Why the big fuss over Donald Berwick and integrative medicine?

And besides, what's wrong with a little quackery now and then if patients want it? Those are the questions posed by Shannon Brownlee over at The New Health Dialogue in a post written in response to Kimbal Atwood's expose of Berwick's apparent fondness for unscientific forms of complementary and alternative medicine. Several of her assumptions and accounts of facts are fundamentally flawed. Let's take a closer look at the post.

Early on she says this about Atwood's post:

In Dr. Atwood’s view, a February 2009 appearance by Dr. Berwick at Bravewell Collaborative-sponsored Summit on Integrative Medicine and the Health of the Public, where he shared the podium with Dr. Mehmet Oz, Dr. Dean Ornish, Senator Tom Harkin, was evidence that he was either naïve or had “gone over to the dark side."

The “dark side?” Don Berwick?? C’mon.

OK, let's not judge Berwick by the company he keeps. But there's much more than his mere appearance at the Summit on Integrative Medicine to raise concern about a fondness for, or at least an openness to quackery on his part. His remarks at the Summit and elsewhere are what should be cause for concern. At the Summit he spoke uncritically about homeopathy and acupuncture and said that he viewed integrative medicine as a “distinguished and important new arena.” On another occasion Berwick said that evidence based medicine may have to take a back seat if patients' preferences so dictate. What more do you need?

She goes on:

More to the point, while alternative medicine is undoubtedly mostly quackery, so is a good bit of allopathic medicine, at least if you define quackery as medical practices that are not based on valid evidence.

She's confused. She's confusing quackery with non-evidence based medicine (it's not the same thing) and poor uptake of best evidence (again, it's not the same thing). After some misinformed questions from commenters I once had to explain this. Here is an excerpt from that post (where I used the word woo to denote quackery):

Frequently asked questions on woo and mainstream medicine

“Woo” is a term for certain implausible and outlandish claims of complementary and alternative medicine (CAM). The term has recently been popularized on the blogosphere by Orac and others and has been a subject of several recent posts of mine which have drawn numerous criticisms and questions. Rather than address them piecemeal in my comment threads I decided a more effective way to answer my detractors would be in the form of a frequently asked questions (FAQ) post, so here goes.

Many of mainstream medicine’s conventional treatments are not evidence based. Aren’t they a form of woo?

No. Although some conventional methods fail to measure up to best evidence they are at least based on known anatomy and physiology. They have some plausibility in the observable biophysical model in contrast to the “vital forces”, nebulous “energy fields” and “non-local powers of the mind” which are characteristic of woo.

You seem to focus a lot of your criticism on woo. Isn’t non-evidence based conventional medicine harmful too?

Yes, of course. Moreover, there are harmless forms of woo just as there are harmless conventional breaches of EBM. All departures from best evidence are problematic and need to be addressed, whether woo-based or not.

Then why make a distinction?

Because of important differences in the ways the problems manifest themselves. Mainstream medicine applies a double standard and that’s what I’m trying to expose. People in the mainstream are appropriately critical of conventional deviations from best evidence and are trying to correct the situation. But due to the nature of the problem---a complex interplay of system and cognitive failures---the fix is not easy. In contrast (and here’s where the real hypocrisy comes in) mainstream medicine uncritically embraces woo, applying to it a much easier evidentiary standard and often no standard at all. The remedy for the problem of woo would be much simpler, too. Mainstream medicine could simply say no. Woo, by definition patently implausible, is easy to spot. There’s nothing complicated about it. It’s not a system problem. It’s there in mainstream medicine purely by choice. That fact raises another important distinction. If mainstream departure from EBM is a complex system problem and woo is there by choice then woo constitutes a more serious ethical problem.

The point I was trying to make there is that from both a policy and a purely scientific standpoint, while there may be areas of overlap, there are important distinctions between quackery and non-evidence based medicine. The problem of poor adherence to EBM is very real in mainstream medicine, but it is a complex mixture of inefficient dissemination of knowledge and external barriers---barriers that leaders in mainstream medicine, some of them anyway, are struggling to overcome. This is quite a different situation from those (and, to complicate matters, there are many in mainstream medicine) who knowingly and willfully promote non-evidence based claims without regard to any scientific foundation.

