Abstract The rapid spread of the SARS-CoV-2 virus has greatly impacted underserved populations. This project aims to understand social and behavioral determinants of COVID-19 testing and variations within sub-groups of this population. In partnership with the largest federally qualified health center in the United States, we will collect survey data and conduct a randomized experiment on 2,160 individuals (540 families) to evaluate the effectiveness of risk-based messaging and incentives that promote repeated testing for SARS-CoV-2 antibodies. In a 2 x 2 (Messaging x Incentive) factorial experiment, participants complete a comprehensive set of social and behavioral surveys to identify determinants of commitment to testing. They are then randomized by strata identified in the survey to receive personalized messaging promoting repeated testing. Messaging will focus upon either (1a) their household risk or (1b) their personal risk of COVID-19. They are also randomly assigned to either an incentive condition that (2a) insures against losing baseline rewards for initial testing, or (2b) that enters them into a lottery with a small chance to win $100 if they complete both tests. Both the insurance and lottery conditions carry the same incentive costs. Our previous work in similar populations demonstrated that adherence to planned health behaviors is higher with insurance-based incentives than cash payments of equal value. This experiment compares insurance-based incentives to lottery incentives that have been shown to be effective in multiple contexts. Finally, we evaluate if social and behavioral determinants of health result in heterogeneous treatment effects that can inform customization of incentive offerings in future programs devoted to increasing uptake of testing or vaccinations among underserved populations.