David R. Henderson  

Reinhardt on Doctors' Monopoly

My Two Modes... The Iron Laws of Pedagogy...
Organized medicine invariably opposes wider scopes of practice and independent practice of nonphysician health professionals, ostensibly not to protect economic turf but to protect the quality of patient care. Curiously, one rarely finds those to be protected by this paternalism vocally on organized medicine's side.

Not many economists today are buying the medical profession's position on this issue. More typically, economists lean toward [Milton] Friedman's more cynical view. They regard professional licensure of any kind - almost always proposed by the very professionals or occupations to be licensed - mainly as a means to endow the licensees with monopolistic market power.

This is from an excellent New York Times blog post by Princeton University health economist Uwe Reinhardt. It's titled "The Dubious Case for Professional Licensing."

Reinhardt quotes extensively from one of the most impressive chapters in Milton Friedman's classic 1962 book, Capitalism and Freedom. I still remember reading that chapter when I was 17 or 18 and being blown away by Friedman's reasoning on an issue I had never thought about but found myself totally persuaded about. Reinhardt is persuaded also. After quoting Friedman, he writes:

Friedman has fewer objections to certification, which practically means that no one without the education and training of a physician could call himself or herself an M.D.; but he would let patients, not physicians, decide from whom patients can seek medical treatments. I share that view.

Reinhardt writes his post in the context of current debates about how much latitude to give nurse practitioners. He, like many other health economists including me, advocates wide latitude. The case is especially strong given the high percent of doctors who will not take Medicaid patients. Reinhardt writes:
On its surface, this concern for the quality of medical care received by Americans seems convincing. Yet a recent paper by Sandra Decker in Health Affairs reports that a third of primary-care physicians (general and family medicine, internal medicine or pediatrics) in the United States do not accept Medicaid patients, presumably because the fees Medicaid pays are considered too low. In California, the percentage of primary care physicians refusing to accept new Medicaid patients falls in the range of 44 to 54 percent.

An economist can understand that physicians refuse to treat Medicaid patients at low fees when the opportunity cost of doing so is treating patients at higher fees. But what is to be done for the patients whom close to 50 percent of California primary care physicians refuse to treat?

What if independently practicing nurse practitioners were willing to see Medicaid patients at Medicaid's fees for the range of primary care services for which nurse practitioners are educated and trained? Would the California Medical Association contend that for patients whom physicians refuse to serve, the next best option is no care at all? Or that properly educated and trained nurse practitioners could render such care, as is done in 17 other states (see, for example, evidence from New York).

Well done, Uwe! BTW, those who don't follow the health economics blogs might not know that Uwe, a strong advocate of heavy government intervention in health insurance and health care, often tangles as a commenter on John Goodman's health policy blog. But this is something all three of us agree on.

HT to Ross Levatter.

Comments and Sharing

COMMENTS (16 to date)
Glen S. McGhee writes:

Randall Collins has a section in his book The Credential Society with the title "The Medical Monopoly" (138-147). Historical reasons are why the medical profession has been the most successful at monopolizing (among other things) who produces doctors and how many are produced, who gets to be a doctor, etc. The hospitals are also to blame, but here too, doctors benefited from them as well -- also, universities benefited immensely from incorporating medical schools into their structures (after the 1910 Flexner Report, sponsored by the AMA's Council on Education, caused many to close down).

Collins (1979) fields some half-serious proposals, but points out that most of the relevant education for doctors takes place on-the-job, and the rest is just so-much splash and dazzle -- much like the shamans of far away and long-ago. And, more recently, my vet said the same thing about animal medicine.

The comments are pretty good too;

Uwe ReinhardtPrinceton, NJ Just because I happen to be lucky enough to be reasonably well to do and to be living in an academic Disneyland does not -- or should not -- mean that I cannot give some thought to the problem of how one might enable less fortunate Americans whom licensed American physicians refuse to treat access to at least some health care, namely the scope of services that NPs can render and for which they are educated and trained.

Licensure has always been a barrier to such solutions. Licensure undoubtedly has some social benefits, but I and many economists are convinced that the costs outweight the benefits.
Oct. 14, 2013 at 10:59 a.m.

About 15 minutes into this interview with Sherwood Schwartz you can learn how the AMA kept one Jewish kid out of medical school in the 1930s. But at least the world got Gilligan's Island.

