Inference for identifying outlying health care providers

• We compare and contrast alternative methods for provider profiling.
• We present a method addressing multiple comparisons.
• Our analyses include real and simulated data.

Provider profiling is the evaluation of the performance of hospitals, doctors, and other medical practitioners to enhance the quality of care. Many jurisdictions have released public report cards comparing hospital or physician-specific outcomes. Given the importance of provider profiling, studying the methodology used and providing enhancements are essential. Ohlssen, Sharples and Spiegelhalter (2007) give a thoughtful evaluation of provider profiling methodology. In particular they are concerned about whether a putative outlier is really an outlier or an observation in the tail of the common distribution for all practitioners, and present methodology to address this issue. In this paper we suggest as an alternative the use of a 100(1−α)100(1−α)% credible region for provider fixed effects to identify outliers. Using both New York State bypass surgery data and simulated data of the same type as that used in Racz and Sedransk (2010), we compare the Ohlssen et al. (2007) approach with standard methodology, and with use of the credible region.