PO42

Abstract

Disparities in race and socioeconomic factors affect patient access to cancer screening, treatment, and clinical outcomes. The aim of this project was to evaluate relationship between race and socioeconomic factors including insurance status, employment status, disability status, and distance from brachytherapy center on clinical outcomes including stage at presentation, number of nodes positive, brachytherapy technique, progression-free survival (PFS), and overall survival (OS). All cervical cancer patients treated with brachytherapy at our institution from 2007-2017 were identified. Race and socioeconomic factors including insurance status, employment status, disability status, and distance from brachytherapy center were recorded. Clinical characteristics including stage at presentation, number of involved nodes, and brachytherapy technique were also recorded. PFS and OS were calculated from date of last brachytherapy fraction, with censorship at date of last follow-up. Correlation was tested between racial and socioeconomic factors and survival outcomes (i.e., PFS and OS) using Cox regression models. Their association with other outcomes was examined with Wilcoxon rank sum tests, Fisher's exact tests, and Spearman's rank correlation coefficients where appropriate. 251 cervical cancer patients were identified, with median follow-up 5.2 years (IQR 2.0-7.7 years). On univariate analysis (UVA), there was no correlation between brachytherapy technique utilized, number of nodes positive, or stage at presentation and race, distance from treatment center, insurance status, employment status, or disability status. UVA did show a significant correlation between PFS and race, insurance status, employment status, and disability status. Significantly worse PFS was seen in non-white group (p=0.036), uninsured group (p<0.001), unemployed group (p<0.001), and disabled group (p=0.041). Similarly, there was significant correlation between OS and race, insurance status, employment status, and disability status. Significantly worse OS was seen in non-white group (p=0.005), uninsured group (p<0.001), unemployed group (p<0.001), and disabled group (p=0.008). On multivariate analysis (MVA), there was no significant correlation between race or disability status and PFS, but there was significantly improved PFS seen in patients with insurance (p < 0.001) and patients who were employed (p = 0.002). MVA showed no correlation between disability status and OS, but significantly worse OS in patients who were non-white (p=0.039) and significantly improved OS in patients with insurance (p<0.001), and patients who were employed (p-0.001). MVA showed no significant correlation between stage and insurance or employment status. MVA showed no significant correlation between histology and employment status; on MVA patients with government insurance were less likely to have squamous histology compared with no insurance (p=0.002). Insurance and employment status are significant predictors of PFS, with significantly improved PFS seen in patients who possess insurance and patients who are employed. White race, insurance, and employment status are significant predictors of OS, with significantly improved OS seen in patients of white race, patients who have insurance, and patients who are employed. There was no correlation between insurance or employment status and stage at presentation.