By Bill Toland
Pittsburgh Post-Gazette, Aug. 15, 2012
With insurers and Medicare hoping that they can cajole doctors and hospitals into providing better care by paying them for good performance, a pair of articles in a top medical journal is now arguing the opposite — that so-called “pay for performance” programs can have a detrimental effect, prompting some physicians to game the system in order to bring about desired results.
Insurers have their own evidence suggesting that paying doctors and hospitals extra money for carrying out certain standardized procedures can bring about improvements in quality, but the writers whose work appears in the peer-reviewed British Medical Journal this morning say that the long-term effect on patient outcomes is unclear.
“Despite a dearth of robust evidence that [pay-for-performance] is clinically effective in health care, payers charge ahead with implementing everywhere an intervention that has proven to work nowhere,” the authors wrote in an editorial that accompanies the main analysis.
Pay-for-performance programs presume that “as long as we just arrange the incentives correctly, people will figure out the exact things to do to maximize the revenue,” said Dan Ariely, one of the authors and a professor of behavioral economics at Duke University.
Often, they do figure it out — but while they are being rewarded for certain outcomes and processes, they are “neglecting some of their [other] obligations,” Mr. Ariely said.
In their editorial, the writers — several of whom are affiliated with Physicians for a National Health Program, which advocates for a single-payer health plan — note that “a growing body of evidence from behavioral economics and social psychology indicates that rewards can undermine motivation and worsen performance.”
The editorial also differentiates between “process-based” incentives — those tied to certain tasks — and “outcomes-based” incentives, which reward or punish a health care provider based on patient results.
Outcomes-based incentives are inherently less fair because they can reward, or punish, a provider for results or risk factors that may be partly out of their control. Some of the incentive schemes try to adjust for those risk factors, but that, too, can lead to unintended consequences.
For example, if a hospital knows that it can earn higher incentive payments by getting better outcomes on more complex medical cases, that can lead to “upcoding,” a practice wherein one diagnosis is re-coded as a more serious one. “Embellishing diagnoses [makes] patients seem sicker on paper, and hence boosts risk adjusted quality scores,” the paper said.
What about the many insurers that tout, say, vastly reduced Methicillin Resistant Staphylococcus Aureus (MRSA) infections and central-line bloodstream infection rates as evidence that pay-for-performance systems are working at hospitals?
“That’s not a scientific method,” said Steffie Woolhandler, a physician and professor of public health at the City University of New York.
A reduced infection rate does not necessarily mean that the payment was specifically responsible for the reduction — in other words, correlation does not equal causation.
Carey Vinson, vice president of quality and medical performance management at Pittsburgh’s Highmark Inc., said that the papers’ concerns are well-known in medical circles and, in some cases, well-founded. “There are limitations to research showing” the effectiveness of pay-for-performance, Dr. Vinson said.
But he also noted that there is plenty of evidence to suggest the current “fee-for-service” model, which reimburses hospitals and doctors for every test and unit of care, isn’t working, either.
“We agree that we’re not there yet,” he said. Insurers must take caution in designing and implementing performance programs, he said, and insurers and providers alike have to find a better way to link quality to payments.
One attempt to do so is Medicare’s new program that will penalize hospitals, starting Oct. 1, if their “preventable” readmission rates are too high, meaning patients are reentering the hospital too soon after discharge.
According to an analysis by Kaiser Medical News, more than 2,000 hospitals nationwide will be penalized, including more than 100 in Pennsylvania and most of the hospitals in southwestern Pennsylvania.
Dr. Woolhandler, and others, argue that “penaliz[ing] low-scorers can make matters worse, effectively punishing patients who have nowhere else to go,” especially those in urban and low-income areas.
Read the editorial at https://www.pnhp.org/sites/default/files/docs/nc/bmj-incentives-edit.pdf.
Read more: http://www.post-gazette.com/stories/business/news/top-journal-dont-bet-on-performance-pay-649020/#ixzz246MNZ730