A reader writes:
You’ve spoken about the government’s failed attempts to pay for performance, and in this last post, concluded that its failure was largely due to selecting bad metrics that do not accurately capture quality. There is at least one program within the government that has been measuring and monitoring quality for nearly 20 years. The Department of Veterans Affairs National Surgical Quality Improvement Program has been collecting and analyzing data on non-cardiac surgery since 1994. As a result of their efforts, the VA has been able to reduce M&M related to surgery. There are a few studies out there that highlight the VA’s success.
However, the VA’s model has not been widely adopted in any other format. I just wanted to highlight one successful model where the government learned what metrics to collect and how to analyze them in the appropriate context.
Let me be explicit. I’m not saying that quality improvement is a bad thing, or that quality initiatives can’t work. I’m saying the way we usually handle “pay for performance”, by picking a few easy-to-measure metrics and then assuming that translates into improved quality, doesn’t work as well as we’d like.
The VA has made enormous strides in the last few decades in working on quality. Almost none of it would be described as “pay for performance”.