Brad writes:

But David, medical certification ensures a minimum level of competence among providers. We can't allow ordinary (and stupid) citizens to see anyone they wish for their medical care. It's far better to price half the population out of good medical care than to allow the whole population to receive an unknown level of care.

MingoV writes:

People who are not physicians, including nurse practitioners, do not realize the complexity of modern medicine. Four years of medical school and a minimum of three years of residency are barely enough to impart an adequate amount of medical knowledge and skills. Most specialists need to add two to seven years of fellowship training. There is no way that the average nurse practitioner (one or two years of post-BS training) or physician assistant (two or three years of post-BS training) can match the knowledge and skills of the average physician.

So, do I (a pathologist who had 4 years of residency and 2 years of fellowship training) support government licensure? No. Do I support government restrictions on scopes of practice? No. My "informed patient" proposal is that information on training and abilities be widely available for each type of clinician, and that each clinician regularly update experience lists. For example, a general surgeon might report 58 appendectomies, 65 gall bladder removals, 12 ulcer repairs, etc. A nurse practitioner might report 140 flu diagnoses, 110 hypertension diagnoses, 15 ankle sprains treated, etc. Ideally, independent organizations could evaluate and publish patient outcomes, but I cannot predict how many clinicians would pay for such services. Clinicians also could participate in a modified BBB program. These non-government approaches would provide more reliable information than government licensure, and would help patients make informed choices.

David R. Henderson writes:

You don't understand certification. Please reread my post, especially the paragraph from Reinhardt about certification.

Julien Couvreur writes:

I suspect Brad's comment was tongue-in-cheek.

Anyhow, I wanted to share some perspective from from nurse friends of mine in France. The anecdotes I heard suggest that nurses there routinely break the law and perform some more advanced procedures. Otherwise they'd have to wait for the doctor to arrive to do it. They said the hospital could not work if they followed the rules.
I admire their dedication, as they seem to expose themselves to possible liability and other charges in order to help patients.

Tracy W writes:

@Mingo V: the question is not whether a nurse practitioner can match a doctor's levels of skills and experience, it's whether they can do better than nothing at all.
I think though your experience list is a very good idea, what would you say happens to beginning medical types?

Jeff writes:

Why not just remove the government-imposed constraints on the number of doctors? To work as a physician in the U.S., you have to have an MD and complete a residency here. The Congress really decides how many residencies there are, and they keep the number low at the behest of the doctors lobby.

Even better, we could quickly increase the supply of doctors and bring down the cost of health care by allowing doctors who did residencies in other countries to practice here.

David Henderson Author Profile Page writes:

Julien, I'm pretty sure you're right that it was tongue-in-cheek. But if Brad had understood the point about certification, he would have made his tongue-in-cheek comment about licensing, not certification. Certification prevents no one from practicing medicine.

Ross Levatter writes:

I must respectfully take issue with fellow physician MingoV above. Not because I disagree with him per se, but because I find his points somewhat...ah...off point. Thus:

Mingo argues, “Most specialists need to add two to seven years of fellowship training. There is no way that the average nurse practitioner (one or two years of post-BS training) or physician assistant (two or three years of post-BS training) can match the knowledge and skills of the average physician.” That’s true, but off point. The issue isn’t categorical, but individual. The question is whether the value represented by an autonomously practicing lesser trained healthcare professional is greater, at the margin, than the cost.

Mingo is well aware of the spectrum of quality among licensed physicians, all of whom received medical school and residency training. I doubt he believes that, because gastroenterologists know significantly more than family practitioners about GI disease, family practitioners should immediately refer all cases of GI-related symptoms to gastroenterologists. Nor does he believe that all physicians who graduated in the lower half of their medical school class should automatically get referrals on all cases from those physicians who graduated in the upper half of their medical school class.

My point is that, if you look at the skill sets and knowledge of all non-MD healthcare professionals and compare them to the skill sets and knowledge of all MDs, you do NOT get a step function, where all in the former group show equal abilities, all below those in the latter group, which again show equal abilities. Instead you get two Bell curves, and THEY OVERLAP. SOME non-MD healthcare professionals (top of their class, many years of experience) do BETTER than SOME MDs (bottom of their class, just out of training) in at least SOME areas. Licensure, and even certification, hides all of this. This is what markets, at least free unregulated markets, quickly reveal.

