I have some serious concerns, as a physician, about pay-for-performance.

When outcomes measures are used, there are many variables outside of the physicians control. These make the actual “performance” as measured by whatever means, highly patient dependant.

If we look at outcomes measures – i.e. less lost work, less sick days, less hospitalizations, lower costs, whatever, and then rate physicians based on these outcomes, physicians will be incentiviezed to select for patients likely to have good outcomes. The 55 year old, history of 2 MIs, diabetic, who refuses to quit smoking and eat twinkies is not a patient I would want on my panel. Their outcomes will be bad, and negatively impact on my “performance.”

On the other hand, if you rate performance based on things such as: physican recommended smoking cessation, physician recommeneded aspirin, i.e. things that can be studied and measured and which the phsycian can do (rather than hoping the patient will do), at least there is some consitency. With true outcomes measures, two physicians with identical methodology, philosophy, and recommendations will have very different “performance.” The physician will be penalized for the patients behavior.

As the source of the original quote above, I must disagree with the premise of some of these arguments. Dr. Centor, my point is precisely that you, as the physician, know best what to do with each and every patient. So, measuring process, defined as things you do, is not helpful. To quote Dr. Lacsamana, “The caring and personal touch of the doctor would be submerged by cold statistics.” Absolutely true, and that is my problem with process measures – we can only say “well, he didn’t do enough counseling, he should have done this or that” – but we weren’t in the room, we did not have the conversation, we don’t know the signs and symptoms, or myriad other contributors to the “outcome”. We do not know what you should have done. Therefore, the fairest way to assess performance is not measuring what you did, but what were the results. I don’t want the health insurance plans that are my clients telling you what to do, I want them to figure out which doctors have the best outcomes and send patients to those docs.

If you want to stand on your head and tell knock knock jokes to one patient and administer oxycontin to the next, and both patients improve and become more functional, who am I to say “that’s bad medicine” – I am not the doctor, and if that is what you need to do to treat the patient, and if the patient gets better, great.

As to your point about “process demonstrably improves outcomes”, what outcomes? what is your basis for that claim? what evidence do you have that doing a certain procedure, test, therapy produces a higher level of functionality, a more productive person, a “healthier” individual? The “outcome” that matters to your patients is that you improved their health, their ability to function, their independence. The “outcome” that matters to the employers that are paying (albeit grudgingly) your bills is employee functionality and productivity.

I have worked extensively in the area of medical guidelines, clinical criteria, disability duration guides, and I have seen very few attempts to link medical care to functionality, productivity, or ADL. Sure, there are a few, but they are far between.

Finally, to the issue of case-mix adjustment – yes it is difficult; yes it is fraught with problems, but payers will not let that hinder their efforts to identify the top performing physicians and send patients to them. And they are working hard on case-mix adjustment techniques and methodologies because most recognize that patients ARE different.

Would you not rather be evaluated based on your results than on whether you complied with some “medical director’s” idea of appropriate treatment?

I’m looking forward to the success of outcomes-based medicine–its time has definitely come. However, reading the comments above, I picture this: a harried physician caught between crafting a tailor-made treatment plan and a backed-up office schedule, while the payor stands there in a green eyeshade with a clipboard and a checklist. Meanwhile, the patient sits there like a confused, passive outsider.

The patient MUST become a more active participant in his care, and he deserves our help. 15-minutes of bedside counseling may have been a breakthrough for this 55 y/o diabetic smoker; but it took him 2 MI’s to get it!

How do we teach every patient the rationale behind their treatment to ensure their compliance? What will reach them?

[...] t quality measures are chosen. All physicians know that much of quality is immeasurable – A comment on pay for performance. Nonetheless, whether the ACP (or any [...]

how about this — patients get to review the eobs. If a patient is delighted with dr x, patient can provide a rating that will up doctor x’s reimbursement rate by 10% or whatever. If patient is unhappy with dr x, patient can downgrade doctor x’s reimbursement by 10%. If doctor x blows patient off and patient x has to go to second doctor for same problem, and patient complains, doctor x can get 75% of visit cut from payment. Why should Dr X get paid for no value added and time away from work? If patient reports that doctor took 45 minutes for a complex issue, shouldn’t that matter in p4p — if doctor breezed through in 2 minutes, didn’t pay attention, and is asking $400, shouldn’t the patient get to turn off the tap? Isn’t the real measure of performance if the patient finds themselves better? Or treated appropriately and with dignity? patients are not so dumb to that they haven’t figured out the pay for procedure gig and have figured out that they mean zero in the process. Doesn’t this address both quality and efficiency — no patient wants to get dragged all over the place with tests and medication they don’t need, but nobody wants to stay sick when they don’t need to either.

If SOMEONE would then publish even minimally reliable data on the doctors, (how many procedures of what kind, how many failures, what percent are complex, level of patient satisfaction, etc. patients could steer themselves to docs whose practices are appropriate to their needs (and not be referred to someone’s golfing buddy), good doctors would prosper, bad doctors would become drug reps and we would all be happier.