Impact of race and payer status on the choice of urinary diversion among patients with localized bladder carcinoma undergoing cystectomy.

455 Background: Prior studies have described significant disparities in the selection of urinary diversion (UD) in patients with localized bladder cancer undergoing cystectomy. Although the choice of UD has not been shown to impact oncologic outcomes, continent urinary diversions (CUD) are associated with lower rates of in-hospital complications and mortality, but higher costs (Farber NJ et al. Bladder Cancer 2018). Male gender, White race, and higher income have been associated with proportionally higher rates of CUD than patients from other races or those without private insurance (Barocas DA et al. Cancer 2014 and Rios EM et al. Urology 2020). Utilizing the California Office of Statewide Health Planning and Development (OSHPD) database, we investigated potential barriers to CUD in patients with bladder cancer undergoing radical cystectomy. Methods: The current procedural terminology (CPT) and the international classification of diseases (ICD)-9/10 codes were used to identify patients with bladder cancer undergoing radical cystectomy from Jan 1, 2012, through Dec 31, 2018. Type of UD and demographic data such as race and payer status were collected. Univariate and multivariable analyses were conducted to determine the association between demographic variables and CUD use. Results: In total, 9,342 patients who underwent radical cystectomy were identified, of which 3,061 had UD status noted. Of these, 13.2% (404/3061) were continent and 86.2% (2,657/3061) were incontinent diversions. CUD use was significantly higher in White patients (14.1%; 320/2276) compared to Asian (12.8%; 24/187), Hispanic (9.5%; 30/316) or Black (5%; 6/119) patients (P=0.01). Use of CUD was significantly higher in patients with private insurance (23.2%; 167/721) compared to those with Medicare (10.2%; 207/2023) or indigent (MediCal/Medicaid; 8.6%; 23/269; p<0.001) coverage. On multivariable analysis adjusting for comorbidities and care setting, Black (OR: 0.30, 0.13-0.69) and Hispanic (OR: 0.57, 0.38-0.86) race were associated with a lower probability of getting a CUD, while male patients (OR 1.88, 1.31-2.71) and those receiving care at academic centers (OR 3.10, 2.38-4.05) were more likely to receive a CUD. Payer status did not show a significant difference between the two procedures. Finally, the presence of chronic kidney disease represented a risk factor for not getting a CUD (OR: 0.61, 0.43-0.85), but not the presence of diabetes and frailty. Conclusions: Black or Hispanic race and female gender were associated with lower rates of CUD when controlling for other factors. We hypothesize that the higher costs for CUD, communication barriers, especially with non-English speakers, comorbidities, and a potential lack of cultural humility could lead to an unconscious bias from the healthcare team. Further research aimed at understanding and addressing these disparities is needed.