AUSTINBAYESIAN METHODS OF INSTITUTIONAL PROFILINGMETHODOLOGY A Comparison of Bayesian Methods for Profiling Hospital Performance

There is a growing interest in the use of Bayesianmethods for profiling institutional performance. In the literature, several studieshave compareddifferent frequentistmethods for clas-sifying hospitals as performance outliers. The purpose of this studywas to compare4differentBayesianmethods for classi-fying hospitals as outcomes outliers, using 30-day hospital-level mortality rates for a cohort of acute myocardial infarc-tionpatients as a test case. The1st Bayesianmethod involved determining theprobability that ahospital’smortality rate for an average patient exceeded a specified threshold. The 2nd method involved ranking hospitals according to theirmortal-ity rate for an average patient. The 3rd method involved de-termining the probability that a hospital’s standardizedmor-tality ratio exceeded a specified threshold. The 4th method involved ranking hospitals according to their standardized mortality ratio. In most of the scenarios examined, there was only marginal agreement between the different methods. In only 4 of 19 comparisons, was there good agreement between the different methods (0.40 kappa 0.75). Methods based on ranking institutions were relatively insensitive to differ-ences between hospitals. These inconsistencies raise ques-tions about the choice ofmethods for classifyinghospital per-formance, and they suggest a need for urgent research into whichmethods are best able to discriminate between institu-tions and which are most meaningful to decision makers