In a cautionary editorial alongside a related article in today's issue of the British medical journal BMJ, leading experts in health policy and behavioral economics argue that pay-for-performance (P4P) schemes – which financially reward doctors and hospitals for hitting specific, numerical targets in such matters as preventing hospital readmissions or prescribing certain drugs – are likely to do more harm than good.
Such schemes are being adopted as a key component of the Accountable Care Organization strategy mandated by the 2010 health reform and are now part of the Medicare program, Massachusetts' cost-control legislation and virtually all major new private health insurance payment contracts.
Yet the editorial, echoing a theme of the accompanying article, says there's very little evidence that P4P has improved patient survival or any other measure of public health.
"Despite a dearth of robust evidence that P4P is clinically effective in health care, payers charge ahead with implementing everywhere an intervention that has proven to work nowhere," the authors write.
Worse still, there is mounting evidence – reinforced by the latest findings in behavioral economics – that such schemes may actually do harm, the authors say.
For example, doctors and nurses may perceive detailed, overly prescriptive financial P4P contracts as "controlling," which can cause them to dissociate from their work, lose their intrinsic motivation to do their very best for the patient, and engage in gaming – the medical equivalent of "teaching to the test."
According to the authors, such gaming – e.g. "upcoding" a diagnosis to another condition that yields a higher payment – is already rife. For instance, labeling a pneumonia patient's condition as "complex" rather than "simple" can increase the hospital's payment by 42 percent.
"Pay for performance inverts medical priorities, making care an instrument for generating money, rather than vice versa," said Dr. David Himmelstein, professor at the City University of New York's School of Public Health and senior author of the editorial. "It can mutate honesty and altruism into accounting and legal trickery."
Dan Ariely, James B. Duke Professor of Psychology and Behavioral Economics at Duke University, co-author of the editorial and the author of numerous research studies and three bestselling books on behavioral economics, including "The (Honest) Truth about Dishonesty," said: "Several studies show that while performance-based rewards can increase output for straightforward manual tasks, they can undermine motivation and actually worsen performance on complex cognitive tasks, such as those required in medicine. The unintended consequence is likely a worsening of care, not its improvement."
Lead author Dr. Steffie Woolhandler, also a professor at CUNY's School of Public Health, noted another hazard: Some physicians in safety-net hospitals may score poorly because of circumstances beyond their control, such as their institution's financial distress. "In such situations, penalizing low-scorers can make matters worse, effectively punishing patients who have nowhere else to go," she said.
Himmelstein and Woolhandler are also visiting professors of medicine at Harvard Medical School and the co-founders of Physicians for a National Health Program, an organization of 18,000 doctors who advocate for a single-payer health care system. PNHP played no role in supporting their research.
The editorial concludes, "We worry that P4P may simply not work because it changes the mindset for good doctoring. However, if P4P schemes must be envisaged then rigorous consideration of their likely benefit prior to their implementation seems essential," referring to an exacting P4P checklist developed by the authors of the related article.
"Why pay for performance may be incompatible with quality improvement," Steffie Woolhandler, M.D., M.P.H., Daniel Ariely, Ph.D., David U. Himmelstein, M.D. BMJ, Aug. 15, 2012. An advance copy of the EMBARGOED article is available to media professionals on request.
The related article: "When financial incentives do more good than harm: A checklist," Paul P. Glasziou, Heather Buchan, Chris Del, et al. BMJ, Aug. 15, 2012.