Pay for Performance: Does It Backfire for Doctors and Hospitals?
Health Letter, January 2013
This article summarizes, with permission, Woolhandler and Ariely, Will Pay For Performance Backfire? Insights from Behavioral Economics, Health Affairs Blog, Oct. 11, 2012, www.healthaffairs.org. Copyright © 2012 Health Affairs by Project HOPE — The People-to-People Health Foundation, Inc.
One of the current trends in U.S. medicine is the concept of pay for performance, or P4P. At first glance, providing financial incentives to doctors or hospitals to promote better care might seem reasonable and appealing. However, there is a notable concern making problematic its popularity: Those who have researched this market-based concept have not been able to find evidence that it actually benefits patients.
A new field of inquiry called behavioral economics has provided a framework for challenging the traditional economic view that financial reward always motivates human behavior — or even that it successfully complements inner motivators, such as helping others or performing a task for its own sake. On the contrary: Research has shown that financial rewards can undermine motivation, thereby worsening performance on intrinsically rewarding work. In other words, P4P may backfire.
Evidence concerning P4P effectiveness in health care
In 2004, a U.K. team began a large experiment evaluating the effects of P4P incentive structures on health outcomes for patients with hypertension. The study examined data from 470,725 British patients with hypertension diagnosed between January 2000 and August 2007, including health data from before and after the incentive programs were initiated. The authors of the study, published in the British Medical Journal in 2011, concluded:
Good quality of care for hypertension was stable or improving before pay for performance was introduced. Pay for performance had no discernible effects on processes of care or on hypertension related clinical outcomes. Generous financial incentives, as designed in the UK pay for performance policy, may not be sufficient to improve quality of care and outcomes for hypertension and other common chronic conditions.
Similarly, The Cochrane Collaboration, which systematically reviews multiple individual studies on a given health care topic to issue conclusions, recently published two reports on the topic of P4P. The first found that “financial incentives may be effective in changing health care professional practice” but found “no evidence that financial incentives can improve patient outcomes.” A second report, reviewing P4P programs in primary care, found “insufficient evidence to support or not support the use of financial incentives.”
Despite this evidence-based skepticism as to whether P4P is effective in improving patient outcomes, both public insurers such as Medicare and some major private insurers seem entranced with this market-based solution and are moving ahead with programs aimed at care in both hospital and outpatient settings. It seems quite difficult for most people deeply involved in P4P ideology to acknowledge that it might just not work in health care.
RCTs in education also point to ineffectiveness
Though the longitudinal studies of the effects of P4P on health care are useful, none of them use the gold-standard methodology of the randomized controlled trial, or RCT. In an RCT, a group of people would be randomly assigned to receive medical care delivered by professionals receiving financial incentives and would be compared to a similarly random group getting care from doctors not offered these incentives. No such trial has yet been done in the health care setting. However, two randomized studies have measured the impact of P4P programs on performance in another field:
One RCT involved 20,000 New York City teachers at 200 “high-needs” schools. The teachers in some schools received financial rewards for higher student outcomes — including higher test scores, graduation rates and attendance levels — and the teachers in other schools did not. Teachers were offered as much as $3,000 to incentivize improved work. In the end, the study concluded that these incentives “did not increase student achievement in any meaningful way. If anything, student achievement declined.”
Similarly, an RCT based in Tennessee failed to show a rise in standardized test scores as a result of P4P programs for those schools in which middle school mathematics teachers were given performance-related bonuses of up to $15,000.
In these studies, monetary bonuses not only failed to improve a quantifiable measure of teacher work quality — student performance — in one study, they actually decreased the quality of that same measure. How might this be explained?
A behavioral economics approach to performance and reward
In articles in Health Affairs Blog (Oct. 11, 2012) and the British Medical Journal (Aug. 13, 2012), Public Citizen’s colleagues Dr. David Himmelstein and Dr. Steffi Woolhandler, along with psychologist Dr. Dan Ariely, examine P4P through a lens of behavioral economics, a new field that studies the effects of psychology on economic decisionmaking.
The authors point out that in the past, the previously separate disciplines of economics and psychology adhered to what could serve as an Eleventh Commandment: People Respond to Rewards. In fact, the authors do not disagree: For straightforward manual tasks, they say, performance pay may in fact increase output.
But according to evidence from experts in social psychology and behavioral economics, when it comes to more complex, cognitive tasks — which would characterize the work of physicians and other health care professionals — rewards can actually undermine motivation and worsen performance. This is especially true when motivation is already high to begin with.
One theory that helps explain this phenomenon is called “crowding out.” This refers to the concept that intrinsic motivators (personal, nonfinancially based reasons for acting, such as altruism or the desire for mastery) may be crowded out or obscured by extrinsic — read: monetary — motivators.
Examples of this concept abound. An RCT was conducted among “frequent (presumably highly motivated)” blood donors. In this study, a payment of about $55 to incentivize blood donation proved to actually decrease the number of donations. Similarly, a Swiss study examining the work of volunteers found that those who went unpaid for their efforts worked an average of four more hours each month compared to “volunteers” who received a small compensation for the same work.
An exhaustive review of 128 studies similar to the two above points to a consistent set of conclusive findings on the notion of crowding out, including:
- Tangible rewards, especially in financial form, can largely undercut motivation for those tasks that are intrinsically rewarding in their own right. (In contrast, “symbolic” rewards, such as praise, may serve to enhance intrinsic motivation.)
- In the case of complex, cognitive tasks, financial rewards have the most negative impact.
- Financial incentives can have the most damaging effects when the monetary figures are large or when they are perceived as “controlling,” as in “associated with surveillance, deadlines or threats.”
From reading these studies, it is easily concluded that despite the headlong rush to implement P4P in our increasingly market-driven health care system, there are far too many doubts about the ratio of risks to benefits to accord it the “ready for prime time” status it now enjoys. As in the approval process for new drugs, any proposed treatment addressing the ills of the health care system in general, such as P4P, must undergo a rigorous and important risk-benefit analysis before subjecting millions of people to a new approach that could do more harm than good.