A Pay-for-Performance Program That May Have Worked

We have a lot of fun mocking the poor research results of various health care quality initiatives.  But sometimes there is research that suggests that a program actually had good results and we enjoy reporting those as well.  An article in the New England Journal of Medicine describes a pay-for-performance program for hospitals in a part of England.      (NEJM Article)     The primary outcome measure was 30 day in-hospital mortality for pneumonia, heart attack or heart failure among patients in the 24 hospitals in the pay-for-performance programs.  The design was a look at 18 months before the program started and 18 months after.  The program, called Advancing Quality, was similar to the Hospital Quality Incentive Demonstration run by CMS, which did not show a significant effect on mortality or other outcomes.  There were 28 quality measures and hospitals in the top quartile of performers could receive a bonus of up to 4% of revenue, which is larger than the incentives used in most other pay-for-performance programs.  Hospitals could also get bonuses if they improved from year-to-year and did better than median performance in the prior year.  The hospitals created an extensive infrastructure to support the program.

During the study period, mortality for the subject conditions decreased everywhere in England, but decreased more in the area where the initiative was introduced.  Across all conditions, there was a significant reduction in mortality, at a relative rate of 6%, or 890 fewer deaths.  The largest reduction was for pneumonia.  While this is positive, the cost both in additional payments and in the infrastructure to support the program was quite large and you have to wonder in this time of spending concerns how many health systems could afford the effort.