Abstract A105: Application of national guidelines diminishes racial disparities in end-of-life cancer care

Abstract

Abstract Introduction: There are known racial disparities in cancer mortality among African American (AA) patients with shorter median survival rates compared to white patients. Utilizing evidence based clinical guidelines allows for equitable treatment of all patients and may eliminate disparities. The goal of this study was to determine whether AA patients were referred to hospice earlier in their treatment course at a tertiary academic cancer center that routinely utilizes nationally recognized treatment guidelines (i.e. – National Comprehensive Cancer Center Network, NCCN). Methods: We retrospectively analyzed 315 charts of patients referred for hospice during 2008–2009. Subjects identified as either AA or Caucasian were analyzed by race, age and socioeconomic category. Subjects residing in geographic neighborhoods corresponding to postal zip codes with median incomes of either > or ≤ $50,000 were used as a surrogate for socioeconomic status. Charts were reviewed to determine how many lines of chemotherapy were administered before hospice referral. Patients were excluded if they were referred to hospice without receiving chemotherapy. Age, number of chemotherapy regimens and socioeconomic status were compared using Wilcoxon rank-sum test within each race. The overall number of lines of chemotherapy stratified by race was compared using Van Elteren's Test. All categorical variables were summarized by year and race and compared using Fisher's exact test. Results: Patient characteristics and outcomes are as listed in the table below. There was no significant difference between the average number of lines of chemotherapy administered to AA or Caucasian patients, respectively (2.92 vs 2.62; p=0.026), and no racial difference in the percentage of patients referred for hospice discussion prior to hospice transfer [AA (92.2%) Caucasian (83.3%); p=0.36). Similarly, adherence to NCCN treatment guidelines was similar between races [AA (89.1%) Caucasian (93.1%); p=0.36]. AA patients were more likely to reside in a zip code with a lower median income (p<0.001). Interestingly, lower income Caucasian patients were statistically found to have been treated with more lines of chemotherapy prior to hospice, (2.81 compared to 2.39; p=0.04). Conclusions: In this retrospective single institution analysis, there was no evidence that AA patients or patients from lower median incomes were referred to hospice earlier than white patients or patients with higher median incomes. These results are encouraging and may be partially attributable to the use of evidence based NCCN treatment guidelines, suggesting that applying such guidelines to all patients help to eliminate racial bias and disparities in cancer care. Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):A105.