RESUMO
BACKGROUND: Palliative care has been associated with improved survival and quality of life, with lower rate of end-of-life health care utilization and cost. We examined trends in palliative care utilization in older pancreatic cancer patients. METHODS: Pancreatic cancer patients with and without palliative care consults were identified using the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database between 2000 and 2009. Trend of palliative care use was studied. Emergency room/intensive care unit (ICU) utilization and costs in the last 30 days of life were compared between both groups using propensity score-matched (PSM) analysis. RESULTS: Of the 54,130 patients, 3,166 (5.8%) received palliative care and 70% received it in the last 30 days of life. The proportion of patients receiving palliative care increased from 1.4% in 2000 to 7.4% in 2009 (P<0.001). Patients with palliative care were more likely to be older, Asian and women. In the unmatched and PSM population, the average visits to the ER in the last 30 days of life were significantly higher for patients who received palliative care, and had a significantly higher cost of care. Similarly, ICU length of stay in the last 30 days of life was higher in patients who did not receive palliative care in both PSM and unmatched patients. Cost of care and number of ICU admissions were not different between palliative and non-palliative care groups in PSM and unmatched patients. CONCLUSIONS: In this study of Medicare patients with pancreatic cancer, palliative care use has increased between 2000 and 2009. Palliative care was largely offered close to the end of life and was not associated with reduced health care utilization or cost.
RESUMO
The purpose of this analysis was to measure the impact of palliative care services on hospital charges in the 5 days prior to death-the most expensive time of a patient's life-and identify hospital service categories and patient financial classes yielding the highest savings from palliative care. The analysis population included UT Southwestern patients admitted to the hospital between October 1, 2013, and September 30, 2016. Palliative care patients were defined as any patient who received at least one completed palliative care order. In order to create an accurate comparison group, a propensity score match was generated to identify patients most likely to have qualified for a palliative care consult. Covariates included in the model were age, sex, race, financial class, and number of comorbidities. Comorbidities were identified using the Elixhauser Comorbidity Index, and all charges were pulled for the 5 days prior to death. Total hospital charges were also reported by hospital service and financial class. Statistical significance was then derived using a gamma distributed log-linked generalized linear model. The final population included in the analysis, post the propensity score match, was composed of mostly white, non-Hispanic males. The majority of the patients had five or fewer comorbidities, and the primary preexisting conditions seen among patients were cardiovascular diseases (36.0%) and cancer (23.4%). The hospital service categories yielding the highest mean savings were pharmacy (mean $2,765; P < .0001) and labs (mean $1,063; P < .0001). Financial classes with the greatest savings were Medicaid and charity/self-pay. Overall, there was a significant difference in charges between those that received a palliative care consult and those that did not. The fact that the highest savings were in pharmacy and laboratory services suggests that unnecessary labs and medications are discontinued in an effort to improve patient care and quality of life while reducing costs during end-of-life care. Palliative care services ease the cost burden of end-of-life services for low-income populations.