Predictive inference for identifying outliers in health care providers
*Michael Joseph Racz, Albany College of Pharmacy and Health Sciences 
Joseph Sedransk, Case Western Reserve University  

Keywords: report cards, posterior predictive probability, residual

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 is essential. Ohlssen, Sharples and Spiegelhalter (2007) present an extensive, 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. We evaluate the Ohlssen et al. (2007) methodology using both NYS bypass surgery data and simulated data of the same type as that used in Racz and Sedransk (2010). We also extend the Ohlssen et al. methodology to permit evaluation of important characteristics seemingly not covered by the Ohlssen et al. procedure.

Ohlssen, D. I., Sharples, L. D., and Spiegelhalter, D. J. (2007), “A Hierarchical Modelling Framework for Identifying Unusual Performance in Health Care Providers,” Journal of the Royal Statistical Society, Ser. A, 170, 865–890.

Racz MJ, Sedransk J, (2010). “Bayesian and frequentist methods for provider profiling using risk-adjusted assessments of medical outcomes,” Journal of the American Statistical Association 105, 48-58