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

by Gary Greenberg

  Frances acknowledged that there was no obvious immediate successor, but he suggested the Department of Health and Human Services, the Institute of Medicine (a section of the National Academy of Sciences), or even the World Health Organization. He mentioned one obvious contender, the National Institute of Mental Health, only to dismiss it as “too research-oriented and insensitive to the vicissitudes of clinical practice.” Since sponsoring the planning conferences at the beginning of the DSM-5 effort, the NIMH had indeed remained on the sidelines—not because nosology was out of its bailiwick, but because the institute had lost faith in the DSM. “Our resources are more likely8 to be invested in a program to transform diagnosis by 2020,” NIMH director Thomas Insel told me, “rather than modifying the current paradigm.”

  The NIMH is placing its bets for a new paradigm on a program it calls the Research Domain Criteria (RDoC), a name that recalls the Research Diagnostic Criteria, the Washington University initiative that led to the transformation of diagnosis in the DSM-III. Just as Spitzer and his colleagues had been confronted with widespread dissatisfaction with the diagnostic system among psychiatrists, so too had Insel heard the discontent among his peers on his frequent trips to hospitals and universities around the country.

  What Insel heard “over and over again9” on his tour was that psychiatrists were tired of being trapped by the DSM. “We are so embedded in this structure,” he told me. He and his colleagues had spent so much time diagnosing mental disorders that “we actually believe they are real. But there’s no reality. These are just constructs. There’s no reality to schizophrenia or depression.” Indeed, Insel said, “we might have to stop using terms like depression and schizophrenia, because they are getting in our way, confusing things.” Thirty years after Spitzer burned down the DSM-II and built the DSM-III in its ashes, psychiatry might once again have to “just sort of start over.”

  Spitzer’s error, at least according to Bruce Cuthbert, the NIMH psychologist in charge of RDoC, was not that he tried to cram psychological suffering into faux medical categories, but that he continued to think of suffering as a function of the mind. “So many of our disorders10 historically have been conceived of as disorders of mind,” Cuthbert says. This leaves scientists in an impossible position. “As scientists we have to measure things,” he says. “What else can we do?”

  But the attempt to measure the mind has led to nothing but dead ends like the old, now discredited theories about depression and serotonin. “There was going to be a one-to-one map between our putative mind diseases and biology,” says Cuthbert. “Whoops! It wasn’t that simple.”

  The way to start over, according to Cuthbert and Insel, is to forget about the mind and look directly to the brain for understanding our suffering. We shouldn’t take the fact that people have been describing disorders of the mind such as melancholia for more than two thousand years or schizophrenia for more than a century as evidence that those diseases exist and then try to find them in the brain, Cuthbert said. Instead we should ask, “What does the brain do? What did it evolve to do? And we know that now.”

  What we know now, Cuthbert explained, is that “there are very specific circuits in the brain that perform somewhat specific things.” Previous attempts to map the brain onto disorders of the mind failed not only because they were looking for mind disorders, but also because they had the brain’s role in psychopathology wrong. The trouble didn’t originate in individual brain structures like the hippocampus or frontal cortex. Neither was it in droughts and floods of neurotransmitters. Rather, it was to be found in circuits of neurons, the pulsating networks that experience builds in the brain, each their own little ecosystem in the vast electrochemical jungle between our ears. Understanding circuits, or even knowing that they existed, wasn’t even possible a generation ago, but now, thanks to MRI, PET, and other brain-scanning technologies, “we know there are circuits for fear,” Cuthbert said. “We know there are circuits that guide us to approach things that are desirable and to go get them, like the food that we need for nourishment. We know there are circuits for memory. So we know something about the organization of these circuits now, and we have an idea that these circuits are involved in lots of different disorders.”

  Cuthbert doesn’t expect neural circuits to map onto the DSM disorders any better than neurotransmitter metabolism has. But then again, this may not matter. If, for instance, researchers can trace the neurocircuitry of the startle response, figure out all the electrochemical events that make an animal blink and hunch and shrink away from a sudden noise, then they will be on their way to understanding the anxiety found in many DSM disorders. The arousal itself having been elucidated, it will no longer be merely a scattered particular waiting to be gathered under the correct diagnosis. Indeed, fictive placeholders will no longer be necessary. They will be replaced by the natural formations that the brain scanners have detected as the sources of this particular kind of suffering.

