Online Program Home
My Program

Abstract Details

Activity Number: 583 - Statistical Methods in Health Services and Performance Profiling
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 2:00 PM to 3:50 PM
Sponsor: Health Policy Statistics Section
Abstract #328556
Title: Hospital Profiling for Quality of End-Of-Life Care via Semi-Competing Risks Analysis
Author(s): Kyu Ha Lee* and Sebastien Haneuse
Companies: The Forsyth Institute and Harvard T.H. Chan School of Public Health
Keywords: Bayesian decision theory; hierarchical modeling; quality of care; semi-competing risks

Although not without controversy, readmission is entrenched as a hospital quality metric. To-date, statistical analyses for hospital profiling based on readmission hinge on fitting a logistic-Normal generalized linear mixed model. In doing so, however, death as a competing risk is ignored. For clinical conditions with a strong force of mortality, such as a diagnosis of pancreatic cancer, ignoring death can have profound effects. In this work, we build on a Bayesian semi-competing risks analysis framework to propose and develop novel multivariate hospital-level performance measures that jointly accommodate readmission and mortality. We also consider a series of hospital profiling-related goals, including the identification of extreme performers and the bivariate classification of hospitals according to whether they have higher-than-expected or lower-than-expected readmission and mortality rates. Towards achieving these goals, we develop a Bayesian decision-theoretic approach that characterizes hospitals on the basis of minimizing the posterior expected loss for an appropriate loss function.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program