The Book of Woe: The DSM and the Unmaking of Psychiatry

by Gary Greenberg

  “The controversy stirred by my critique7 of DSM-5 is a terrible moment in the history of psychiatry,” he told me the month after the annual meeting. “This is the worst thing to happen to the field’s credibility since Rosenhan—and psychiatry is a field that especially requires credibility to be effective. I know I have done grave harm.”

  Frances reasoned that the damage to misdiagnosed and overdiagnosed patients was a graver harm than undermining psychiatry’s credibility with the truth. It was the kind of calculation Cincinnatus might have made, hoping to hasten his return to the farm. Frances had gambled that the fragile edifice Spitzer had erected and he had reinforced would withstand the weight of the truth, that one of the guardians of the noble lie could reveal it and yet somehow preserve the authority the lie had purchased. And even as he fended off more antipsychiatrists drafting him unwillingly into their cause and more attorneys eager to use his own criticisms to undermine his (and his profession’s) credibility, he continued to be certain he’d made the right choice.

  It’s possible he was compelled by unconscious inner necessity to blurt out the truth, or that contrition or self-loathing or that old Freudian notion, the death instinct—the inbuilt yearning for the chaos that the lies of civilization, noble and otherwise, hold at bay—drove him. He would say that what he did was much simpler than that, that it grew from an easy calculus, nearly bureaucratic in its plainness: that the only chance to preserve the DSM’s hard-won authority was to stop the APA from going ahead with the worst of its ideas—especially those, like removing the bereavement exclusion, that would badly cashier the reputation of his beloved profession. He would also say that it doesn’t require vast sophistication to grasp the reality: that a language by which two doctors can agree on a name for a patient’s subjective suffering is a signal achievement no matter how contrived, and worth preserving despite its many flaws.

  In this, he may have overestimated the value of that language. He may also have overestimated the tolerance of Americans for bullshit. But above all, Allen Frances may have overestimated himself.

  Chapter 20

  Or maybe it was just me he overestimated.

  In October 2012, I join Frances and Manning in a hotel room near Harvard Medical School, where Frances is scheduled to address a bioethics seminar. A documentarian working on a film about the DSM is setting up her equipment. She has finally caught up with Frances after a four-month chase. He has decided that she should interview the two of us together. It’s not entirely clear to me if she is on board with that idea. Since June, Frances has mostly been quiet about the DSM. He is still blogging for The Huffington Post and Psychology Today and the Psychiatric Times, where he has weighed in on gun control and the presidential election and offered “to stop being an amateur columnist” if David Brooks would “stop being an amateur psychologist.” But the DSM has never been far from his mind, and as soon as the lights are on and the camera is running, he is back to it and drawing me into his explanation of all that has gone wrong with the DSM-5. I may be an upside-down Jesuit and he a world-weary rationalist, but for the moment, we’re just a couple of friends on the inside of the same joke. The filmmaker seems entertained, although it’s possible she is simply egging us on in hopes of capturing some outrageous Francesism on film. But he’s become more careful. In fact, he tells the camera, he’s learned his lesson, the one about how impertinent remarks might, in the wrong hands, turn his attacks on DSM-5 into attacks on psychiatry.

  The director describes an Internet video she has seen, put out, she says, by Scientology’s Citizens Commission on Human Rights, in which the narrator somberly intones that even the head of the DSM-IV thinks psychiatry is bullshit. Frances looks over at me, vindicated.

  The thing I don’t understand, he tells me (and I’m working from memory here; I didn’t tape our meeting), is that you think the words in the DSM are capable of great harm. So why aren’t you worried about the harm your words can do?

