Saturday, April 22, 2006

P4P is Unethical

Point/Counterpoint on Pay-for-Performance is a fascinating look at the latest buzzword, P4P, pay for perfomance. Dr.Robert Centor of DB's Medical Rants faced off against Dr. Michael Barr.

On the face of it, who could object to P4P? A doctor will be paid based on his performance. The better the quality of medicine he practices, the more patients he saves and improves, the more he will be paid. Oh, wait, that's not what P4P really means. According to Dr. Barr, speaking in support of P4P, the physicians who will be most highly compensated are "physicians who are perceived to be delivering higher quality for lower cost."

I have been thinking about that phrase for several days, and getting increasingly angry. That's because P4P as described by Dr. Barr is fundamentally unethical.

First, an example:

If an oncologist treats 10 patients at a cost of $1,000 each, and 9 survive, he has saved 9 lives for $10,000.

Now imagine that a second oncologist treats 10 similar patients. The first 9 respond to the $1,000 treatment, but, again, the 10th patient does not. However, this oncologist refuses to give up and creates a new treatment regimen. This regimen fails, too, and the patient is hospitalized with multiple complications of his disease. On the next try, the oncologist comes up with a life saving regimen, and patient #10 lives. Of course, between the first failed attempt, the second failed attempt, the hospitalizations and the third successful attempt, an additional $10,000 has been spent. The second oncologist has saved 10 lives for $20,000.

Whose "performance" is better? Who should be paid more, the doctor who managed to save 9 people at a total cost of $10,000 or the doctor who saved everyone by refusing to give up and creatively designing new treatment regimens at a cost of $20,000?

Obviously, the second oncologist is the "better" doctor. However, I suspect that a P4P system would consider him less cost effective and would penalize him accordingly. In fact, they might penalize the doctor quite severely since he cost the insurance company double the "average" amount spent by the first oncologist.

The example above shows the perverse results of a P4P system which uses "higher quality for lower cost" as its benchmark. It is not surprising that such a system would deliver perverse results because such a system is unethical on its face.

The fundamental relationship in medicine is the doctor-patient relationship. Society and the law recognize this by privileging this relationship in comparison to other types of relationship. The doctor has a MORAL and and legal obligation to put the patient's interests and well-being above his own. Obviously, not every doctor will do that. There are some doctors who might recommend expensive treatments purely to enrich themselves. However, we understand those doctors to be unethical, and they may even be subject to legal action.

In dramatic contrast, however, P4P attempts to inject the insurer into the relationship. Even more objectionable, the insurer asks the doctor explicitly to balance the patient's interests against the doctor's financial interest. This is fundamentally unethical and should be banned as a result.

American democracy is a rights based system, not a utilitarian one. Ethically and legally, you are not allowed to violate a person's rights even if it will increase overall happiness or utility. Each person in a democracy is shielded from the power of others and the power of government by these rights.

Similarly, the sanctity of the doctor patient relationship is a moral right. Insurers are not free to violate it simply because it may free up money to care for others (or more likely to profit the insurance company). Furthermore, it is UNETHICAL for an insurance company to ask doctors to violate this patient right.

Doctors should stand firm on this important point. We should refuse to participate in any system that is unethical on its face and we should aggressively charge the industry with their ethical violations.


Blogger Tom Leith said...

This is going to ramble a bit, and it is kinda rough. I have been asked to help explore more fully these questions over at . I have not started yet -- I am still gathering thoughts.

> The doctor has a MORAL and legal
> obligation to put the patient's
> interests and well-being above
> his own.

Very well, then. Let us outlaw both medical and hospitalization insurance, eliminating once and for all the entirely artificial distinction, created 80 years ago by physicians mind you, between "professional" and "technical" services. Henceforth, it shall be illegal to enter into any mutual insurance contract to finance medical risk of any kind. Further, hospitals shall bill doctors directly for all services and materials they order, or are reasonably necessary for the care of a patient he has admitted to the institutions. Doctors shall directly bill patients or their guardians or their estates for their services, including any services provided to the patient under a subcontracting arrangement, as with a hospital. The doctor's risk is the risk of not being paid by the patient, and as you say he is MORALLY OBLIGATED to take this risk because the doctor-patient relationship is sacred. We can surely count on doctors to honor their MORAL obligations, and faithfully serve any patient who comes along without regard for their own well-being.