I just saw my internal medicine PCP for a case of acute conjunctivitis. Even as a radiologist, I could make the diagnosis (though there is never any imaging for acute conjunctivitis.) Nonetheless, despite being a board-certified specialist MD, I had no idea what the latest thoughts are for proper antibiotic treatment of acute conjunctivitis. My IM doc did. He did not have to refer to an ophthalmologist. I suspect most non-MD nurse practitioners could have done equally well, especially since my PCP just glanced at my eye from across the room and wrote the ‘script, without further examination.

The “complexity of modern medicine” that Mingo mentions cannot be denied. What can easily be denied, however, is the implication that doctors cannot possibly ever be overtrained, such that the marginal cost exceeds the marginal benefit. The fact I know the difference between Carney’s triad and Carney’s complex [two rare syndromes that have nothing to do with one another] does not give me a heads up on treating acute conjunctivitis over a nurse practitioner.

I of course admire, indeed, esteem, MingoV for his opposition to professional licensure and wish he and I were not among a small minority of physicians who take this view.

Many thanks to David H for the HT.

MingoV writes:

@Tracy W writes: I think though your experience list is a very good idea, what would you say happens to beginning medical types?

New physicians have years of experience as residents (and sometimes as fellows). That experience can be documented and listed.

New nurse practitioners and physician assistants receive less experience during their training. For these folks, I recommend starting as supervised affiliates of physicians. Once they develop enough experience, they can strike out on their own.

MingoV writes:

@Ross Levatter

I am well aware that clinicians have overlapping spectra of knowledge and experience. I do not expect all GI problems to be treated only by gastroenterologists. But, I also do not expect a nurse practitioner to perform upper GI endoscopic exams.

I did not imply that doctors cannot be over-trained. I stated that training varies from three to ten years. I don't expect family practice physicians to complete a four year residency and a six year fellowship.

My main point is that clinicians of all types do not need to be licensed by the state, and that non-government methods of assessing knowledge, experience, and skills are likely to provide better information than licensure.

Ross Levatter writes:

Well, MingoV, my apologies if I read your stress on the years of extra training physicians receive and the complex nature of modern medicine as implying that this training was beneficial and that doctors have not been, perhaps cannot be, overtrained. I'm sure we agree this training comes with costs as well as benefits.

It is of interest that, at a time when much of general anesthesia is performed by nurse anesthetists (something that two generations ago--when it was less complex--one couldn't imagine anyone but MDs doing), you believe it impossible that nurse practitioners cannot be trained to perform upper endoscopies. Certainly radiology technologists can be trained to do barium fluoroscopy of the upper GI tract as well as radiologists, something known since articles demonstrating that appeared in the late 1960s, but which have been prevented by accreditation agencies for almost half a century, even while RTs doing ultrasound essentially do the entire study on their own and merely present their findings to the MDs, who routinely simply report out the study without ever picking up the US probe.

If you're old enough, you no doubt recall how, before additional hospital regulations put an end to the practice, surgical appliance reps, not MDs but often with a nursing background, would often demonstrate their wares to surgeons by scrubbing up and assisting in the OR, showing the surgeon how to install a new device.

My point in going into such detail, likely boring or unclear to non-physicians, is to demonstrate how many instances of who exactly does what has been determined by happenstance and tradition rather than actual issues of competency ("Can a non-MD be trained to do THAT?") and certainly (and unfortunately) not be markets.

V writes:

[Comment removed pending confirmation of email address. Email the webmaster@econlib.org to request restoring your comment. We have attempted to contact you previously. We'd be happy to publish your comment. A valid email address is nevertheless required to post comments on EconLog and EconTalk.--Econlib Ed.]

Dr.Tom writes:

@Ross Levatter

The vast majority of cases of conjunctivitis does not need antibiotics. Certainly the few cases that do benefit from them can't be diagnosed from across the room.

Not to try to poke a hole in your trust of your internist (kudos to you for having one. Most doctors seem to have just curbside care) but (as I am sure you know) there are a lot of things done in clinical medicine just because 'its always been done that way' rather than because those decisions are evidence based.

Perhaps our licensure and credentialing isn't all it's cracked up to be.


Comments for this entry have been closed
Return to top