Identifying priority neighborhoods for mobile cancer screening using georeferenced data

Abstract Background: Mobile screening units (MSUs) are an evidence-based approach known to reduce barriers and increase access to preventive services such as cancer screening. Less is known about how to plan and prioritize where MSUs are deployed to maximize impact and reduce the burden of disease. We used a geographic-based approach to create an index to identify which census tracts were of greatest priority for cancer screening, thereby creating a priority scoring metric for deploying the MSU across our 7-county cancer center catchment area. Methods: We assessed publicly available data reported at the census tract level for cancer relevance, choosing those data that were most likely to be associated with disparities in cancer screening or outcomes. This included data from the Social Vulnerability Index (socioeconomic status, percent of residents without health insurance, percent of residents reporting minority race or ethnicity, percent renting their housing, and percent with no transportation) as well as data from the Centers for Disease Control and Prevention (CDC) Places on the percent screened for breast cancer, cervical cancer, and colon cancer. Data was transformed from CDC Places to be percentage in need of breast cancer, cervical cancer, and colon screening rather than percentage screened. To construct the index, each of the variables was ranked from highest to lowest across all census tracts in Pennsylvania and New Jersey with a non-zero population (n=1,184). A percentile rank was then calculated for each census tract over each of these variables. Finally, an overall percentile rank for each tract was calculated. Index scores could range from 0.000000 (low priority for cancer screening) to 1.000000 (high priority for cancer screening). Results: There are 1,184 census tracts within our cancer center catchment area. Gloucester County NJ has the fewest census tracts (n=63) while Philadelphia County PA has the most census tracts (n=384). The mean index score across all census tracts was 0.400593 (SD = .198556) with a range of 0.000000 to 0.938343. Philadelphia County had the highest mean index score (0.400953) while Bucks County PA had the lowest mean index score (0.281576). In Philadelphia County, 76% of census tracts are above the mean index score, indicating high need for cancer screening, while in Bucks County, only 22% of census tracts are above the mean index score. A catchment area map by census tract visualizes the index score across the counties. Conclusion: A cancer screening index for census tracts in a cancer center catchment area can be used to prioritize resources such as a MSU in an evidenced-based manner. Next steps involve validating the index score against cancer disparities data in our catchment area and creating disease specific index scores to use with cancer specific screening initiatives. Citation Format: Krista Mar, Yawei Song, Khaldoun Hamade, Maria Katerina Alfaro, Alex Wrem, Christopher McNair, Amy Leader. Identifying priority neighborhoods for mobile cancer screening using georeferenced data [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr A017.