Individual-level barriers to breast cancer screening are well documented. Although community factors, such as area deprivation or availability/distribution of medical providers, are independently associated with breast cancer incidence, survival, and with several behaviors (e.g., walking and smoking), their association with screening for breast cancer has received very little attention. The use of geographic information systems (GIS) and multilevel modeling facilitate new and exciting opportunities that could shed new light on potential barriers to screening that could be used to increase breast cancer screening by multilevel interventions. In order to develop effective multilevel interventions to increase screening and to identify where to implement such interventions, it becomes important to investigate the association of community indicators at different geographic levels (zip code, census tract, block group). GIS analyses will be combined with quantitative epidemiological methods to examine the relationship of neighborhood characteristics to breast cancer screening. The goals of the proposed, population-based study are to determine areas where breast cancer screening is underutilized in St. Louis, if women living in neighborhoods characterized by adverse socioeconomic conditions are less likely to be screened for breast cancer, and if this association depends on the geographical classification used to obtain these community indicators. We will collect primary data of 1400 randomly-selected women in the City of St. Louis via telephone interviews (sociodemographics, barriers to screening) and use existing data from the 2000 census (area deprivation and residential segregation), the Food and Drug Administration (mammography location), the Missouri Department of Health and Senior Services (location and distribution of primary care physicians), the City of St. Louis, and the federal Department of Housing and Urban Development. GIS will be used to geographically link the data and construct multiple data layers by respondent address. Advanced multilevel and spatial analyses will be conducted to identify underlying factors for the spatial variation of breast cancer screening and where to target interventions aimed at increasing screening could be implemented.