If anyone is laughing, I pause now to allow them to compose themselves...

I think it is under these conditions that definitions for "rights" and "adequate" and MORAL obligation should be tested. Moral teachers, including modern popes, have said that all human beings have a right to adequate healthcare. They do not, however, tell us what this means. We do have some hints: Pope John Paul II has said that every living human being has a right to warmth, cleanliness, and nutrition and hydration so long as the body is capable of utilizing them. He says these things are never disproportionate even when artificially provided, and his standard of futility is quite strict. The popes have said also that burdens and benefits may be licitly compared; and that burdens to people other than the patient count as true burdens. But again: it is left to us to reduce this principle to practice. Peter Singer has said that it is wrong even to buy at one's own expense medical services when someone else could benefit more from having them -- one has an obligation to forego having the services oneself, and has a further obligation to obtain the services for others. He says we each have a moral obligation to bear increasing burdens up to the point that we are hurt more than anyone is helped. That all this went out the window for him when his own mother was nearing the end of her life is not evidence he is wrong about it.

So, the popes and Singer seem to agree it is not clear that anyone has a "right" to be cured of cancer at all, or leaving aside the definition of cost, that anyone has a right to be cured at a cost to others of ten times the expected "cost". The cure of cancer may well fall beyond the bounds of "adequacy". I personally think it can.

Medical services are first of all services. Doctors and everyone else involved in the delivery of these services expect to be paid as barbers, janitors, auto mechanics, investment advisors, and nurses, laundry workers, and cooks expect to be paid. If I own a restaurant and somebody comes is and plunks down a crispy, new $10 bill, and orders "Surf `n Turf" who will say he should be served lobster and a wonderful New York Strip Steak rather than a couple of fried shrimp and a burger? Don't we all have a "right" to food? And does anyone say a "right to adequate food" mean it should be prepared by exquisitely trained chefs and obsequiously served in beautiful surroundings? What if he comes in broke and says, “I have a right to food”? Must I do the cooking?

If a patient comes in with $10 resources and $100 wants, he hasn't a right to $90 from somebody else. The $90 may be given as charitable or humanitarian aid, and there may well be a moral obligation to provide charitable or humanitarian aid, but an obligation on the part of one person does not create a right on the part of another person to anything in particular. A bag of beans and rice, a jar of Cajun spices, a cheap pot, and fifteen minutes' instruction will adequately feed a man, and feed him better than much of the world eats today. If Peter Singer is right, even geography does not matter: we should all be eating beans and rice, and it should be illegal to cultivate anything but beans and rice until malnutrition is eliminated from the face of the earth. And so long as there is one child at risk of FAS (or something like this) nobody should be cured of cancer.

> Whose "performance" is better?
> Who should be paid more, the
> doctor who managed to save 9
> people at a total cost of $10,000
> or the doctor who saved everyone
> by refusing to give up and
> creatively designing new
> treatment regimens at a cost of
> $20,000?
> Obviously, the second oncologist
> is the "better" doctor.

This begs the questions:
1. What is performance?
2. Who is paying?

So what is “performance”? Is it "better" in medicine to behave as though non-technical constraints do not exist, and leave it to others to clean up the mess later? Or to face all constraints up-front? If the patient has made financing arrangements that pay for more or less "average" treatments but ends up with an unresponsive disease, what should be done? Has the doctor a MORAL obligation to attempt a cure if it is technically possible, even when it means personal bankruptcy for him and his family? Should the patient (and doctor) be able to demand open-ended funding for a doctor's "creativity" for as long as the patient manages to survive the attempts? I think the answer to both these last questions is “no”.

