Racial disparities in receipt of guideline care and cancer deaths for women with ovarian cancer.
e18081 Background: Receipt of guideline care improves survival for women with ovarian cancer yet care may vary by race. We assessed the receipt of guideline care and cancer deaths by race for women with incident ovarian cancer. Methods: This retrospective cohort analysis used the National Cancer Institute’s Patterns of Care data for black and white women diagnosed with all stages of ovarian cancer in 2002 and 2011 (n = 5356) with follow-up through 12/31/14. Data included patient characteristics, type of surgery and chemotherapy, and provider factors. Multivariate logistic regression was used to create the standardized percentages (predictive margins) of women receiving guideline treatment by race group. Cox proportional hazards models were used to assess the unadjusted and adjusted risk of ovarian cancer death by race. Results: At diagnosis, for black women, 35.6% had Stage III and 37.2% had Stage IV disease compared with 44.1% Stage III and 24.2% Stage IV for white women. More black women had surgery in large teaching hospitals (47.6%) than white women (39.7%) but use of gynecologic oncologists (GO) was similar for black women and white women- (62.1% vs 58.8%). In regression models, the standardized percent of black women receiving guideline care was significantly lower than for white women (29.2% vs 38.5%). The unadjusted hazards ratio (HR) for death was significantly higher in black women (HR = 1.32) yet after controlling for patient and provider factors and receipt of guideline care, black women did not have a significantly higher risk of death (HR = 1.08). Conclusions: Rates of guideline care are very low for all women with ovarian cancer, significantly more so for black patients. Low use of guideline care among black women cannot be explained by provider factors as a large percent of black women consulted a GO or received care in large teaching hospitals. Research is needed to address how to increase guideline care among black patients as we found that race was not associated with the risk of death when guideline care was included as a factor in multivariate survival models.