Longer-term Geospatial Food Access and the Incidence of Breast Cancer in Metropolitan Chicago

Diet quality contributes to breast cancer (BC) risk and is shaped in part by the neighborhood food environment. Yet, the long-term impact of residential food environments on BC incidence remains largely underexplored. We linked residential histories of 7,396 BC cases and 21,900 controls in Chicago (1990-2019) to food outlet data from the National Establishment Time Series. Cumulative (time-weighted), inverse distance-weighted (IDW), food access scores were derived along with nearest distance metrics for "healthy" and "less healthy" food access. Associations with incidence were stronger for walking than driving distance-based measures and for nearest distance-based measures over IDW measures. In multivariable logistic regression models, BC incidence decreased monotonically with shorter walking distance to the nearest healthy food outlet, reaching a 60% lower incidence for shorter vs. longer walking distances (OR=0.42, 95% CI=0.38, 0.48). Similarly, incidence increased monotonically with shorter walking distance to the nearest less healthy food outlet, reaching a 150% greater incidence for shorter vs. longer walking distances (OR=2.49, 95% CI=2.19, 2.83). This is the first study to use residential histories to define long-term, time-weighted geospatial food access metrics in BC epidemiology, highlighting how cumulative neighborhood environments have the potential to shape cancer risk and informing targeted interventions.