Comparative Effectiveness of Community-Based vs Clinic-Based Healthy Choices Motivational Intervention to Improve Health Behaviors Among Youth Living With HIV A Randomized Clinical Trial

IMPORTANCE Youth living with HIV make up one-quarter of new infections and have high rates of risk behaviors but are significantly understudied. Effectiveness trials in real-world settings are needed to inform program delivery. OBJECTIVE To compare the effectiveness of the Healthy Choices intervention delivered in a home or community setting vs a medical clinic. DESIGN, SETTING, AND PARTICIPANTS This randomized clinical trial was conducted from November 1, 2014, to January 31, 2018, with 52 weeks of follow-up. Participants, recruited from 5 adolescent HIV clinics in the United States, were youths and young adults living with HIV aged 16 to 24 years who were fluent in English, were currently prescribed HIV medication, had a detectable viral load, and had used alcohol in the past 12 weeks. Individuals with an active psychosis that resulted in an inability to complete questionnaires were excluded. Data were analyzed from May to December, 2019. INTERVENTIONS Participants were randomized to receive the Healthy Choices intervention in either a home or clinic setting. Four 30-minute individual sessions based on motivational interviewing to improve (1) medication adherence and (2) drinking behavior were delivered during 10 weeks by trained community health workers. In session 1, participants chose which behavior to discuss first. Using motivational interviewing strategies, the community health worker elicited motivational language, guided the development of an individualized change plan while supporting autonomy, delivered feedback, and addressed knowledge gaps. Session 2 focused on the second target behavior. In subsequent sessions, community health workers reviewed the individualized change plan, monitored progress, guided problem solving, and helped maintain changes made. MAIN OUTCOMES AND MEASURES Primary outcomes were viral load and alcohol use change trajectories during 52 weeks of follow-up. Alcohol use severity and frequency were measured using the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) (scores range from 0 to 33, with higher scores indicating greater severity of alcohol-related problems) and number of drinks consumed each day over a 30-day period, with timeline followback. RESULTS A total of 183 young people living with HIV (145 male [79.2%]; mean [SD] age, 21.4 [1.9] y) were randomized to the home setting (n = 90) or clinical setting (n = 93). Using growth-curve analysis, both groups showed declines in viral load after the intervention: among participants with available viral load information, in the home group, 12 participants (21%) had an undetectable viral load at 16 weeks, 12 (22%) at 28 weeks, and 10 (20%) at 52 weeks; in the clinic group, 16 participants (24%) had an undetectable viral load at 16 weeks, 20 (39%) at 28 weeks, and 18 (35%) at 52 weeks. However, the clinic group maintained gains, whereas those counseled at home had a significantly different and increasing trajectory during follow-up (unstandardized beta = -0.07; 95% CI,-0.14 to -0.01; P = .02). A similar pattern was observed in ASSIST scores during follow-up, with reduced ASSIST scores in the clinic group (unstandardized beta = -0.44; 95% CI,-0.81 to -0.07; P = .02). CONCLUSIONS AND RELEVANCE In this trial, the Healthy Choices intervention resulted in improvements in viral load and alcohol use over 12 months. Unexpectedly, the clinic setting outperformed home-based delivery for viral suppression.Although cross-sectional differences in ASSIST scores were nonsignificant, clinic delivery did improve the trajectory of ASSIST scores during follow-up. Thus, clinics may be the more effective site for interventions aimed at viral load reduction for young people living with HIV.