Edward Norton

Edward C. Norton, PhD, is an Economist and Professor in the School of Public Health at the University of Michigan.

Pay-for-performance programs aim to reward hospitals that achieve high quality at low cost.  Yet designing the incentives to meet those goals is challenging.

Consider the dilemma that the Centers for Medicare and Medicaid Services (CMS) faced when they added a performance metric for episode payment to a program that previously had included only quality metrics.  Instead of rewarding only hospitals with both high quality and low cost, they gave bonus payments to some hospitals whose low quality was outweighed by low costs.

In short, CMS paid a bonus to some low-quality hospitals after the addition of a spending metric to Medicare’s Hospital Value-Based Purchasing (HVBP) program.  Several MVC Team Members just published a new study showing this unintended consequence.

Here are the main results of that national study.  In fiscal year 2015, CMS expanded its Hospital Value-Based Purchasing program by rewarding or penalizing hospitals for their performance on both spending and quality.  This represented a sharp departure from the program’s original efforts to incentivize hospitals for quality alone. How this change redistributed hospital bonuses and penalties was unknown.  Using data from 2,679 US hospitals that participated in the program in fiscal years 2014 and 2015, we found that the new emphasis on spending rewarded not only low-spending hospitals but some low-quality hospitals as well. Thirty-eight percent of low-spending hospitals received bonuses in fiscal year 2014, compared to 100 percent in fiscal year 2015. However, low-quality hospitals also began to receive bonuses (0 percent in fiscal year 2014 compared to 17 percent in 2015).  All high-quality hospitals received bonuses in both years.

What does this mean for MVC and the BCBSM pay-for-performance program?  The simple answer is that BCBSM only measures episode payments, not other quality measures.  Obviously BCBSM also cares deeply about quality of care, but this P4P program is focused on only one measure, thus avoiding the problem that CMS found.  CMS is now considering a minimum quality threshold to avoid rewarding low-quality, low-spending hospitals.

The MVC team hopes to use its experience studying the national P4P programs to design and improve similar P4P programs in Michigan.


Questions or comments?  We’d love to hear from you.  You can comment on this article, or fill out the contact form at the bottom of the page.