Provider Profiling for Quality of End-of-Life Care
*Sebastien Haneuse, Harvard T.H. Chan School of Public Health
Keywords: Profiling; Readmission; Death; Medicare Data
To monitor quality of care in the U.S., the Centers for Medicare and Medicaid Services (CMS) reports, among other measures, hospital-specific 30-day readmission rates. The focus of these efforts is on health conditions with low mortality, including pneumonia and heart failure. Expanding these efforts to include a broad range of increasingly prevalent ‘advanced’ health conditions, such as Alzheimer’s disease and cancer, for which the force of mortality is relatively strong is problematic because the current CMS approach ignores death as a truncating event. In this talk, we present an alternative framework for profiling of providers and quality of care assessments on the basis of outcomes that are subject to a strong force of mortality. The approach builds on a semi-competing risks framework that is extended to the clustered data setting. A novel hierarchical model is proposed along with an efficient algorithm for estimating model components, including provider-specific effects central to provider profiling. The work is motivated by and applied to a study of end-of-life care for patients diagnosed with pancreatic cancer using data from Medicare Part A.