The Bigger Picture: Pay for What?

November 1, 2007

Pay-for-performance may be inevitable, but that doesn’t mean the current programs are working well.


We’re told there’s too little uniformity in the way physicians diagnose and treat patients, especially those with chronic illnesses such as diabetes, and that lack of standardization is part of the reason healthcare costs are so high. Something has to be done. So why not pay doctors more for sticking to what the evidence says works in managing chronic illnesses? The patients would be better off, and it would be cheaper to treat them even after paying physicians what amounts to bonus money. So everyone wins, right? Great care gets great pay. Philosophically, I truly love the concept. And any fix for the episode-focused reimbursement system is welcome.

Trouble is, pay-for-performance isn’t working as advertised.

Recently, Medicare announced the first-year results from its Physicians Group Practice Demonstration. Under the demonstration, participating practices could get cash for meeting evidence-based guidelines in diabetes.

Each of the 10 participating practices met Medicare’s measurements goals. But only two actually made any money. Why?

Well, it turns out payment isn’t performance-based, after all. It’s based on how much money the practices saved Medicare, explains Chet Speed, vice president of public policy and political affairs for the American Medical Group Association.

Most of these practices made significant and expensive changes in technology and work flow - like creating patient registries and disease-management plans - yet a group that saved Medicare 1.9 percent in spending didn’t even get a thank-you note because the government takes 2 percent off the top.

The 10 participants were all large groups. They could make the requisite changes without too much pain, but it’s very doubtful that the average solo, small, or even midsized practice could make the same investments - especially if the return is zero.

Medicare’s pay-for-performance concept also assumes that beneficiaries primarily see a single physician who would get all the credit for controlling their chronic illnesses. Here’s the reality: According to a study published recently in the New England Journal of Medicine, in a typical year beneficiaries saw a median of two primary-care physicians and five specialists working in four different practices. For a third of Medicare patients, the assigned physician changed from one year to the next. It would be a bummer to counsel a patient extensively only to have the compensation for the resulting outcomes go to someone else.

There is also a lot of skepticism about Medicare’s new pet, the Physician Quality Reporting Initiative. Launched in July and running through December, the government is offering to pay you very small amounts of money to report data on various patient outcome measures. “It’s basically a pay-for-reporting initiative,” says Speed. “Congress wanted physicians to get into the practice of reporting data. … [But] a lot of physicians feel it’s … a fairly onerous process, and the payment is considered not to be worth all the additional time and effort.”

Still, all this might be acceptable if pay-for-performance truly did make medicine more evidence-based. But that seems to be happening, anyway. According to an article in the March 2007 Journal of General Internal Medicine, the proportion of primary-care physicians reporting that clinical practice guidelines had a large or very large effect on their practice increased significantly from 1997 to 2005, from 16.4 percent to 38.7 percent.

That’s a cultural shift, and not one based on the bribery implicit in pay-for-performance. And it just feels more real than the “lipstick on a pig” machinations of most pay-for-performance plans.

Pamela Moore, PhD, is senior editor, practice management, of Physicians Practice. She can be reached at pmoore@physicianspractice.com.

This article originally appeared in the November 2007 issue of Physicians Practice.