Background Multiple approaches that can contribute to reducing obesity have been proposed. These policies may share overlapping pathways, and may have unanticipated consequences, creating considerable complexity. Aiming to illuminate the use of agent-based models to explore the consequences of key policies, this paper simulates the effects of increasing neighbourhood availability of good food stores, physical activity infrastructure and higher school quality on the reduction of black/white disparities in body mass index (BMI) in the USA.

Methods We used an agent-based model, with parameters derived from the empirical literature, which included individual and neighbourhood characteristics over the life course as determinants of behaviours thought to impact BMI. We systematically varied the strength of the 3 policy interventions, examining the impact of 125 different policy scenarios on black/white BMI disparities.

Results In the absence of any of these policies, black/white BMI disparities generally increased over time. However, we found that some combinations of these policies resulted in reductions in BMI, yielding decreases in the black/white BMI disparity as large as a 90%.

Conclusions Within the structure of relationships captured in this simulation model, there is support for the further use of agent-based simulation models to explore upstream policies as plausible candidates for the reduction of black/white disparities in BMI. These results highlight the potential insights into important public health problems, such as obesity, that can come from uniting the systems science approach with policy analysis.