Background. Although persons living with HIV (PLWH) are living longer, little is known about factors that contribute to their stopping antiretroviral therapy (ART), what prompts re-starting, and what contributes to sustaining/persisting with ART in times of hardship when risk for discontinuation may be high. We explored these phenomena, focusing on Black and Latinx (BL) PLWH from low SES backgrounds and with serious barriers to ART using a sequential explanatory mixed methods design.

Methods. Participants were BL-PLWH (N=320) with non-suppressed viral load, recruited in New York City. Participants completed a structured assessment battery with reliable/valid measures. A subset (N=41) was randomly selected for in-depth interviews. Bivariate linear regression was used to examine socio-demographic, background, and health-related predictors of 1) rates of stopping/starting ART and 2) the longest period on ART (in months). Qualitative data were analyzed with a systematic content analysis approach, and qualitative and quantitative results were integrated.

Results. Participants were mostly men (72%), Black (70%), 47 years old on average (SD=11 years) and had lived with HIV for an average of 19.7 years (SD=9.1 years). Participants had stopped/started ART an average of 10.3 times (SD=23.5 times). The longest interval of sustained ART was X=38.6 months (SD=48.4 months). Higher rates of stopping/starting ART were associated with being younger, male, transgender, less time living with HIV, more adverse childhood experiences, lifetime homelessness, lower motivation for ART, and moderate-to-high-risk cannabis use. Fewer months of sustained ART use were associated with being Black, higher social support, moderate-to-high-risk alcohol use, and stigma. Qualitative results indicated a convergence of multiple intersecting risk factors for stopping/starting ART and highlighted challenges inherent in managing HIV over decades in the context of poverty. Primary among these was unstable housing, which contributed to social isolation and mental health and substance use problems, which, in turn, often prompted participants to sell ART. Quantitative/qualitative findings were complementary and when taken together described risk factors for stopping ART, protective factors, and the mechanisms by which risk/protective factors operated.

Conclusions. Findings highlight both challenges and strengths inherent in ART persistence and point the way to policy, structural and social-level interventions to support BL-PLWH.