Sunday, January 30, 2011

Is Robert Whitaker's book Anatomy of an Epidemic acceptable heresy, or dangerous nonsense?

Over the past year in the world of psychiatry there has been a small hubbub about a book that was released last April entitled Anatomy of an Epidemic: Magic Bullets, Psychiatric Drugs, and the Astonishing Rise of Mental Illness in America. Its author is Robert Whitaker, a science journalist who has an expertise in issues involving psychiatry, having written a previous book entitled Mad In America: Bad Science, Bad Medicine, and The Enduring Mistreatment of the Mentally Ill. The gist of Whitaker's book is that the very treatments used for psychiatric illness (that is, drugs) for the past generation may, in fact, be driving the epidemic in the first place, prolonging and worsening psychiatric symptoms.

Needless to say it is a provocative hypothesis, but for the most part it has not yet taken hold in the collective consciousness in the manner of Rachel Carson's Silent Spring, to which vague comparisons could be made. As of now it ranks only 19th in the category for "Mental Illness," being edged out by such books as Will I Ever Be Good Enough? Healing the Daughters of Narcissistic Mothers as well as The Sociopath Next Door (which particularly sounds like a remarkably silly book). Part of its modest showing has to do with the media coverage, mostly limited to smaller outlets such as Salon and New Scientist; a local NPR station in Boston covered Whitaker's address to the department of psychiatry at Harvard earlier this month. Time had a brief dispatch, and The Boston Globe published a review, somewhat oddly, by a pediatrician specializing in lung and sleep disorders. Without the media flap, the book's sales appear to have faltered.

This last review by Dr. Daniel Rosen is particularly interesting because he finishes his review, which roundly criticizes Whitaker's contentions, by making a dark allusion to the dangers of agreeing with Whitaker's thinking. "Thabo Mbeki [the president of South Africa at the turn of the century] refused to accept that AIDS was caused by the HIV virus, believing instead that it was a side effect of malnutrition and the medications used to treat AIDS itself," Rosen noted. "Those who would seize the opportunity to cast psychiatry as a discipline into the rubbish heap without consideration for the benefits it has brought to so many would do well to remember how Mbeki’s inability to distinguish between theory and fact exacted such an enormous toll in human life and suffering."

It's an interesting little maneuver because, without explicitly saying so, Rosen essentially proclaimed that Whitaker's book belongs to a completely different class of book than the kind a reader would naturally assume he was discussing, which is to say that Rosen was accusing Whitaker of writing dangerous nonsense rather than acceptable heresy. And the two categories are as different as they could be in the world of science and medicine.

The typical mechanism of scientific progress relies on the establishment of heresies. Let's say we have some model about how stomach ulcers come into being: people think that it's related to stress, and doctors who read the literature prescribe a bland diet and tell people to chill out. Then along comes someone who's been doing biopsies of patients with ulcers and reports a new bacteria that's pretty much impossible to grow, but is clearly there, and thinks that it's this weird new bacteria, and not a stressful lifestyle, that is the cause of ulcers. This someone--after doing further research on his little bacterium--writes papers, gives talks at various conferences, and essentially argues with the scientific establishment that he's right and the old way of thinking about ulcers is wrong. Eventually, enough people are convinced by his data and adopt the new model. What I've just described actually happened and involves the discovery of the bug Helicobacter pylori by Dr. Barry Marshall in Australia in the early 1980's; by the late 1990's the H. pylori model was universally accepted. It's what you'd call standard, acceptable heresy, and science thrives on such a process.

The problem that very simple-minded people frequently are unable to appreciate, however, is that just because science works by overturning conventional beliefs, it doesn't mean that anything goes, or that just because some theory isn't shared by mainstream scientists, it must by definition have some merit. So a lot of writers and thinkers of science distinguish this form of thinking from "acceptable heresy" by regarding it as "dangerous nonsense." In the world of science, the two most dangerous forms of nonsense is that a) evolution isn't really true, and b) man-made global warming doesn't exist. In the world of medicine, the two would be a) vaccines cause autism, and b) HIV is not the cause of AIDS.

