In early February 2006, a national forum on pay-for-performance (or P4P, as it has become known over my strenuous objections) will convene in Los Angeles. Leading experts and policymakers will be there to debate the appropriate role and structure of one of the key underpinnings of the quality movement. Given recent developments, they will have a lot to talk about.
It's too early in the game to be drawing final conclusions, but we now have enough data from a cornucopia of private- and public-sector payment initiatives to start reframing the debate.
For starters, some good news: There is now hard evidence that programs that include payment incentives push providers to adhere to standards of care and improve patient safety. Unfortunately, it seems that the extra cash is going to the highest-performing providers, while the biggest jumps in improvement may be coming from those on lower quality strata.
That was the finding of a study in the Oct. 12 issue of the Journal of the American Medical Association. The researchers also found that in only one of three clinical areas studied was there a differential in quality improvement among groups eligible for bonuses and control groups that weren't. Extra reimbursement that amounted to less than 1% of total practice revenue wasn't a big enough carrot.
That finding was underscored by another major study of seven different P4P models, which found that it may be the reporting of comparative data on physician performance that drives improvement, not simply payment incentives. And data from the CMS-Premier Medicare hospital demonstration project also seem to support that view. Even with relatively paltry bonuses, hospitals in that program showed significant improvements in each of five clinical areas, even those facilities that failed to qualify for extra reimbursement.
A rather large monkey wrench was thrown into the works last week by a study of current pay-for-performance models that found many models may inadvertently penalize providers who treat a larger share of underinsured or uninsured patients, jeopardizing access to care.
Finally, the Institute of Medicine, believing that P4P initiatives have been muddied by disparate program designs and clinical guidelines, calls for a universal set of performance measures and creation of a National Quality Coordination Board under HHS. The report has drawn some industry criticism for its recommendations, in large part because quality stakeholders are threatened by a loss of control and providers fear loss of revenue if comprehensive payment incentives include sticks as well as carrots.
All of this evidence should prompt those at February's conference to look at some new models. In fact, they should heed my earlier warnings and change the moniker for their movement. It should be called payment for improvement. Providers ought to be rewarded for meeting minimum benchmarks. There could be accelerating incentives for achieving each tier of performance, with the tiers being recalibrated regularly to reflect overall quality advances. Incentives overall should surpass 5% of total provider revenue.
There should also be public reporting of all the data, so that consumers and health plans can make rational decisions on which providers to use. That should obviate the need for cuts in payments to chronically poor performers.
Finally, the latest study on access underscores the need for a national program, as the IOM suggests. Some way must be found to compensate safety net providers so they aren't penalized for treating the medically indigent.
Obviously, this is going to be an arduous process, with continuous monitoring and refining needed. We're not even at the end of the beginning of the quality movement, and there are going to be plenty of mistakes made before we can reach that place where we can limit those other kinds of mistakes, the ones that harm patients.