Abstract
Assessing and reporting hospital performance has become a worldwide phenomenon. How performance is measured and ranked may determine results. This study evaluates the implications of alternative methods of aggregating hospital performance measures. Data are from patient abstracts and financial Medicare reports from U.S. acute hospitals in 2009, as cleaned and prepared by the Thomson-Reuters 100 Top Hospitals initiative. We evaluate the stability of individual hospital rankings under rank averaging and Data Envelopment Analysis (DEA). Rankings of individual hospitals can shift across four quartiles with inclusion or exclusion of different performance dimensions or when using rank averaging versus DEA. Comparing DEA models with rank averaging, only one third of cases remain in the same quartile, and one third move two or more quartiles. Two proposed DEA models are more similar. Selection among commonly used variables substantially affects the results in ranking hospitals and the method of ranking (DEA vs. r...
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