Which brings us back to Dr. Rosen's comparison of Robert Whitaker and AIDS denialists: it's Rosen's way of telling people that they shouldn't even bother, that Whitaker's work is not acceptable heresy, but rather dangerous nonsense. And Whitaker got the point of that right away: he posted a same-day reply in his blog at Psychology Today decrying that particular rhetorical trick, calling it "a bit over the top" in what wins points for remarkable restraint.

The question is: who is right? Should we agree with Dr. Rosen, think of Whitaker as a purveyor of dangerous nonsense, and throw it on the ash heap of silly attacks on medicine, or should we give Whitaker's argument a wide berth and read his book with care? For me, this is not an idle question as I haven't read the book, and I've been wondering whether I should put it on my reading list, and if so, how high should it go?

The answer is that I think Whitaker's probably solidly in the camp of acceptable heresy (which doesn't mean I think he's right, only that I am taking his contentions quite seriously). Anatomy Of An Epidemic is going on my reading list, and may go pretty high, for at least three reasons:

a. Whitaker clearly did his homework. One doesn't need to have a Ph.D. in statistics, or be a psychiatrist, to see that Whitaker researched his subject thoroughly, and for me that counts a great deal. Far too often people hostile to science rely on clinging to single studies as proof of their rightness. Whitaker sought to undertake what appears to be a comprehensive survey of the literature and ask the singularly heretical question is all of our modern therapy making a difference for the better? But even a causal glance at the reviews makes clear that he has done due diligence in trying to assess what the entire field knows. It's entirely possible that his interpretation of the data is flawed, as psychiatrist-blogger Dan Carlat thinks (part 1 here and part 2 here), but he isn't cherry picking data in order to further his pet theory, an intellectual strategy that drives me nuts.

b. Studying outcomes in psychiatry is a tricky business. One of the reasons why AIDS denialism is dangerous nonsense is because it's really not that hard to follow the progression of HIV viral load through sickness and death. In other words, outcomes are pretty clear. I have personally witnessed what happens to some patients who have stopped taking their antiretrovirals (small, nonstandardized nonexperimental data to be sure, but persuasive data nonetheless). In other fields, like cardiology, studying the effect of this-or-that drug is comparatively easy, because you are often measuring death as the main outcome, which as outcomes go is pretty fixed. In psychiatry, by contrast, outcomes are much more slippery: "feelings" are notoriously hard to standardize, and even in more severe cases such as schizophrenia, I remain skeptical that you can easily and cleanly reproduce results in clinical trials. As a consequence, I'm inclined to listen to multiple interpretations of effectiveness data, including those that question the value of the entire enterprise--and I'm inclined to listen to it in a way that I wouldn't if I hear an attack on HIV meds, for which there is far too much data to suggest anything other than their effectiveness.

c. Other people whose opinions I respect think highly of him. It's of course the weakest of the three reasons, but still counts for something. Particularly when one has to deal with sorting garbage from gems, it helps to have a trusted figure offer their two cents. E.g. Carlat: "[Anatomy Of An Epidemic] is the work of a highly intelligent and inquiring mind--a person who is struggling to understand the nature of psychiatric treatment.  Put it on your reading list, and join the debate."

It is, and I will.


  1. As one who was lucky to see Mr. Whitaker personally speak and who has read his book, I can't recommend it highly enough. I am sure alot of people will be greatly interested in your thoughts after you read it.

    That is a very interesting comparison to Dr. Barry Marshall. By the way, he treated one of my relatives who unfortunately had problems with the meds he prescribed which was definitely not his fault.

    Actually, it sounds like according to this site, he endured alot of hardship before his theories were accepted:

    Sadly, even if someone with greater credentials than Mr. Whitaker came up with studies like Barry Marshall did regarding the issues raised in Anatomy of an Epidemic, I doubt they would be accepted.

