August 12, 2007 10:03PM

P4P: I Disagree with Arnold Kling and Dr. Bob

Arnold Kling agrees with Dr. Bob regarding existing efforts to pay physicians for quality — known as pay‐​for‐​performance, or P4P — rather than for the volume and intensity of the services they deliver. Dr. Bob argues:

  1. “High quality — while not invariably more expensive — is often so.”
  2. “[B]y and large,” the guidelines that physicians are supposed to follow “don’t exist — except in a few relatively straightforward areas of medicine.”

I agree with those statements, but I disagree with their conclusions. 

Though the first statement is true, it is also true that a lot of the expensive stuff that doctors deliver is not high quality. For 30 years, researchers at Dartmouth Medical School have found it very easy to demonstrate that some doctors do a lot more expensive stuff than other doctors do (e.g., specialist consultations, hospital stays, etc.). But they have found it very, very hard to find any evidence that that extra stuff makes patients any healthier or happier. Thus, a lot of the expensive stuff that doctors do isn’t high‐​quality care.

Though the second statement is true, it is also true that where evidence‐​based guidelines do exist, patients still don’t get the “high‐​quality” care that the guidelines recommend. According to Elizabeth McGlynn and her colleagues, patients receive such recommended care only about 55 percent of the time. (I put “high‐​quality” in quotes because not every patient should receive what the experts recommend. But it would be a stretch to say that 45 percent of patients are outliers.) Even when evidence‐​based guidelines exist, doctors don’t follow them.

Quality suffers both because physicians don’t do enough of what they should, and do too much of what they shouldn’t.

Physicians generally support P4P incentives that pay them more to do more of the former. But they really hate P4P incentives that penalize them for doing too much of the latter. (The American Medical Association supports only two types of P4P incentives: those that increase the incomes of some physicians, and those that increase the incomes of all physicians.)

I think third‐​party P4P, where insurers reward providers for high‐​quality care, is a fine idea – provided the patient gets to choose her insurer. For a good overview of the quality problems in the U.S. health care sector, the difficulties in implementing third‐​party P4P, and why the federal government should stay out of the P4P business, read my law journal article.