  The question makes me think of the infamous line with which Janet Malcolm opened her book The Journalist and the Murderer. “Every journalist who is not too stupid or too full of himself to notice what is going on knows that what he does is morally indefensible.” Malcolm was writing about the way Joe McGinniss had seduced and betrayed Jeffrey MacDonald, promising exoneration and then penning indictment. But Frances is accusing me of more than luring him into candor with assurances that my criticism of psychiatric diagnosis was tempered by a recognition of its uses, and then using him as a cudgel in my own crusade against psychiatry. Indeed, he has always insisted he doesn’t care about his own image, and while that may be a little too much protest, it’s not hard to believe that his worries are genuine, and that my real betrayal is using his comment to harm the people who need to have confidence in their doctors (and keep taking their drugs) to get better. If the discrepancy between opportunity and knowledge remains under wraps, it seems, that’s not bullshit. That’s wisely deploying the placebo effect. That’s medicine.

  The camera is rolling, or whatever it is that digital cameras do. I’m pinned and wriggling, scrambling for some way to explain why I don’t seem to care about what happens when people glimpse what’s behind the curtain. It isn’t the first time a psychiatrist has warned me that criticizing the profession would lead to dire consequences. It’s the profession’s stock response to anyone who attacks it, and I have a stock rebuttal: that I am sure more people have been hurt by the DSM, or at least by the treatments that follow diagnosis, than by anything I ever wrote. Yet it seems inadequate to the moment. I say some words, but they don’t really make sense, and they surely don’t answer his question. Frances sits back in his chair.

  He has the right to his satisfaction. It is true that I didn’t give a moment’s thought to the question of whether reporting Frances’s comment (along with a lot of evidence that he is right) would hurt anyone. I always figure people are better off with the truth, which is probably why I went into both the therapy and the journalism businesses—and why I get angry when one of those professions hides its own uncomfortable truths. But as much as I like the way that sounds, maybe I’m just too full of myself to see that I’m using Frances and the patients, that they have become character and audience, and that I’m using truth as well, not as a virtue but as a narrative device, as the MacGuffin for exposing humbuggery and chronicling comeuppance, and that to undermine the already shaky foundations of a profession that offers the last and only hope for some patients—that has succeeded, at least in some cases, at quelling their hallucinations, modulating their mood swings, allaying their anxiety, and restoring them to some semblance of normal functioning—and to bring low the confidence man at the expense of his potentially satisfied customers is simply indefensible.

  • • •

  But then again, so is psychiatry, at least when it comes to the DSM. And not because the DSM-5 was botched or because the profession is a cabal of Pharma collaborators, although it harbors its fair share of both incompetents and conspirators, but because even at its best, even in the view of honest and eloquent men like Steve Hyman and Allen Frances, psychiatric diagnosis is fiction sold to the public as fact. And not the Supreme Fiction that Wallace Stevens says begins “by perceiving the idea / Of this invention, this invented world,” but a fallen fiction whose authors, if they are to hold on to their power, must insist that they have gathered together the scattered particulars of our suffering and sorted them according to their natural formations, even as they harbor the knowledge that they have done no such thing. That knowledge can be locked up, like Leibowitz’s Memorabilia in its monastery, but it will always escape when the DSM is opened for revision and doctors once again argue over matters that their science cannot settle.

  Later that day, Frances is once again called upon to defend psychiatry against his own charges. In an elegant wood-paneled room at Harvard, just after he has told a group made up mostly of doctors why expert consensus—the method that has
yielded the DSM—is both necessary and dangerous to public health, Arnold Relman, the eighty-nine-year-old former editor of The New England Journal of Medicine, professor emeritus at Harvard, and a longtime critic of for-profit health care, suggests that this tension is worse in psychiatry than in other specialties because psychiatric experts lack biological findings that can anchor diagnosis in something beyond the symptom. Where is psychiatry’s pneumococcus, Relman seems to be asking. Frances has fielded this question before, and he has a ready answer: that diagnostic uncertainty and lack of treatment specificity haunt all of medicine.

