Abstract C14: Decomposing Racial Disparities in Breast Reconstruction Surgery

Abstract

Abstract Background: Breast Reconstruction Surgery (BRS) is an important component of overall breast cancer treatment and is known to improve psychosocial and quality of life outcomes for patients who undergo mastectomy. Clinical guidelines recommend that BRS should be offered to all women who undergo mastectomy. However, significant racial disparities in receipt of BRS have been observed in literature. African Americans are known to be less likely to receive BRS as compared to Non-Hispanic Whites (NHWs) even after controlling for clinical characteristics. Not much is known about the factors that contribute to this disparity. Racial differences in barriers to receipt of BRS may explain the disparity. These barriers are grouped into patient, provider and area level factors. Objective: To determine the degree to which racial disparity in the receipt of BRS can be explained by observed factors at the patient, hospital and area level using Blinder-Oaxaca decomposition method. Data: State inpatient databases (SIDs) for 2010-2012 from seven diverse states were used to identify patients with breast cancer. Information on hospital-level characteristics was derived from the American Hospital Association (AHA) data and the area-level data was derived from the Area Health Resource File (AHRF) and the University of Michigan Racial Residential Segregation Measurement Project. Analyses: The primary outcome was receipt of BRS with mastectomy compared to mastectomy alone. The Fairlie modification to the Oaxaca-Blinder decomposition method was used to decompose the difference in likelihood of receiving BRS between Non-Hispanic Whites and Blacks into components that are explained by each observed characteristic. The observed characteristics included in the model are patient-level barriers: lack of insurance and higher comorbidities; hospital-level barriers: hospital-level factors that reduce the likelihood of offering culturally competent and appropriate care (racial mix catered to by the hospital, availability of knowledge and resources, market competition, etc.); and area-level barriers: racial residential segregation, socio-economic status of the area and availability of plastic surgeons. Findings: Racial differences in insurance status can explain 22% of the disparity and racial differences in socio-economic status of the county can explain 13% of the disparity in receipt of BRS between African American and NHW women. Other factors that contribute significantly to the racial disparity in BRS are higher rates of obesity, higher number and severity of comorbidities in African Americans and higher likelihood of seeking care at hospitals that cater to racial minorities. Lower likelihood of seeking care at an NCI designated Cancer Center by African American women also contributes significantly to the racial disparity in receipt of BRS for these women. On the other hand, African American patients are likely to be younger, reside in urban counties with high density of plastic surgeons and seek care at larger hospitals located in regions with high market competition. These factors contribute significantly to reducing disparities in receipt of BRS for African American women. Living in racially segregated neighborhoods reduces disparities in receipt of BRS for African American women. This could be explained by better psychosocial support in segregated neighborhoods. Conclusions: Known barriers to access significantly influence receipt of BRS for African American women. These findings help guide policy makers towards the barriers that have a high and significant impact on disparities. Selectively addressing these barriers may provide a focused approach to mitigating racial disparities in receipt of BRS. Citation Format: Jaya Shankar Khushalani. Decomposing Racial Disparities in Breast Reconstruction Surgery. [abstract]. In: Proceedings of the Ninth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2016 Sep 25-28; Fort Lauderdale, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(2 Suppl):Abstract nr C14.