Using MOST to Optimize an HIV Care Continuum Intervention for Vulnerable Populations

More than half of persons living with HIV/AIDS (PLHA) in the U.S. are insufficiently engaged in HIV primary care and not taking antiretroviral therapy (ART), mainly African Americans and Latinos. In the proposed project, two experienced and productive behavioral scientists will employ a potent and innovative research methodology, the Multiphase Optimization STrategy (MOST), to develop a highly efficacious, efficient, scalable, and cost-effective intervention to increase engagement along the HIV care continuum. Whereas randomized controlled trials (RCTs) are valuable for evaluating the efficacy of multi-component interventions as a package, they are not designed to evaluate which specific components contribute to efficacy. MOST, a pioneering, engineering-inspired framework, addresses this problem through highly efficient randomized experimentation to assess the performance of individual intervention components and their interactions. The investigators propose to use MOST to engineer an intervention to increase engagement along the HIV care continuum for African American and Latino PLHA not well engaged in care and not taking ART. Further, the intervention will be optimized for cost-effectiveness. This efficiency and cost-effectiveness are critical in a time of constrained resources, and will also increase the intervention's future scalability. NIH has signaled its interest in MOST, and this is the first study to apply it in the field of adult HIV treatment. A similar set of multi-level factors impede both HIV care and ART initiation for African American and Latino PLHA, primary among them individual (e.g., substance use, distrust, fear), social (e.g., stigma), and structural-level barriers (e.g., difficulties accessing ancillary services). Guided by a multi-level social cognitive theory, the study will evaluate 5 distinct intervention components (i.e., Motivational Interviewing sessions, pre-adherence preparation, support groups, peer mentorship, and patient navigation), each designed to address a specific barrier to HIV care and ART initiation. These components are well-grounded in the empirical literature and were found acceptable, feasible, and promising with respect to efficacy in a preliminary study. Study aims are: 1) using a highly efficient experimental design, identify which of 5 components contribute meaningfully to improvement in viral suppression, and secondary outcomes of ART adherence and engagement in HIV primary care; 2) identify mediators and moderators of component efficacy; and 3) using a mathematical modeling approach, build the most cost-effective and efficient intervention package from the efficacious components. A heterogeneous sample of African American and Latino PLHA (with respect to age, substance use, and sexual minority status) will be recruited with a proven hybrid sampling method using targeted sampling in community settings and peer recruitment (N=512). This highly innovative and significant study, which addresses a high-priority research area (NIH NOT-OD-15-137), will produce an HIV care continuum intervention for the nation's most vulnerable PLHA, optimized for cost-effectiveness, and with exceptional levels of efficacy, efficiency, and scalability.