“Personalized Health Care Outcome Analysis of Cardiovascular Surgical Procedures: An Instrumental Variable Tree Approach” by Mr. Guihua Wang
PhD Candidate in Technology and Operations
Stephen M. Ross School of Business
University of Michigan
This study addresses the challenges of generating patient-centric information about hospital quality and analyzes the impact of information on enabling patients to receive better care. Methodologically, we develop a new Instrumental Variable (IV) tree approach by incorporating an IV into a tree-based method to correct for potential endogeneity issues in heterogeneous treatment effect analysis using observational data. Empirically, we designate hospitals as different treatments and apply the IV tree to study the outcome differences between thirty-five New York hospitals for cardiovascular surgeries. We find that the outcome differences between hospitals are heterogeneous across different patients. By comparing scenarios with patient-centric and population-average information, we show that 803 of patients can benefit from using patient-centric information and their complications can be reduced by 67.43. We also illustrate how patient-centric information can enhance pay-for-performance programs offered by payers and guide hospitals in targeting quality improvement efforts.