PROJECT SUMMARY Despite its universal provision of ART, South Africa remains the epicentre of the HIV pandemic, illustrating the urgency of addressing the disconnect between accessibility and utilization of care within the highly mobile population of the ‘economic engine’ of Africa. Migration is an established driver of the HIV epidemic in South Africa: mobility increases risk of HIV acquisition and death, and, after diagnosis, delays initiation and disrupts engagement in HIV care. Thus, migrants disproportionately miss the health and economic benefits of ART, and their partners face increased risk of onward transmission. However, little is known about the effect of migration on HIV outcomes and HIV care utilization preferences. The proposed research will address this knowledge gap using data from two health and demographic surveillance systems (HDSSs) that exist in South Africa: the Africa Health Research Institute HDSS and the Agincourt HDSS. These HDSSs are ideal data sets for this research given that, within them, migrants are two to four times as likely to die from HIV/TB as their non-migrant counterparts; analyses across these cohorts will highlight heterogeneity in migration patterns, epidemic experiences, and geographic regions. Data from the Africa Health Research Institute HDSS collect biomarker and geolocation data yearly, uniquely enabling the study of HIV outcomes and geospatial analyses (Aim 1). We will use geospatial analyses to create measures of migration distance, pattern, and clustering; we will create a Cox proportional hazards model to evaluate the effect of migration on viral suppression among persons with HIV (PWH) who have initiated ART. Characterizing the effect of migration on care selection among PWH (Aim 2) will be feasible and important to undertake in both populations. We will create a latent class transitional model, which is ideal for identifying heterogeneity in preference structures in care utilization, in each population to identify unobserved or ‘latent’ classes of care preferences using observed data on characteristics of care that PWH utilized (e.g., fees, distance, type of provider/facility), and the individual-level predictors (e.g., migration, sociodemographic characteristics) of belonging to each class. Identifying the effect of migration on HIV care utilization, which relates to its effect on HIV outcomes, can inform the creation of urgently- needed migrant-specific HIV policies within a health system that currently is unresponsive to human mobility. Specifically, this research may be used to contextualize interventions to increase long-term HIV care engagement among migrants in South Africa, whose migration-health relationship is likely a harbinger for sub-Saharan Africa, maximizing this work’s contribution to ending HIV within the world’s highest-burden continent.