Even if it is true that the needs of any individual patient trump every need of the unlucky doctor to whom he presents, do his needs or desires trump absolutely every other need? If not, is there no role for a doctor is protecting those other needs? If not the doctor, then whom?

When we ask about doctors being paid more or less, we should ask, "paid by whom?" The National Institutes of Health? Pfizer? The patient? Or the patient's insurance group?

What is the doctor being asked to do? Groundbreaking medical research? Apply the research of others in concrete cases? Take the best possible care of the patient within the set of constraints, including the financial constraints that come with that patient?

If we replace mutual insurance by contract with social insurance, none of these questions will go away.

Moving from questions of obligations toward an individual to more concrete examples of P4P:

> Whose "performance" is better?
> [...] the doctor who managed to
> save 9 people at a total cost of
> $10,000 or the doctor who
> saved everyone by refusing to
> give up and creatively designing
> new treatment regimens at a cost
> of $20,000?

This isn't generally the question in P4P. The question is more like:

Whose "performance" is better? [...] the doctor who managed to save 10 people and billed $20,000 or the doctor who saved eight clinically equivalent people and billed $50,000? The doctor who actually gets a few patients to quit smoking, or the doctor who refuses even to address the issue with his patients? The study of practice variation says these are the more realistic questions. Yes, I know I know risk adjustment is not perfect. I also know that "six" is a very small sample. So I don't ask questions about six patients, I ask about the doctor's processes and practice patterns, trusting that what is being taught in the best medical centers today is an improvement over what was being taught in the best medical centers 20 years ago, and trusting that the results are replicable.

> In dramatic contrast, however,
> P4P attempts to inject the
> insurer into the relationship.
> Even more objectionable, the
> insurer asks the doctor
> explicitly to balance the
> patient's interests against the
> doctor's financial interest.
> This is fundamentally unethical
> and should be banned as a result.

I deny the major premise, and the minor premise. The patient has invited the insurer into the relationship. The insurer is not "injected". When the patient does not want the insurer, the patient can pay out of pocket. When the doctor does not want the insurer, the doctor can refuse to accept assignment. Both can decide whether they are better off. Refer back to the gedankenexperiment at the top of the page.

Further, the insurer does not ask the doctor explicitly to balance the patient's interests against the doctor's financial interests: the insurer asks the doctor explicitly to respect the contractual agreement made by the patient with the insurer, or the terms under which philanthropic funding is being provided. She is being asked to educate the patient about these things, and present the reasonable options available under the funding constraints. She is being asked to provide these services in a diligent and conscientious way. In return for providing these very valuable services, she will be paid a performance bonus.

This said, a P4P that asks only about "financial performance" and not "clinical performance" is not the kind of P4P I think anyone should tolerate. Unfortunately, this is the easiest kind to implement.

I am just getting started thinking about a systematic approach to The Ethics of Medical Services Financing.

6:19 PM  
Blogger Ex Utero said...

Dr Amy,

this is unrelated but I just want to complement you on your persistence and your excellence on Neonatal Doc's Home birth Blog. You have a lot more patience than I do. I'd like to think we might have saved a life or two with that blog.

Ex Utero (a.k.a. cherubs in the land of lucifer)

10:30 PM  
Blogger Amy Tuteur, MD said...

Dear Ex Utero,

Thank you for your kind words. For those who may not have seen it, I participated in an extended debate about homebirth on Neonatal Doc.

I'm sure I didn't convince any homebirth advocates, because for them, it is a matter of faith, not evidence.

The string of comments actually included a comment from a woman whose baby had died from a preventable cause at a homebirth. She blamed no one and did not think that it reflected poorly on homebirth. If the advocates of homebirth could ignore a preventable death among one of them, they were hardly likely to be swayed by statistics.

7:45 AM  
Blogger Amy Tuteur, MD said...

Dear Tom,

I never said that there was no moral role for the insurance company; what I said was that it is unethical for the insurance company to interject themselves into the relationship between the doctor and the patient. If the insurance company does not believe that a particular test or procedure is needed, they should tell the patient that they will not pay for it and the patient will have to pick up the cost if he wants to follow the doctor's advice. This is very different from trying to pressure the doctor not tell the patient about his options in the first place.

