Comparison of end of life care practices and resource utilization by ovarian cancer patients of varying racial, socioeconomic and geographic backgrounds in Texas

The purpose of this study was to assess differences in utilization of palliative care resources and achievement of end of life quality care goals among patients with ovarian cancer in Texas and examine the racial-ethnic, socio-economic and geographic disparities in these outcomes. ^ Using Texas Cancer Registry database which had been previously linked to Medicare payer systems, we selected ovarian cancer patients who died from 2000 to 2012, had at least 30 days from diagnosis to death and at least 13 months of continuous Medicare coverage before death. Descriptive statistics and multivariate logistic regressions were conducted to evaluate predictors of end of life quality care, using metrics from the National Quality Forum and American Society of Clinical Oncology Quality of Practice Initiative.These metrics included avoidance of all of the following: chemotherapy in the last 14 days of life, an ICU admission in the final 30 days of life, more than one emergency room (ER) or hospital admission in the last 30 days of life, an invasive procedure and a life-extending procedure. Additional quality of end of life care metrics included enrollment in (and remaining enrolled in) hospice prior to death, enrollment in hospice prior to the last three days of life and enrollment in hospice as an outpatient (instead of as an inpatient). ^ A total of 4,003 patients were included with 3,009 (75%) white, 637 (16%) Hispanic, 307 (8%) black and 50 (1%) other. Median age at death was 76. Eighty-five percent had an advanced stage of ovarian cancer at diagnosis. Seventy-nine percent lived in urban settings. Two-thousand eight hundred and sixty-nine (72%) women enrolled in Hospice prior to death but only 2,543 (64%) died while enrolled in hospice. The median time a patient was enrolled in hospice prior to death was 19 days. In the final 30 days of life, 427 (11%) had more than one ER visit, 554 (14%) more than one hospital admission, 647 (16%) ICU admission, 789 (20%) invasive care and 454 (11%) life-extending care. In the final 14 days of life, 389 (10%) received chemotherapy. After controlling for relevant covariates, several outcomes differed by race. Hispanic (OR 0.77 [0.62-0.97]; p=0.04) and black (OR 0.73 [0.55-0.98]; p=0.02) patients were less likely to enroll in hospice and die in hospice (Hispanic (OR 0.77 [0.63-0.94]; p=0.01); black (OR 0.65 [0.50-0.85]; p=0.001)). Hispanic (OR 1.39 [1.08-1.78]; p=0.009) and black (OR 1.40 [1.02-1.92]; p=0.03) patients were more likely to go to the ICU and have \u3e1ER visit (Hispanic (OR 1.42 [1.06-1.92]; p=0.02); black (OR 2.27 [1.63-3.22]; p\u3c.001)). Black patients (OR 1.34 [1.003-1.79]; p=0.04) were more likely to have invasive care. Hispanic (OR 1.37 [1.02-1.84]; p=0.03) and black (OR 2.12 [OR 1.52-2.95]; p\u3c0.001) patients were more likely to receive life-extending care. ^ Disparities also existed between women living in census tracts with higher versus lower percentages of poverty and women living in rural versus urban locations. The highest percentage of poverty in a census tract was associated with decreased odds of ICU admission (OR 0.66 [0.46-0.93]; p=0.02] and decreased odds of receiving an invasive procedure in the final 30 days of life (OR 0.73 [0.57-0.95]; p=0.02) compared to the census tract with the lowest percentage of poverty. Women living in rural areas had decreased odds of having ICU admission (Rural OR 0.78 [0.61-0.998]; p=0.04) and of invasive procedure (Rural OR 0.80 [0.65-0.99]); p=0.04) compared to women living in urban areas. ^ In conclusion, risk factors for receiving intensive and invasive care were belonging to a minority race, higher socioeconomic status or urban geographic area. This identifies which groups are more at-risk to experience disparities in end of life care.