Abstract for: Do Theory-Derived Interventions Hold Up in Complex Systems? A Modeling Study of HIV Drug Resistance

HIV drug resistance (HIVDR) threatens the long-term effectiveness of antiretroviral therapy (ART) in people living with HIV. Increasing HIVDR underscores the need for targeted interventions. While qualitative system mapping has highlighted stigma, income, and quality of care as key leverage points for intervention design, quantitative system dynamics analyses, particularly in Tanzania, remain limited. This study aimed to develop a model to examine the influence of these factors on HIVDR dynamics. This study was conducted in the Dar es Salaam Urban Cohort Study area in Tanzania. A system dynamics model was developed using Stella Architect to simulate the HIV care continuum. The model was calibrated to cohort data (2019–2023). Key leverage points were incorporated, and sensitivity analyses assessed the effects of stigma, income, quality of care, contact frequency, and testing rates on outcomes among people living with HIV. Sensitivity analysis showed that stigma, income, and quality of care had a limited impact when modeled individually, despite being identified as deep leverage points. In contrast, transmission-related factors such as contact frequency and testing rates strongly influenced resistance trajectories. These findings suggest distal social determinants may have a greater impact when addressed jointly. However, further data and external validation are needed to strengthen model calibration and generalizability. While social determinants appeared quantitatively muted in this formulation, they remain critical targets for integrated, multi-level interventions. System dynamics modeling, combined with implementation science, offers a valuable framework for designing and testing such interventions. Language and grammar improvement