ART adherence among index cases over time between arms. Among the 66 index cases, those in the intervention arm who had a partner with HIV had a significantly higher adjusted mean change (AMC) in self-reported ART adherence immediately post-intervention (one-month follow-up) than index cases in the control arm with partners with HIV (AMC 16.55, 95% CI [2.77, 30.34], p = 0.038); this effect was maintained at the six-month follow-up (AMC 17.05, 95% CI [3.14, 30.95], p = 0.033). No significant differences between arms were found among index cases who had partners without HIV. No significant differences in electronic monitoring device-measured adherence or viral suppression were found between arms. The dyad-based intervention significantly increased self-reported ART adherence among PWID with partners with HIV, indicating the importance of considering dyad-level factors in interventions.Clinical Trial Registration: NCT03555396.