  Cuthbert pointed to a chart titled “Draft Research Domain Criteria Matrix.” Its rows list the five natural formations the NIMH is interested in: negative valence systems, including threat, fear of loss, and frustration; positive valence systems, including motivation, learning, and habit; cognitive systems (attention, perception, and memory); systems for social processes (facial expression identification, imitation, attachment/separation fear); and arousal/regulatory processes (stress regulation). Its columns are eight units of analysis—such as genes, molecules, and cells—which the NIMH would like researchers to use to investigate the domains. So, for instance, a scientist interested in working memory (a cognitive system) might want to look into the dorsolateral prefrontal cortex, while a researcher in the negative valence domain could propose a study about the hypothalamic-pituitary axis or the bed nucleus of the stria terminalis or corticotrophin-releasing factor. Ultimately, so Cuthbert and Insel hope, the matrix will be filled in with knowledge about these domains, and the neural substrates of the distress of people with attachment/separation fear or difficulties in regulating stress will be elucidated, pinpointed, and presumably targeted for treatment, without any need for recourse to putative mind diseases.

  Cuthbert was not clear about whether it is the mind that is putative or only its diseases, although the fact that most of this research is going to take place on animals is a clue to the relative place of the human mind in this scheme. And, indeed, it is hard to see how the idea that human consciousness is something more than the sum of its parts—an idea that, however muted, still lingers in the notion that a state of mind such as depression can be something real, something that surpasses and unites its scattered particulars—can survive an effort such as RDoC. Not that the program signals the death knell of the self (and not that psychiatry, for all its influence, could slay the idea of human agency that has developed over the last five thousand years or so), but it does seem to signal the profession’s intent to complete its abandonment of the mind as the location and source of our suffering, an effort that began in earnest when Spitzer kicked psychoanalysis out of the DSM and that might end as it turns to circuits and systems whose primary virtue is that they can be measured. It seems to signal a future in which diagnosticians will let the brain talk in its own language of inputs and outputs, of ganglia and dendrites, of myelin sheaths and afferents; tell its owner what it is saying; and then provide treatments that are no longer targeted at mythical chemical imbalances or fictive disorders but at the faulty circuits that are causing distress.

  At the APA meeting in Honolulu, Insel laid out his vision of that future. He showed an animation illustrating the differences in brain development between children with ADHD and normal kids, differences that he said originated very early in life. “We call this attention deficit/hyperactivity disorder11,” he said, but “think about it for a moment. Attention, that’s cognition; hyperactivity, that’s behavior; so this is a cognitive-behavioral disorder. That’s the way we define it, the way we characterize it, the way we
study it, the way we treat it.” He looked over at the PowerPoint screen, where the two kinds of brains were still projected. “This to me looks like a disorder of cortical maturation. Imagine if we took everyone with myocardial infarction and said they had a chest-pain problem. Yes, these kids do have attention deficit, they do have hyperactivity, that is part of it, but if you don’t begin to think about ADHD as a disorder of cortical maturation, you’ll never ask the key questions . . . Maybe we could use this as the target to develop treatments instead of always thinking about the observable symptoms.”

  The same is true of schizophrenia, Insel continued, and presumably of other psychiatric disorders. “Behavior is a late manifestation of brain disorders, so if mental disorders are brain disorders and we’ve only allowed ourselves to define them based on manifest symptoms and signs, we’re talking about getting into this game in the ninth inning. And in medicine we don’t do so well when we get into the game very late.” Psychiatrists, not to mention their patients, can’t afford to wait until people actually suffer to intervene—and, if RDoC is successful at laying bare the neural and genetic substrates of our suffering, they won’t have to. They will be able to render a diagnosis before there is even trouble, based on what they can see in the brain.

  And since the brain is nothing but electricity and meat, since, that is, it is real, brain-based diagnosis will also be real, not reified real but really real, and psychiatric nosology will finally put paid to the century-old promissory note—not by finally connecting signs and symptoms to biology and chemistry, but by getting out of the mind business and placing all the money on the brain. In the system the RDoC envisions, there will be no more reminders to clinicians not to think of diagnoses as actual diseases, no more worries that the DSM is taken too seriously, no more whining about epistemic prisons, no more fights over symptom counts or disorder names, no more exclusion criteria, no more doomed attempts to ride the royal road of descriptive psychiatry to the kingdom of anatomical pathology, no more unworkable definitions of mental illness, and, above all, no more bullshit.

