Development and calibration of a mathematical model of HIV outcomes among Rwandan adults: Informing achievement of global targets across sub-populations in Rwanda

Background: We developed and calibrated the Central Africa-International epidemiology Databases to Evaluate AIDS (CA-IeDEA) HIV policy model to inform achievement of global goals, overall and across all sub-populations, in Rwanda. Methods: We created a deterministic dynamic model to project adult HIV epidemic and care continuum outcomes, overall and for 35 subpopulations (age group, sex, HIV acquisition risk, urbanicity). Data came from the Rwanda cohort of CA-IeDEA, 2004–2020; Rwanda Demographic and Health Surveys, 2005, 2010, 2015; Rwanda Population-based HIV Impact Assessment, 2019; and literature and reports. We calibrated the model to 47 targets by selecting 50 best-fitting parameter sets. Targets reflected epidemic, global goals and other indicators. Best-fitting sets minimized the summed absolute value of the percentage deviation (AVPD) between projections and targets. Good performance was mean AVPD ≤5% across best-fitting sets and/or projections within target confidence intervals; acceptable was mean AVPD >5%–15%. Results: Across indicators, 1,843 of 2,350 (78.4%) model projections were a good or acceptable fit to calibration targets. For HIV epidemic indicators, 247 of 300 (82.3%) projections were a good fit to targets, with the model performing better for women (80.3% a good fit) than for men (62.3% a good fit). For global goals indicators, 97 of 100 (97.0%) projections were a good fit; model performance was similar for women and men. For other indicators, 708 of 950 (74.5%) projections were a good or acceptable fit. Fit was better for women than for men (percentage virally suppressed only) and when restricting targets for number on ART to 2013 and beyond. Conclusions: The CA-IeDEA HIV policy model fits historical data and can inform policy solutions for achieving global goals across all sub-populations in Rwanda. High-quality population-based data and novel approaches that account for calibration target quality are critical to ongoing use of mathematical models for programmatic planning.