Improvements in Geospatial Disparity of Breast Cancer Stage at Diagnosis under Medicaid Expansion in Ohio, A Space-Time Cluster Analysis

Authors: Dong W, Kim U, Koroukian SM, Rose J

Category: Cancer Health Disparities
Conference Year: 2021

Abstract Body:
The purpose of the study is to examine whether geospatial disparities in advanced-stage breast cancer at diagnosis were reduced after the implementation of Medicaid expansion under the Affordable Care Act in Ohio. Methods We examined the association between the implementation of Medicaid expansion and changes in geospatial clustering of advanced-stage breast cancers in Ohio. The data were obtained from the Ohio Cancer Incidence Surveillance System and the study included women age 20-64 years diagnosed with invasive breast cancer between 2010 and 2017. We used an ordinal-based space-time scan statistic on SEER (Surveillance, Epidemiology, and End Results) summary stage, local, regional, and distant, corresponding to order 1, 2, and 3, where increasing numbers represent more severe disease. The study period was dichotomized into pre-expansion (2010-2013) and post- expansion (2014-2017). We compared the presence of clusters of more severe stage cases pre- and post-expansion. We also conducted a subanalysis of women with private insurance, allowing for an examination of existing spatiotemporal trends in stage at diagnosis independent of Medicaid expansion. Results Our study population included 33,537 women, after excluding unknown stage/unstaged cases. Based on the distribution of cases by stage in a given area compared to that of the state, we observed changes in clusters of more severe stage cases from pre- to post-expansion. Specifically, there were eight space-time clusters detected, of which two were statistically significant (p<.05) and six were non-significant (p>.4). All of the clusters were in the pre- expansion period, but none was identified in the post-expansion period. These clusters were in the four largest metropolitan areas in Ohio with the two significant clusters in the city of Toledo and a suburb of Cincinnati. Patients in these areas were more likely to be African American, unmarried, have no insurance, and live in a disadvantaged neighborhood in terms of income, education, and vehicle availability. Conclusion Our study suggests that addressing disparities in health insurance through policy interventions such as Medicaid expansion may reduce disparities in breast cancer stage at diagnosis among women living in disadvantaged neighborhoods.

Keywords: Breast cancer, spatial cluster, Medicaid