Variation in Palliative Care Program Performance for Patients With Metastatic Cancer.

Abstract

While specialist palliative care is associated with improved end-of-life quality metrics for patients with advanced cancer, its effectiveness may differ between hospitals. To examine variation in palliative care program performance on end-of-life care quality metrics. Retrospective cohort study of palliative care programs that participated in the National Palliative Care Registry, 2018-2019. Medicare data for patients age ≥65 who died with metastatic cancer were aggregated on a program-level. Variation in program performance on outcomes (use of hospice, hospice enrollment ≥3 days, use of intensive care in the last 30 days of life, and use of chemotherapy in the last 14 days of life) was quantified by risk-standardized outcome rates (RSOR) and adjusted median odds ratios (aMOR). The cohort comprised 235 palliative care programs who delivered care to 33,015 patients. There was substantial variation in use of hospice (median RSOR 65.6%, interquartile range (IQR) 57.5%-74.3%), hospice enrollment ≥3 days (median RSOR 53.6%, IQR 48.6%-58.2%), and use of intensive care (median RSOR 14.1%, IQR 13.1%-15.3%), but not use of chemotherapy (median RSOR 1.5%, IQR 1.4%-1.5%). Variation was greatest for hospice use (aMOR 1.48 [1.39-1.57]), suggesting that patients at programs with high hospice use would be 48% more likely to use hospice than if they received care at programs with low use. We found variation in most end-of-life quality metrics for patients with metastatic cancer. Further work is needed to better understand why variations exist and whether such variations reflect a difference in quality of care.