ASPO Abstracts
Neighborhood Characteristics and Breast Cancer Mortality in an Urban Setting: Multilevel Analysis
Category: Cancer Health Disparities
Conference Year: 2019
Abstract Body:
Background: Although efforts have been made to decrease breast cancer disparities, recent changes in breast cancer screening recommendations can influence racial disparities of breast cancer in U.S. cities with the largest African American populations. The goal of thisstudy was to examine the relationships between race and neighborhood characteristics, respectively, on breast cancer death among women diagnosed with breast cancer in a multiethnic, urban population. Methods: We obtained PA cancer registry data describing female breast cancer patients residing in Philadelphia County (2005-2014). Patients were geocoded to census tract of residence at time of diagnosis. Census tract population and socioeconomic measures were obtained from the US Census American Community Survey. Neighborhood characteristics included %Black, %unemployed, median household income, %families income below poverty, and Gini index. The percent Black was grouped by three (low, middle, and high) based ontertile and below poverty was categorized into two (<30%, ≥30). Patient characteristics included race/ethnicity (Whites, Blacks, Hispanics, and Asians) and age at diagnosis (<50, ≥50). We used multilevel statistical models using Stata. Results: Our sample of the 4 majorrace/ethnic groups included 10,801 women: Whites (47%), Blacks (46%), Hispanics (4%) and Asians (3%). In multivariate logistic analyses, living in a high percent Black and high povertyneighborhood was related to higher odds of breast cancer death. Patients who lived in neighborhoods with high percent Black residents had higher breast cancer mortality than those in low (aOR=1.26, 95% CI: 1.02-1.55). Patients who lived in neighborhoods with more than 30% of residents below poverty had higher mortality than those in neighborhoods with less than 30% below poverty (aOR=1.25, 95% CI: 1.08-1.45). At the individual level, being older and beingBlack was associated with higher odds of mortality. Conclusion: We observed that individual (age,race) and neighborhood characteristics (%poverty, %Black) related to breast cancer mortality. Future studies are needed to examine how individual socioeconomic and geographic factors interact to predict breast cancer incidence and mortality among diverse populations.
Keywords: Neighborhood characteristics;breast cancer mortality; multilevel analysis;race