Pay-for-performance under scrutiny

Years ago, many health insurance companies were excited at the prospect of pay-for-performance programs. The premise was that providers often did not act in the best interests of their patients because they weren't paid to and were too busy doing lower-value work. By paying doctors for performance (quality), then doctors would have an incentive to focus on the high-value work, which often centered around prevention. The idea was that everyone would then experience a win: providers would be paid more for doing quality work, patients would have better health outcomes, and insurance companies would overall pay less because emergency operations would be averted. That was the theory.

STAT published an opinion piece that compiled a number of studies showing that the confidence in these programs might be misplaced. It appears that objectors were right about at least one unfortunate side effect: some providers appear to have denied care to individuals because those individuals were sick enough that providers thought they would effectively be penalized for treating them. Some of the objections appear addressable by a different risk-adjustment algorithm; one can imagine, for example, a risk-adjustment algorithm under which some providers would actually seek out the sickest and poorest patients. Risk-adjustment has been frequently debated, and coming to a consensus would not be easy. Some of the other objections also seem difficult to address, such as causing some providers to game the system by incorrectly characterizing infections. Although the list of studies were presented as negatives against pay-for-performance, a number of them appear to qualify their conclusions, citing methodological limitations or mixed results. The overall effect of pay-for-performance programs is undoubtedly a difficult one to fully understand.