**Background:** Cancer remains a major public health concern in the US, with obesity and overweight status associated with 13 types of cancer accounting for 40% of all cancers diagnosed annually. Food deserts (areas with limited access to supermarkets and healthy food options) and food swamps (areas with an excess of proinflammatory food options like fast-food restaurants and convenience stores relative to healthy options) represent key barriers to healthy eating. Prior research has been limited, with only one US study examining food deserts and breast/colorectal cancer mortality, and none comprehensively assessing the association of both food deserts and food swamps with obesity-related cancer mortality across all US counties.
**Methods:** This ecologic cross-sectional study analyzed 3038 US counties (96.7% of all US counties) using 2010-2020 county-level cancer mortality data from the CDC and 2012-2020 food environment data from the USDA Economic Research Service. The outcome was age-adjusted obesity-related cancer mortality rates (per 100,000 population) for 13 cancer types, dichotomized at the median (≥71.8 vs <71.8 per 100,000). Food desert scores were calculated as the proportion of each county's population with low income (≤200% federal poverty threshold) and low access to grocery stores (>1 mile urban/>10 miles rural). Food swamp scores were calculated as the ratio of fast-food restaurants and convenience stores to grocery stores and farmers markets. Both were categorized into tertiles (low, moderate, high). Covariates included race/ethnicity, age, median household income, poverty rate, adult obesity rates, and adult diabetes rates. Generalized linear mixed models with binomial distribution and logit function were used, reporting adjusted odds ratios (AORs) with 95% CIs.
**Key Results:** Of 3038 counties, 758 (25.0%) had high obesity-related cancer mortality. Counties with high mortality had higher percentages of non-Hispanic Black residents (3.26% vs 1.77%), higher poverty rates (19.00% vs 14.40%), higher adult obesity rates (33.00% vs 32.10%), and higher adult diabetes rates (12.50% vs 10.70%) compared to low-mortality counties. Both food desert and food swamp scores were positively correlated with obesity-related cancer mortality (ρ=0.12 and ρ=0.08, respectively). Counties with high food desert scores had 59% increased odds of high obesity-related cancer mortality (AOR, 1.59; 95% CI, 1.29-1.94) compared to low food desert counties. Counties with high food swamp scores had 77% increased odds (AOR, 1.77; 95% CI, 1.43-2.19). A dose-response relationship was observed: counties with high food swamp scores had more than 2-fold increased odds of high obesity-related cancer mortality (AOR, 2.10; 95% CI, 1.67-2.63) compared to low food swamp counties. After adjusting for age, race, and poverty rate, high food swamp counties still showed nearly 30% increased odds (AOR, 1.29; 95% CI, 1.03-1.63), though further adjustment for adult obesity rates rendered associations non-significant.
**Clinical Implications:** This is the first study to comprehensively assess the association of both food deserts and food swamps with obesity-related cancer mortality across all US counties. The findings suggest that food swamps may be a more novel and comprehensive indicator of unhealthy food environments than food deserts, with a stronger association with obesity-related cancer mortality. The results highlight the potential importance of addressing systemic barriers to healthy food access, particularly in communities with lower socioeconomic status and higher proportions of racial/ethnic minority groups. The authors note that historic discriminatory practices such as redlining may contribute to these disparities. Future research should incorporate individual-level data, examine potential mediating roles of obesity and social determinants of health, and consider community-based interventions such as community gardens to improve healthy food access in underserved areas.