Profiling healthy neighborhoods through retail food spatial analysis: the case of Madrid

Abstract

Creating healthy neighborhoods is essential for enhancing residents’ quality of life, with the local retail food environment being a key factor. Madrid’s neighborhoods vary significantly regarding access to nutritious food options, potentially linked to their primary function as residential or tourist areas. This study proposes and applies a structured methodology, adapted from the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework, to profile the city’s neighborhoods using a modified Retail Food Environment Index (mRFEI) derived from OpenStreetMap (OSM) data. OSM, an open-source platform providing freely accessible geographic information, was used to identify and classify food outlets as “Healthy” or “Unhealthy” based on specific shop and amenity tags selected via expert knowledge. The mRFEI, calculated as the ratio of healthy to unhealthy outlets per neighborhood, was analyzed spatially using geospatial techniques. Results, visualized via interactive maps differentiating mRFEI scores below and above 100, reveal significant heterogeneity across Madrid, often indicating a lower relative presence of “Healthy”-classified outlets in central areas compared to several peripheral ones. By mapping and analyzing the distribution of food retailers, this study generates valuable insights to support informed policymaking aimed at promoting equitable access to healthy food options and contributing to health-focused urban development in Madrid. While demonstrating the utility of open data for large-scale assessment, this study acknowledges limitations inherent in OSM data and index simplification, suggesting avenues for future validation and refinement.