Statistical Considerations in Assessing Hospital Performance

• Activity Number: 398
• Type: Invited
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Sharon-Lise Normand* and Ann Lovett and Robert Wolf and Treacy Silbaugh and Katya Zelevinsky and Matthew Cioffi and Caroline Wood
• Companies: Harvard Medical School and Harvard Medical School and Harvard Medical School and Harvard Medical School and Harvard Medical School and Harvard Medical School and Harvard Medical School
• Keywords: Bayesian inference ; hierarchical models ; predictive inference ; standardization ; Mass-DAC

Abstract

Assessing the quality of care delivered by hospitals involves observational data that are clustered within hospitals. While most often a single outcome, such as death following a surgical procedure, serves as the unit of observation, increasingly composite endpoints are utilized. In this talk, I discuss the key parameters of interest, the roles of sample size and uncertainty, and choice of shrinkage targets for policy making. Hospital performance of coronary artery bypass surgery and of percutaneous coronary intervention illustrate issues.