  It’s a version of the argument Steve Mirin and Darrel Regier once made to the editors of The Washington Post, and it was not entirely wrong. There are plenty of illnesses that are described purely in terms of their symptoms—chronic headache, for example, or idiopathic neuropathy—and devastating diseases, such as multiple sclerosis and cancer, that seem unlikely to have a single form caused by a single pathogen like pneumococcus. And while it is true that those diseases are often diagnosed by lab studies, if only to rule out other known causes of their symptoms, psychiatry is still not so different from other specialties in this sense.

  But even if medical nosology, taken in the aggregate, is as fictive as psychiatric nosology, even if many of its diagnoses are merely descriptions of the problem in a medical language, still it would have its pneumococcus and its polio and its diabetes, not to mention its heart diseases and bone fractures, its blood counts and biopsies and X-rays, its antibiotics and vaccines, its cobalt-chromium stents and titanium joints, its brain surgeries and organ transplants. Even if its unknowns far surpass its knowns, medicine undeniably has its slam dunks. Even when they are found by accident, as they often are, and even when they seem miraculons, as they often do, these are not miracles or mere serendipities, but the discovery of the natural laws that govern our suffering. Medicine’s sure knowledge of those laws saves our lives and earns doctors our deference.

  This is precisely what psychiatry lacks. Without a single mental disorder that meets the scientific demands of the day, let alone enough of them to make the DSM more than an invented world, and with its claim to “real medicine” still mostly aspirational, it cannot make good on its assertion that psychological suffering is best understood as medical illness. So it must guard its position jealously. Lacking confidence in itself, psychiatry must work ever harder to command ours. This is what unites the APA, with its circle-the-wagons paranoia, its deceptions and duplicity and tortured language, and Allen Frances, with his invocations of Leibowitz and his warnings about patients gone wild. He and Darrel Regier may be bitter opponents, but they both have the fear that comes with knowing the fragility of the edifice they share.

  • • •

  The APA had at least one opportunity for a slam dunk in the DSM-5. In an article published in The American Journal of Psychiatry, an international group of seventeen prominent men—including clinicians, psychoanalytically minded personality theorists, historians of medicine, biological psychiatrists, critics of biological psychiatry, and Bob Spitzer—urged the DSM Task Force to include in the DSM-5 a disorder they called melancholia. “Melancholia,” they wrote1, is “a syndrome with a long history and distinctly specific psychopathological features.” Melancholia is Winston Churchill’s black dog, Andrew Solomon’s noonday demon (an image he borrowed from Isaiah), William Styron’s darkness visible—a form of depression noted by doctors since Hippocrates and characterized by an unshakable despondency and sense of guilt that arises from nowhere, responds to nothing, and dissipates for no apparent reason.

  The authors drew on thirty years of research to describe five clinical characteristics by which melancholia could be distinguished from other kinds of depression. Among those characteristics were biological findings that set the melancholics apart, notably hypercortisolemia and disturbances in sleep architecture. A sleep study could show whether or not patients had the reduced deep sleep and increased REM time characteristic of melancholia. And a dexamethasone suppression test (DST)—in which patients were given a synthetic steroid to see if it suppressed the activity of their own hormonal system—could determine whether their cortisol, a stress hormone, was in overdrive. Patients who meet the criteria for melancholia are much more likely than other depressed people to show this abnormality. They are also much more likely than other depressives to respond to two treatments: tricyclic antidepressants (drugs discovered in the late 1950s and in wide use before the Prozac era) and electroconvulsive therapy (ECT), better known as shock treatment, and they show less response to both placebos and cognitive-behavioral therapy. Melancholia, the proponents concluded, was a “distinct, identifiable, and specifically treatable2 affective syndrome.” It might even be, although they didn’t say this, a type of depression that actually was the result of a real chemical imbalance, a disorder onto which our biochemistry could be mapped.

  The proposal included plenty of standard scientific evidence—clinical and lab studies, case histories, literature reviews—and, with its tie to cortisol, melancholia seemed to fit in with emerging theories about depression and stress. So you would think that the APA would have leaped at the opportunity to finally prove to dismissive doctors in other specialties and to a skeptical public that, at least in this one case, psychiatrists were real doctors treating real diseases that could be discerned with real tests and treated with real cures.

