Background: There are race-based differences in bladder cancer survival. To better understand this phenomenon, this study was designed to assess the statistical contributions of tumor, treatment, and access variables to race-based differences in survival.
Methods: Data were extracted from the National Cancer Data Base on black and white adults with muscle-invasive bladder cancer from 2004 to 2015. The impact of tumor, access, and treatment variables on differences in survival was inferred by the performance of sequential propensity score-weighted analyses in which black and white patients were balanced with respect to demographics and health status (comorbidities) tumor characteristics, treatment, and access-related variables. The propensity score-weighted hazard of death (black vs white) was calculated after each iteration.
Results: This study identified 44,577 patients with a median follow-up of 77 months. After demographics and health status were balanced, black race was associated with 18% worse mortality (hazard ratio, 1.18; 95% confidence interval [CI], 1.12-1.25; P < .001). Balancing by tumor characteristics reduced this to 16%, balancing by treatment reduced this to 10%, and balancing by access-related variables resulted in no difference. Access-related variables explained 40% (95% CI, 22.9%-57.0%) of the excess risk of death in blacks, whereas treatment factors explained 35% (95% CI, 22.2%-46.9%). The contribution of tumor characteristics was not significant.
Conclusions: In the models, differences in survival for black and white patients with bladder cancer are best explained by disparities in access and treatment, not tumor characteristics. Access to care is likely a key factor in racial disparities in cancer.
Keywords: health care disparities; health services accessibility; observational study; propensity score; urinary bladder neoplasms.
© 2019 American Cancer Society.