Pay-For-Performance Solutions May Lie With Patients, Not Doctors

The problem with pay-for-performance programs may be that small incentives are tied to outcomes for a large population of patients, a Journal of the American Medical Association (JAMA) article suggests. The article’s authors propose, instead, that large incentives be tied to outcomes for smaller populations of patients at risk for negative outcomes. Doing so could make the extra effort needed to qualify for incentive payments more feasible for clinicians and support the government’s aim of rewarding providers for their efforts, rather than rewarding them for outcomes they did not affect.


Pay-for-performance programs reward physicians whose patients achieve desired outcomes. Traditional programs target all patients meeting certain criteria, e.g. patients discharged for congestive heart failure, and seek a specific outcome, e.g., no readmissions within 30 days.  However, according to the article’s authors, Aaron McKethan, Ph.D, of the Gillings School of Global Health at the University of North Carolina at Chapel Hill and Ashish K. Jha, MD, MPH, of the Department of Health Policy and Management at the Harvard School of Public Health, many patients can expect good outcomes regardless of care, possibly due to factors including good social support systems and primary care. 

Targeted Programs

Instead, the authors suggest that prediction models identify patients at risk for bad outcomes or missed clinical goals. Physicians could then aim their extra resources at these patients in a targeted program.  Furthermore, incentives could be increased if applied only to patients in this smaller subset. As an example, a program that might pay up to $50 for each achieved outcome in a population of 20,000 patients could be revamped to pay up to $250 for each achieved outcome in a smaller, targeted population. Doing so would help physicians target their efforts where needed and justify resource investments needed to improve care. The government would benefit because, as the authors note, such incentives would “reward high-quality health care professionals, not health care professionals whose patients are likely to do well irrespective of incentives.”