  But you would be wrong. Melancholia not only failed to gain inclusion, it was not even given much consideration. Only five days after he had received the group’s proposal, in October 2008, mood disorders work group member William Coryell was already telling Max Fink, one of melancholia’s main proponents, that the odds were very long. The main obstacle was exactly what Fink and his colleagues thought was one of the great strengths of the proposal: the biological tests, especially the DST. “I believe the inclusion of a biological measure3 would be very hard to sell to the mood group,” Coryell wrote Fink—and not because the test was unreliable. “I agree there is more data to support using the DST for melancholia than for using any other measure for any other diagnosis,” he conceded. Even so, the DST would be “very hard to sell since it would be . . . the only biological test for any diagnosis being considered.” Coryell didn’t finish the thought, but the implication was obvious: a test for melancholia would make the lack of biological measures elsewhere in the DSM that much more glaring. It was a success that would only highlight the APA’s failures. (Coryell declined to comment.)

  Sixteen months later, when the APA posted its first draft of the DSM-5, Coryell was proven correct. Melancholia didn’t even show up in the mood disorders section; it had merited only a single line in a section of “conditions proposed by others”—a category it shared with Parental Alienation Syndrome and Male-to-Eunuch Gender Identity Disorder, among others.

  “I [am] flabbergasted4 that our suggestion . . . has been excluded from consideration,” Fink wrote to Coryell. “Carving out a well-defined type of mood disorder, one that carries with it the promise of homogeneous samples and optimized treatment outcomes, is a small step in the development of the classification, but it is one that has been extracted from Nature grudgingly, and deserves greater attention and consideration within . . . DSM-5.”

  “I believe you and your colleagues5 are fundamentally correct,” Coryell replied. But, he added, his belief had “not been shared by any of the other work group members,” so there was “no point in pursuing it further.” Coryell ventured a new explanation for the outcome: that the proposal would “entail a fundamental change in the boundaries” of a diagnosis (MDD) that was “among the most enduring and stable” of the DSM’s categories. “Evidence for such a sweeping modification would need to be quite extensive and compelling.”

  But an ambiguous research record hadn’t stopped the work group from gerrymandering the bereavement exclusion out of MDD, nor was it stopping other work groups from considering destabilizing changes li
ke the removal of Asperger’s or the introduction of entirely new diagnoses like DMDD. On the other hand, those diagnoses had one advantage over melancholia: they didn’t threaten to introduce a biological measure into the DSM and make the rest of the book look bad in the bargain. Offered a key to one of the cells of its epistemic prison, the APA had decided that the cost of freedom was too high.

  • • •

  In June 2012, the APA posted a change to the permissions policy on its DSM-5 website. “The APA owns all products6 generated by the Work Groups developing DSM-5,” it declared. This included, they asserted, not only proposed criteria, but also the discussions that led to the work groups’ decisions. To those who wondered how this squared with the insistence that this was the most transparent DSM ever, the APA issued a reassurance: “Requests will be considered for permission to describe the criteria and development process in narrative form.” The organization wasn’t trying to erase history, only to control it.

  There was no explanation of this change. It was hard not to think that it had something to do with my having shown up at the annual meeting and peppered the presenters with questions, from which the APA’s communications experts could only conclude that their embargo had not stopped me from writing my book.

  On the other hand, the new policy also came shortly after Allen Frances put his scathing description of the development process on the op-ed page of The New York Times. The APA, he charged, was guilty of “arrogance, secretiveness7, passive governance and administrative disorganization.” It had failed to rein in its experts, who had in turn (and predictably) manufactured new disorders, heedless of the fact that “new diagnoses in psychiatry can be far more dangerous than new drugs.” And now he had come to a reluctant conclusion—that the APA “is no longer capable of being sole fiduciary of a task that has become so consequential to public health and public policy,” and that it should be stripped of the diagnostic franchise.