Healthcare Access Domains Mediate Racial Disparities in Ovarian Cancer Treatment Quality in a US Patient Cohort: A Structural Equation Modelling Analysis

• Type of Material: Article
• Authors: Tomi Akinyemiju, Duke University; Quan Chen, University of Kentucky; Lauren E Wilson, Duke University; Rebecca A Previs, Duke University; Ashwini Joshi, Duke University; Margaret Liang, University of Alabama Birmingham; Maria Pisu, University of Alabama Birmingham; Kevin C Ward, Emory University; Andrew Berchuck, Duke University; Maria J Schymura, New York State Cancer Registry, Albany NY; Bin Huang, University of Kentucky
• Language: English
• Date: 2023-01-01
• Publisher: AMER ASSOC CANCER RESEARCH
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © 2023 by the American Association for Cancer Research
• Final Published Version (URL): http://dx.doi.org/10.1158/1055-9965.EPI-22-0650
• Title of Journal or Parent Work: CANCER EPIDEMIOLOGY BIOMARKERS & PREVENTION
• Volume: 32
• Issue: 1
• Start Page: 74
• End Page: 81
• Grant/Funding Information: This research was funded by the National Institutes of Health/National Cancer Institute (T. Akinyemiju, R37CA233777). R.A. Previs is supported by a grant from the NIH K12 HD103083
• Keywords: STAGE; Oncology; SURVIVAL; OUTCOMES; AVAILABILITY; RACIAL/ETHNIC DISPARITIES; Science & Technology; TRENDS; Life Sciences & Biomedicine; TREATMENT GUIDELINES; IMPACT; RESOURCES; Public, Environmental & Occupational Health; ADHERENCE
• Research Categories: Health Sciences, Oncology; Biology, Biostatistics; Health Sciences, Medicine and Surgery; Health Sciences, Obstetrics and Gynecology

Abstract

Background: Ovarian cancer survival disparities have persisted for decades, driven by lack of access to quality treatment. We conducted structural equation modeling (SEM) to define latent variables representing three healthcare access (HCA) domains: affordability, availability, and accessibility, and evaluated the direct and indirect associations between race and ovarian cancer treatment mediated through the HCA domains. Methods: Patients with ovarian cancer ages 65 years or older diagnosed between 2008 and 2015 were identified from the SEER-Medicare dataset. Generalized SEM was used to estimate latent variables representing HCA domains by race in relation to two measures of ovarian cancer-treatment quality: gynecologic oncology consultation and receipt of any ovarian cancer surgery. Results: A total of 8,987 patients with ovarian cancer were included in the analysis; 7% were Black. The affordability [Ω: 0.876; average variance extracted (AVE) = 0.689], availability (Ω: 0.848; AVE = 0.636), and accessibility (Ω: 0.798; AVE = 0.634) latent variables showed high composite reliability in SEM analysis. Black patients had lower affordability and availability, but higher accessibility compared with non-Black patients. In fully adjusted models, there was no direct effect observed between Black race to receipt of surgery [β: -0.044; 95% confidence interval (CI), -0.264 to 0.149]; however, there was an inverse total effect (β: -0.243; 95% CI, -0.079 to -0.011) that was driven by HCA affordability (β: -0.025; 95% CI, -0.036 to -0.013), as well as pathways that included availability and consultation with a gynecologist oncologist. Conclusions: Racial differences in ovarian cancer treatment appear to be driven by latent variables representing healthcare affordability, availability, and accessibility. Impact: Strategies to mitigate disparities in multiple HCA domains will be transformative in advancing equity in cancer treatment.