Integrating expert knowledge in a GIS to optimize siting decisions for small-scale healthy food retail interventions

Richard Casey Sadler

The availability of healthy foods in a neighborhood remains a key determinant of diet and diet-related disease in disadvantaged communities. Innovative solutions to the 'food desert' problem include the deployment of mobile markets and healthy corner store initiatives. Such initiatives, however, do not always capitalize on the principles guiding retail development and the possibilities of GIS-based data. Simultaneously, community partners are not always engaged effectively in the planning for such interventions, which limits acceptability and suitability of such work. This paper highlights the results of a participatory mapping exercise to optimize the siting of a planned healthy food retail intervention in Flint, Michigan. Potential sites are chosen by engaging experts in a three-stage mapping process that includes the analytic hierarchy process and point allocation of five key variables (including food access, socioeconomic distress, population density, access to transit, and proximity to neighborhood centers), as well as direct mapping of suitable sites. Results suggest a discrete set of areas-primarily in the northwestern quadrant of the city-where small-scale healthy food retail interventions might be most strategically located. Areas with the most consistent overlap between directly mapped sites and very high levels of suitability align well with neighborhoods which are distant from existing grocery stores. As a community-based strategy, this increases the opportunity for effectively improving neighborhood access to healthy foods by optimizing the potential sites for healthy food interventions. Community partners have already been active in using these results in project planning for just such an intervention.