Pay for Performance

In our reading for this week, Leape and Berwick note (with a hint of exasperation, perhaps?) that, "despite the widely disseminated message from the IOM that systems failures cause most injuries, most individuals still believe that the major cause of bad care is bad physicians." In other words, while people tend to blame the healthcare system as a whole when it comes to problems with cost or access, they tend to blame individuals when it comes to problems with quality of care.

And why shouldn't they? After all, the doctor-patient relationship is at the core of the "quality" concept. And I'm not talking about all the other myriad forms of "quality" here; I'm just talking about the quality of medical care, not quality of access or coverage or portability or what have you. Quality of care is clearly determined by some sort of concensus between doctors and patients. It is essentially defined (or rather, undefined) by the tension between these two populations: does "quality" mean patient satisfaction or desirable medical outcomes? Is the patient's perception or the physician's a better measure of quality? Or, to put it another way: if the patient is happy, has the doctor done his or her job? Is the doctor's job to make a patient happy? Or is his job to make a patient healthy?

Patient satisfaction and patient health are obviously two very different things, but the fact that they're also very much related makes determining the minimal standards for "quality" very difficult. For instance, I strongly believe that the pay-for-performance (P4P) payment system is the right approach to improving quality of care. Since the doctor-patient relationship is what essentially determines quality, it makes sense to tackle the problem from the ground-up by changing physician incentives, rather than from the top-down by imposing systemic restrictions. However, creating incentives under P4P is very tricky. Do you pay doctors more for better medical outcomes for their patients, or for better patient satisfaction? Neither is objective, but you can't have quality care plus the PERCEPTION of quality care (an arguably more powerful political force) unless both agree. So do we award doctors for lower blood sugar levels, or for making patients feel better able to manage their diabetes? Ideally, we would do both, but which should take precedent in case the two measures disagree?

This is all the more difficult to figure out on a policy level because the US has paradoxically the most advanced, well-equipped, and cutting edge medical facilities of any nation (a physician-side "quality" measure) but also the worst continuity of care (a patient-side "quality" measure). US patients are not happy with their doctors, as this graph plainly shows:

Yikes. But for those of you who think more preventative medicine is the answer, here's a graph from the year before:



So even though US patients are plainly GETTING their primary or preventative care (if we understand that to be breast cancer screenings, blood pressure checks, etc.), they still FEEL that their primary care (or "regular") doctors aren't talking to them enough about what's going on.

I'm not posting this to be critical of either the doctors giving the care or the patients recieving it, but simply to illustrate that it is entirely possible for one group to say that the first graph is inevitable and the second wonderful, and for the other group to say that the first graph is terrible and the second inconsequential.