Racial and ethnic disparities in quality of diagnostic evaluation for uterine cancer among Medicare patients.

Abstract

5600 Background: Uterine cancer is the most common gynecologic malignancy. As there is no routine screening for uterine cancer, appropriate evaluation of presenting symptoms is crucial for early diagnosis. We examined racial/ethnic disparities in the quality of diagnostic evaluation received by Medicare patients with uterine cancer. Methods: Using the Surveillance, Epidemiology and End Results (SEER)-Medicare database, we identified patients ≥65 years with uterine cancer diagnosed in 2008-2017 who presented with abnormal uterine bleeding (AUB, the most common symptom of uterine cancer). Patients with other cancer or unknown race/ethnicity were excluded. Race/ethnicity was categorized as White, Black, Hispanic, or Other (including American Indian/Alaska Native and Asian/Pacific Islander). Quality of diagnostic evaluation was measured by: 1) not receiving guideline-recommended diagnostic procedures (yes/no), and 2) number of days between AUB presentation and first diagnostic procedure. We compared these measures across racial/ethnic groups using multivariable regressions adjusting for patient sociodemographic characteristics, comorbid conditions, histology, and tumor grade. We further examined racial/ethnic differences in stage at diagnosis before vs. after accounting for these quality indicators. Results: The sample included 23,017 patients. Compared to White patients, Black, Hispanic, and Other patients were younger; more likely to reside in a metropolitan area, lack preventive care in the past year, or have comorbidities; and tended to have uterine cancer of non-endometrioid types and higher grade (p<0.001 for all). A larger proportion of Black than White patients did not receive guideline-recommended diagnostic procedures (Black: 9.6% vs. White: 5.0%, p<0.001; adjusted odds ratio [aOR] = 1.45, 95% CI: 1.20-1.74). Time from AUB presentation to first diagnostic procedure was longer for Black than White patients (90th percentile: 34 vs. 21 days, p<0.001; adjusted mean ratio = 1.38, 95% CI: 1.21-1.57). Black patients were more likely than White patients to be diagnosed at regional (aOR = 1.18, 95% CI: 1.04-1.33) or distant stage (aOR = 1.27, 95% CI: 1.06-1.52), rather than localized stage, even after adjusting for other characteristics. Further adjusting for quality of diagnostic evaluation slightly reduced Black-White differences in stage at diagnosis (regional: aOR = 1.16, 95% CI: 1.03-1.32; distant: aOR = 1.23, 95% CI: 1.03-1.48). Hispanic and Other patients also experienced delay in time to first diagnostic procedure, but they did not differ from White patients in receipt of guideline-recommended diagnostic procedures or stage at diagnosis. Conclusions: There were racial/ethnic disparities in the quality of diagnostic evaluation received by Medicare patients with uterine cancer, which may partially explain Black-White differences in their stage at diagnosis.