Complex systems approaches to identify policy levers to reduce racial/ethnic disparities in diet and obesity in cities

PROJECT SUMMARY Racial/ethnic disparities in diet and obesity are remarkably consistent across U.S. cities. First, we will use group model building to systematically engage academic, policy, and community stakeholders to build capacity for systems thinking, develop and refine a "map" of the multilevel factors that drive diet disparities, and identify policy levers to reduce diet disparities in cities. The need for this work is motivated by the lack of an existing conceptual framework that explicates mechanisms via which obesogenic environments and systematic structural disadvantage disproportionately affect minorities and lead to disparities. Previous research and existing conceptual frameworks have identified myriad influences on diet among the general population, but a more specific conceptual framework can advance understanding of social, environmental, and policy factors that work in combination to constrain healthy food choices of Blacks, Latinos, and other racial/ethnic minorities. Second, we will implement an agent-based simulation model (ABM) to examine how residential segregation, the inequitable distribution of food outlets, the lower price of unhealthy foods, and income inequality work in combination to constrain food choices of racial/ethnic minorities and lead to diet and obesity disparities. The ABM bridges lines of research conducted by our group and others that have used ABM to examine how food access and food prices separately affect diets. By integrating these separate modeling paradigms, we can examine how diet disparities emerge due to intersecting disadvantage in food access and affordability. In the ABM, individual-agents in a virtual city make a series of daily decisions about where to shop for food, what types of food to purchase, and what to eat. Each decision is based on simple rules that reflect influences on food purchasing and diet, including household food budgets; travel costs to food stores; between-store variation in price, inventory, and quality; and the prices of 12 nutritionally important food categories (e.g., protein, whole grains) and 6 beverage categories. We use gold standard data regarding household income and food spending, food prices and purchasing, and diet. We propose two uses for the ABM: First, we will assess the impact of job and income loss related to the COVID-19 pandemic and federal policies that restrict eligibility and enrollment of immigrants in food assistance programs - both of which have a disproportionate effect on minorities and thus are likely to exacerbate disparities. Second, we will engage policy stakeholders to inform dissemination and evaluate how scaling up existing pilot programs (e.g., healthy food delivery, multiplying the value of SNAP dollars spent at farmers' markets, increasing healthy food access in minority neighborhoods) and implementing current policy proposals (e.g., USDA proposal to replace SNAP with "harvest boxes") will exacerbate or reduce diet disparities. The ABM is grounded in the Philadelphia context, but the research questions and findings are highly relevant to diet disparities in essentially all U.S. cities.