BID Seminar: ‘Reputation in Patient Hospital Selection: Empirical Evidence from Australia’

Research Cluster

Speaker:Lorenz Adams , University of Exeter
Date: Tuesday 21 May 2019
Time: 12.30
Location: Streatham Court B

Further details

This paper studies the role of reputation in individual hospital selection. Our analysis focuses on birth episodes in the Melbourne metropolitan area. We examine the decision behaviour of women with private health insurance in selecting one of 18 local hospitals. Firstly, we examine how observed hospital performance and unobserved hospital reputation influence a patients willingness to travel between her home and the chosen hospital. In addition, we estimate the probability of individual hospitals being selected from the set of available alternatives. We use a multinomial probit (MNP) model estimated by a Bayesian method. In this study, we allow for correlations across hospitals, due to similarities across hospitals attributable to geographic locations and other characteristics, including hospital ownership. We find significant hospital specific differences in the willingness to travel, which suggests patients travel further to visit more reputable institutions. This result is complemented by substantial variation in the probability of specific hospitals being chosen. A comprehensive simulation study is conducted, using the estimated results from the MNP model. We examine how a new hypothetical hospital will alter the selection mechanism. By varying the geographic location, the performance of the new hospital and other clinical characteristics, we highlight the differences in its impact on the population’s hospital choice. This exercise provides valuable evidence and guidance to social planners and policy makers.