Second, the fundamental assumptions that underlie P4P are wrong. Doctors do not profit at all from most of what they recommend. They make no money from prescribing drugs or tests. Therefore, in order to get them not to recommend drugs and tests, you must penalize them. Moreover, such penalties don't even have the benefit of offsetting a profit motive that might cause a doctor to recommend a surgical procedure that he will be paid for. They are simply punitive.

In addition, P4P relies on the notion that over prescribing by doctors is at the heart of the healthcare crisis. The heart of the healthcare crisis is COST and doctors have no control over the cost of drugs, tests, etc.

If you want to apply P4P, why not start there? Tell the drug company that the price you will pay for a drug or procedure will be determined by how effective it is. If the drug does not work as advertised, if it is not as effective as substitutes, or if it causes side effects that require additional expenditures, withhold payment from the drug company! They are the people responsible for both the price and the effectiveness of the drug. It seems clumsy, arbitrary and, above all, punitive to punish the doctor for the drug companies' actions in overpricing medication.

8:02 AM  
Blogger Tom Leith said...

What I had said is that the insurance company has not injected itself into the doctor-patient relationship: the insurance company has been invited in, mainly by the patient.

If a doctor wants to work within a patient's feasible choice set and refrain from mentioning infeasible options, I do not see that as inherently unethical, and I daresay this kind of thing has routinely been in every doctor-patient interaction since classical antiquity. I wonder what motive there might be for a doctor making recommendations outside of a patient's feasible choice set. As for the notion that doctors do not profit from most of what they recommend, I have learned enough about referral patterns in the last few years to see through this, and I know that pecuniary considerations are not the only considerations. Further, pandering to patients will keep them coming back. Do not try to tell me this isn't done, or Nexium would not be advertised on television, antibiotics would not be overused. There would probably be fewer imaging centers. I do not know what else. I doubt that surgery is overprescribed, except maybe c-section births.

I do wish the terms of insurance contracts were better understood by patients. Some plain language like: "Although we hope to work in partnership with you and your doctor, when we disagree with your doctor about medical necessity or likelyhood of success, you should of course feel free to follow your doctor's advice, but we will not pay anything for it. We have the final say with respect to what we will pay for. This will leave you no worse off than you would've been without us in the first place. Our liability will be limited to the refund of your premium for the year, less any payments we have made during the year on your behalf." Plain English Policy Language would be good. It would also be good if insurance companies would make their expected methodologies better known to doctors (and patients). This could save a lot of headache.

I personally like QUALY-style methodologies, as we have discussed over on The Healthcare Blog, better than apparently ad hoc approaches, but QUALY has problems too. However we approach this, there has got to be a limit to the claims an individual may make upon whatever risk-pool he is in.

The P4P programs I have heard of do not rely on the notion that over prescribing by doctors is at the heart of the healthcare crisis. They make quite the opposite assumption: that Best Practices are not followed, and this amounts to underprescribing and misprescribing. The intellectual basis for this assumption is the study of practice variation. 75% of the healthcare spend is for chronic diseases and their complications. We have examples of where chronic conditions are managed better, sometimes three times better, than than they are in other places. Yet the differences persist. P4P is aimed at narrowing the differences.

Insurers do most of what you want with respect to the price/efficacy ratio of drugs when they do formulary design and tiered copayments in the benefit design. It isn't perfect, but no system will be.

The only basis for saying that any medication is overpriced is "poor sales". Everything is priced based on the value it provides to the person paying for it, not on the marginal cost of production. If a patient is convinced he gets three or five times the value out of Nexium that he gets out of omeprazole, then more power to Astra-Zenica. What can you do in order to provide three or five times the value to your patients for only a little incremental cost?


4:01 PM  
Anonymous Anonymous said...

If an insurance company can decide on how and what is needed to treat my illness, then the insurance companies need to be managed by Doctors with extensive experience.

8:28 PM  

Post a Comment

Links to this post:

Create a Link

<< Home