Goldman Sachs CEO Lloyd Blankfein is into it. So are education secretary Arne Duncan and Gawkereditor A.J. Daulerio. Policy wonks champion it, charter school teachers embrace it, and professional athletes cash in on it. In our data-driven age, “performance pay” is the Next Big Thing. The logic is seductive: collect numbers, cut them up six different ways, and let a computer decide who’s worth how much. No favoritism, no bias, no human mess. The winners rise to the top, the losers don’t, and the whole system improves as a result of the competition.
Now, thanks to the Affordable Care Act, performance pay is coming to a hospital near you. Under a system that went into effect in October, the Medicare funds of some 3,500 U.S. hospitals are pegged to a dozen different metrics, including patient experience and outcome. The hospitals will lose a percentage of their usual federal dollars, with the opportunity to earn it back for showing improvement. The top performers can earn back more than they lost—not unlike a Christmas bonus—while the underachievers have to live with the financial hit.
The only question is, when it comes to medicine, does pay-for-performance actually work? Two recent studies, both in the New England Journal of Medicine, go in search of answers and arrive at different conclusions.
The more recent research, from a team of British physicians, looks at performance pay and patient mortality in 24 hospitals across northwest England. Using data from the National Health Service, the authors compared survival rates in the 18 months before a program was implemented with patient outcomes after it went into effect. Would a competitive, pay-for-performance schema encourage doctors to work harder, and smarter, to keep their patients alive?
The researchers found that, when adjusted for risk factors, 30-day mortality in the test hospitals dropped 1.3 percentage points, with significant gains especially among patients with pneumonia (less so among patients with heart failure and cardiac arrest). Over the year-and-a-half study period, these improvements equated to some 900 lives saved. The pay incentives seemed to work.
England’s program was modeled on a 2003 trial, known as the Hospital Quality Incentive Demonstration, conducted in the U.S. by the Centers for Medicare and Medicaid Services. But, as the authors are quick to point out, the British version was both better funded and more amply supported than its American predecessor, which likely had something to do with its success. In England, the hospitals were only rewarded for improvements—not penalized for failures—and administrators at competing care centers sat down regularly to share lessons learned. Bonuses were twice as large as they had been under the American HQID, and twice as many were handed out. In other words, the prizes felt real—and attainable.
The British findings stand in stark contrast to the results of the HQID trial itself, which appeared this spring in the Journal.
Under the six-year HQID, which ran in 252 hospitals across the country, more than six million patients underwent bypass surgery, or were treated for pneumonia, heart failure, or heart attack. Researchers from Harvard’s School of Public Health compared those patients’ outcomes—as evidenced by 30-day mortality—with control data from more than 3,000 non-pay-for-performance hospitals. They found that, between 2003 and 2009, survival rates at performance-pay centers did not differ significantly from standard hospitals. Whatever financial incentives might have done to improve patient experience and clinical care, they didn’t save lives.
The authors called the evidence “sobering” and noted that the performance-pay program mandated by the Affordable Care Act and drawn up the federal government looks an awful lot like HQID.
As of October 1st, of course, the point is moot. Whatever the data say, performance-pay is now a cornerstone of American Health Care 2.0. Don’t expect a cure-all.
