Food deserts and swamps in major Brazilian cities: an analysis of inequities of race, income, and territory

Background Food deserts and swamps describe food environments characterized by limited access to healthy foods or by an overabundance of ultraprocessed foods. This study examines food deserts and swamps in Brazilian municipalities and their association with race, income and regional inequalities. Methods An ecological study analyzed 220,576 census tracts from 319 Brazilian cities (> 100,000 inhabitants). Food establishments were identified using Brazilian Annual Social Information Report (RAIS) data and classified according to Locais-NOVA methodology. Food deserts were defined by a density of establishments selling fresh or minimally processed foods below the 25th percentile, and food swamps by a density of establishments selling ultra-processed foods above the 75th percentile. Socioeconomic variables included the percentage of Black population and income. Poisson regression with robust variance estimated associations between socioeconomic variables and food deserts and swamps prevalence, stratified by macro-regions. Results Overall, 77% of census tracts were food deserts and 25% as food swamps. Census tracts in the highest tertile of Black residents had 24% higher food desert prevalence compared to the lowest tertile, while food swamp prevalence was 54% lower. Higher income was associated with lower prevalence of food deserts (South:PR:0.69; 95%CI: 0.66–0.71) but higher prevalence of food swamps (North:PR: 3.55; 95%CI: 3.17–3.98). Associations evidencing inequality were strongest in the North and Northeast regions. Conclusions Food deserts are highly prevalent in major Brazilian cities and are strongly associated with socioeconomic and racial inequalities. These findings highlight regional disparities and support geographically targeted policies to improve access to healthy foods.