Brownlee's examples of non-evidence based mainstream medicine are misinformed. She writes:

Just look at the recent history of widespread medical practices that turned out not to work -- or worse, not only did they not work, they also harmed patients. High dose chemo for breast cancer comes to mind. And don’t forget lidocaine for cardiac arrythmia and cardiac stents to prevent heart attacks. Bee pollen and homeopathy may turn out to be pseudoscientific bunk, but at least for the most part they’re not as dangerous as some widespread medical therapies that aren't based in sound science and valid evidence.

First off, I'll give her a pass on breast cancer (not my area of expertise by any means; I'll leave that to Orac if he chooses to weigh in). Lidocaine I know a little about. It certainly did work. It's still an alternative in the ACLS algorithms. Prophylactic lidocaine for routine use in myocardial infarction has been abandoned but it did work, in what it was purported to do. Here's the classic paper of Lie and colleagues demonstrating that it when lidocaine was used prophylactically in MI patients it decreased the rate of V fib from 9% to zero. Not only did it work but it worked extraordinarily well. That finding has never been refuted although lidocaine was gradually abandoned for this use for complex reasons. Beta blockers came along. Lidocaine could not be demonstrated to decrease mortality because defibrillation was universally effective. There were side effects---not life threatening when the drug was used properly but they were there. By the mid 80s we were in the reperfusion era, making primary V fib less of a threat.

And cardiac stents to prevent heart attacks? What's she talking about? There are well defined benefits of coronary stenting but no one with any authority in mainstream medicine claimed that stents prevented heart attacks. (With apologies to Orac, holy strawman fallacy, Batman!). In fact, about the time stents came into common use in the late 1980s mainstream thought leaders were saying just the opposite. We already knew by that time that most heart attacks were due to rupture of mildly stenotic plaques, lesions which no one would ever consider targets for revascularization.

To be completely fair, Brownlee could have come up with examples that would have better supported her claim about the shortcomings of mainstream medicine. There was the use of oral antiarrhythmic drugs to suppress ventricular arrhythmias which turned out to be lethal (class Ic drugs, as demonstrated in CAST), not worth the side effects or extracardiac toxicity (tocainide, mexilitine) or not as good as device therapy (amiodarone). The problem here was not that the drugs didn't “work” but that over-reliance on surrogate endpoints was hazardous, even while the recommendations were based on the best scientific principles and evidence then known. CAST was conducted because investigators were aware of proarrhythmia but didn't know it balanced out against potential benefits. The study was completed and lessons were learned. You will never see that kind of scientific progress with any of the forms of woo such as homeopathy or acupuncture.

As to her point that fairy dust at least is not dangerous? Of course it's not, because it does nothing in the patient's body. But that misses an important point. No one that I know of, certainly not Atwood, probably not even Orac, would deny patients the right to avail themselves of harmless quackery. It's quite another thing, though, for supposedly reputable and scientific organizations like the Institute of Medicine, or the leader of the IHI (and now CMS) to endorse it, or, worse, for a government health plan to pay for it.

Brownlee does go on to make a valid point: we in mainstream medicine don't do a good enough job of informing patients about the evidence. And I'm not trying to make excuses but there are external barriers, including time and economic constraints, that make that process very difficult, as I demonstrated here and here.

She concludes:

When it comes to the ethical duties of physicians, do we really need to fret about low risk alternative treatments? It’s probably more important that doctors and patients share decisions, and that we make sure that patients have the tools they need to make their own choices about any treatment, whether it's alternative or mainstream medicine.

No, we don't need to fret about alternative treatments---at least we shouldn't fret about patients seeking them out for themselves, as long as they're harmless. But we should be outraged at the ethical failure of purportedly scientific, esteemed mainstream organizations like the Institute of Medicine, mainstream thought leaders and our government promoting pseudoscience and quackery.

Oh---and for another expose of Berwick's fondness for woo and a brilliant analysis about what's really, really wrong with his extreme consumerism see this post from David Gorski at Science Based Medicine.