    Psychiatry is simply too entrenched in the meds come heck or high water philosophy in my opinion.

  2. Thanks for your comments, Anon. One thing that I think is pretty clear after reading about Mr. Whitaker is that he's got a serious amount of street cred...or whatever the equivalent of "street cred" is among the world of journalists.

    Barry Marshall is but one example in a line of researchers whose theories were highly unorthodox at the outset and have since become the standard; anyone familiar with Judah Folkman's history will know how his thoughts on angiogenesis (how new blood vessels form as a key component of cancer tumor growth) were regarded as something approaching outright lunacy when he first proposed it in the early 1970s, but is now accepted. I could go on in this vein for some time. Most scientists are aware that this is the case and is one reason why we're cautious as hell when we think we've discovered something important (and is also why I, for one, am immediately skeptical when a scientist "goes public" with some discovery before submitting to the usual process of weeding the idea out, cf. Cold Fusion in the late 1980s).

    Comparing Whitaker to researchers like Marshall isn't quite fair to either party, since Whitaker isn't proposing a new mechanism about how biology works (as Marshall did), and Marshall wasn't calling for an intellectual overhaul of his profession (as Whitaker argues needs to happen in clinical psychiatry). It's a matter of perspective: Marshall may have had a radical idea but he was still very much part of the everyday scientific system; Whitaker is an outsider questioning the entire foundations of a subspecialty of medicine.

    As it turns out, this has happened before in psychiatry, and quite recently, i.e. standards of Shrink practice in the 1950's would be considered barbarous today. In my field of Infectious Disease, though, I can go all the way back to the 1920's and earlier and see that my professional forbears shared pretty much the same assumptions I do, and their methods can only be questioned on the level of technology and whether the results were adequate. That's just not so in psychiatry just yet, and while my sense is that we've made huge advances in the last 20-30 years at understanding and treating psychiatric conditions, I also think it wouldn't be totally surprising to learn over the next 20 years that future physicians will regard the kind of meds that I prescribed as a student and resident little better than the lancets used by bloodletters in the 18th century. Which is, boiled down, why I am going to be reading his book.

  3. I've just read Whitaker's remarkable and impressive book. Then, I decided to spend an hour or so researching what critiques or opposition he has out there among psychiatric circles.

    There is none. None that is legitimate and based on any credible counter-evidence, that is. There are lots of psychiatrists floundering around talking vaguely about how meds "help people," but there is absolutely no solid evidence to rebut what Whitaker is saying. That's because he's right.

  4. Thanks for your thoughts, Laura, and apologies for the delay in posting your comment.

    As a clinician and (still so far) a scientist, I am a little gun-shy about using sweeping statements like "Whitaker's right and none of his opponents' critiques are legitimate", so I'd frame things a little differently.

    I think his arguments are certainly provocative and that there are a lot of hard questions that psychiatry and psychiatrists should be asking about their profession. I also think that many of the critiques of Whitaker are not well thought out.

    That said, not all of Whitaker's arguments are spot-on; for instance, Dan Carlat's review of Anatomy notes--I think correctly--that one of Whitaker's key arguments (that the expansion of psychiatric illness in this country coincides with the expansion of psychiatric meds, and that the latter has caused the former) is weak. But like Carlat, I think that Whitaker is asking some very important questions about the uncritical acceptance of a biopharmacological model of neuropsychiatry that may be a lot of smoke and mirrors.

    My field is infectious disease, where most, though not all, diseases can be studied more easily because they have clear outcomes. We learned about the magic of penicillin and sulfa because people who previously died managed to survive nasty infections. It's a lot harder to measure subjective feelings over the course of depression, and harder to know how long to treat depression, and in which people depression treatment is most effective. So Whitaker asks crucial questions. Based on what I have seen, there's a moderate chunk of docs out there that chuck out prescriptions for SSRIs without much thought for how long they want to treat their patients, and by what measures should they adjust--or stop--their treatment.

    Best, Billy