Research Activities, September 2012: Patient Safety and Quality: A five-point checklist can help public report sponsors avoid misclassifying the performance of health care providers

Research Activities, September 2012: Patient Safety and Quality: A five-point checklist can help public report sponsors avoid misclassifying the performance of health care providers

A five-point checklist can help public report sponsors avoid misclassifying the performance of health care providers

A new study concludes that a five-point checklist can help public report sponsors avoid misclassifying the performance of health care providers. This misclassification can lead to lower quality, greater waste, or patient harm in health care, note RAND researchers. They believe that by publicly explaining how they addressed the five points in the checklist, report sponsors can give performance reports greater methodological transparency and improve the chances that such reports will give rise to better, more efficient care. The checklist highlights key methodological options for report sponsors to consider in creating a report and helps report users decide if they trust the information in the report.

The first point includes measuring and addressing systematic misclassification, such as reporting higher mortality rates for patients of providers treating an older population than for patients of providers treating a younger population. Checklist point two focuses on measuring and addressing random misclassification, which can occur when there is little provider-to-provider variation in true performance (e.g., if it is uniformly high) or when there is high measurement error. The third checklist point is to use composite scores appropriately, for example, clarifying that the provider with a high composite rating could have a great spread among the individual scores.
Checklist point four is to perform sensitivity analyses to ensure that choosing alternate methods would not create large changes in the performance scores. The fifth checklist point is to measure the effects of reporting to verify that the goal of the public report is being met, whether to help patients choose better providers or motivate providers to make quality improvements.
This study was funded by the Agency for Healthcare Research and Quality (Contract No. 290-07-10022). More details are in "A five-point checklist to help performance reports incentivize improvement and effectively guide patients," by Mark W. Friedberg, M.D., M.P.P., and Cheryl L. Damberg, Ph.D., M.P.H., in the March 2012 Health Affairs 31(3), pp.612-618.

— DIL

Editor's Note: Drs. Friedberg and Damberg were funded by AHRQ to develop a more detailed User's Guide on this topic, which is available at http://www.ahrq.gov/qual/value/perfscoresmethods.