Analysis of Census Tract-Level Racial Disparities in Food Access from 1998 to 2009 in Cook County, IL

Background: African Americans (AA) in the United States have higher rates of chronic health problems than their White counterparts. Research has shown that obesity is linked to chronic health problems and that geographic areas with limited access to healthy food have higher rates of obesity. The segregated nature of Cook County, IL, allows examination of disparate geographic access to quality grocery stores in AA or White communities, and investigation of how food access has changed with neighborhood demographics over time. Methods: In this analysis, we examined change in availability of quality grocery stores by census tract over the years of 1998 to 2009, a time of transition for areas of Cook County. Geographic location of high grocery stores, using data obtained from Dun & Bradstreet, was mapped using kernel density (KD) estimation methods in ArcGIS and average KD scores for each census tract were used to divide tracts into quintiles of risk. Comparisons of high and low risk tracts were made by examining census-level sociodemographic variables. Multivariate longitudinal logistic models were also constructed to determine significant predictors of high risk areas over time. Results: Census-level variables such as low income, AA race, and poverty were significantly (p<0.05) associated with low density of quality grocery stores. In multivariate analyses, change in demographic variables at the census level over time predicted change in density of quality grocery stores, resulting in a decrease in quality food access in increasingly predominant AA communities from 1998 to 2009.