P4P is Dead. Long Live P4P!

P4P (pay for performance) is all the rage in healthcare. And why not? In the face of the damning evidence that we “get it right” only about half the time, criticizing linking higher payments to better care seems frankly un-American.

And there is accumulating evidence that P4P may work. A study earlier this year by my friend Peter Lindenauer found that P4P led to about a 3 percent improvement in quality measures (compared to public reporting alone) over two years. The Centers for Medicare & Medicaid Services (CMS) surely believes in it, promoting their P4P experiments at every turn. And where CMS goes, private payers usually follow.

But P4P is really tricky. As in…

  • How much do you pay for better performance? In most industries, bonuses of less than 5-10% are felt to be all-but-meaningless. But few believe that P4P bonuses will be more than 5%, because…
  • Where does the money come from? The UK National Health Service’s P4P experiment appeared to be going swimmingly until a smart actuary noticed that it was busting the NHS budget.
  • In fact, given budgetary realities, the bonuses for the winners are surely going to come from lower payments to poor performers. But what happens when safety net hospitals are forced out of business by the cuts needed to generate bonuses for well-heeled hospitals?
  • How can you be sure that you’re not skewing priorities inappropriately (like spending more energy on giving pneumovax to hospitalized patients – an element of the P4P program, but one with scanty supportive evidence – than on discharging them safely).
  • Finally, do you pay the top performers or the top improvers? I like to ask this question when I lecture on the topic: “There’s a terrific hospital that began at the 90th percentile and stayed there. Then there’s another, a plucky improver that started at 10th percentile and improved to 40th. Which one gets the bonus?” The audiences always split 50-50, which means that the (relatively wimpy) P4P payment will have to be split, in the best Solomonic tradition, as well.

The Lindenauer study has been characterized as being mildly supportive of P4P. It was, but it – together with a prior New England Journal study – also provides the biggest surprise of the quality movement. Here it is:

Transparency, via public reporting, works incredibly well, despite the fact that there is little evidence that patients are looking at or acting on the results!

This raises the key question about P4P, one that I don’t hear many people asking. Not, “Does P4P lead to improvement?” but rather, “Is any additional improvement that P4P generates above that generated by transparency alone worth all the administrative hassles and political battles?”

I’m guessing that the answer will be “no”. Does that mean that P4P is dead?

Far from it.

At my hospital, staff – including the residents – now receive bonuses when the organization performs well on the publicly reported quality measures. And, throughout the country, I see more and more senior leaders wading into the quality bonus pool.

What’s going on here? Basically, transparency is generating terrific social pressure within healthcare organizations, both from the embarrassment and shame that accompany poor scores and the elation that accompanies good ones. This pressure, in turn, is leading institutions (now hospitals, soon group practices) to figure out how to improve their performance. Everyone begins these efforts without P4P – buying bagels for the docs, distributing glossy newsletters – but sooner or later, most will feel the need to put some “skin in the game.” At that point, bonusing (determined by local rules and distributed through local channels) for better quality performance will become nearly ubiquitous. All driven by public reporting.

And I think that’s a good thing. I trust the hospital or the multispecialty group to thoughtfully move money around – respecting local conditions and culture and being careful not to do something silly and harmful – more than I do a 3rd party payer with an arms-length relationship to providers and patients.

So long live P4P! But before payers roll it out everywhere, let’s give public reporting and transparency a chance to generate versions of P4P within healthcare organizations. I suspect that it’ll work better.

1 Comment

  1. WRS on October 20, 2007 at 8:03 pm

    We’ve always have P4P–we’ve been paid for good performance and we’ve been paid for lousy performance.

Leave a Comment