  • • •

  On the other hand, maybe not. In 2020, or whenever RDoC comes to fruition, after the animals have been startled or frustrated or taught a new maze and had their brains duly mashed and assayed, after the matrix has been filled in, Insel and Cuthbert or their successors will still have to name those circuits and then define all those words. They will have to say exactly what those measurements of neurons in the stria terminalis or cortisol in the spinal fluid are measuring. And then they will have to do what Kraepelin and Salmon and Spitzer and Frances and First and Regier and every other would-be psychiatric nosologist has had to do: figure out what fear of loss is and where it leaves off and attachment/separation fear sets in, and how much of each is pathological and when and whether to say that the measured symptoms add up to an illness. They will once again be faced with the fact that there is very little that is important about us that can be defined in such a way as to measure it, and that numbers and words may be incommensurable vocabularies, two irreconcilable languages in which to understand us.

  The mind may well be an illusion, something the brain does to entertain us while it goes on about its business, whatever that business is, but it’s a gorgeous illusion and very convincing. I’ll bet you think you are in there reading this, just as I am sure I am in here writing it, and when you are doing that, or when you are anxious or depressed, I’ll bet you are pretty sure it’s because of something other than some crossed-up brain circuits. The mind is also a resilient illusion. The idea that we are agents, that our brains serve us just as our other organs do—in this case providing us with the means to author our lives—and that this is in some sense what it means to be human, has survived all sorts of assaults, and it may survive this one as well.

  But then again, popes and dictators and philosopher kings have never had so many drugs at their disposal or a huge scientific-looking book claiming to list the natural varieties of our suffering. Nor have they been able to hold out the tantalizing possibility of elucidating the brain’s role in consciousness, of finding us in its hundred billion or so neurons, its five hundred trillion synapses, its ten-to-the-millionth-power possible connections. Neither have they presided over a populace quite so eager to turn over their (and their children’s) troubles to their brains and to the doctors who claim to know how to understand them, or quite so willing to gobble down mind-altering medications whose mechanisms of action and long-term effects are as unknown as their capacity to blunt feeling is known. So conditions might be ripe for a neuroscience-inflected psychiatry to usher in a new understanding of ourselves as the people of the brain and for us, with the help and encouragement of our doctors and the drug companies, to become the kind of selves who believe in and benefit from that understanding, and for whom the RDoC’s matrix is the troubleshooting manual.

  Not that an assault on human agency is what psychiatrists like Insel are after. Despite the creepy Minority Report overtones of his idea that we can be mentally ill (and ready for treatment) before we actually do anything, it’s impossible to spend an hour or so with him and Cuthbert or, for that matter, with any of the doctors with whom I have spent so much time over the last couple of years and think that they are motivated by anything other than a wish to relieve suffering. Their purpose in cataloging our troubles is surely not to turn us into Shrink McNuggets. But they are in the grips of forces bigger than they are, bigger than any of us. It’s not their fault that medicine is a service industry, that diseases are market opportunities, and that a book of them is worth its weight in gold.

  • • •

  After my visit with Frances in Boston, I e-mailed to ask him to name a diagnostic category that in his view made the strongest case for psychiatric diagnosis.

  “Why do you hate psychiatrists12 so much? Is it because I pinched your cheek?” he wrote back.

  I persisted. I wanted to hear about a slam dunk, the psychiatric equivalent of strep or diabetes, a single diagnosis that indicated a single pathology and a single treatment. But I would have settled for less, just one solid example of the value of a diagnostic system.

  “Really silly questions,” he replied. “Your bias is showing.”

  Frances did offer a defense. Not for the first time, he told me that “psychiatry done badly can be very harmful, psychiatry done well within its proper competence can be noble. The trick is to develop a healing relationship, to care for the person not just the disorder, to diagnose and treat cautiously, and to see the healthy part of the person not just the sick.” All of which is inarguable, if a little hazy, but it doesn’t really answer the question of why pulling off that trick requires a thousand-page catalog of disorders that are not real—other than to inspire confidence among bureaucrats, and among people who are comforted when a doctor names their suffering. Nor does it explain how exactly that book can keep psychiatry within its “proper competence.” Neither does it acknowledge Freud’s warning that medical education is the worst possible training for people who take on the troubles of the psyche, a warning issued long before the medical-industrial complex turned suffering into a commodity and psychiatry into a profession in which clinician communication must be efficient and the “healing relationship” must be established in ten-minute medication management visits.

  “It isn’t bias to be skeptical,” I wrote. “And it’s not antipsychiatry to question psychiatry. And it’s not silly, in the context of a book about diagnosis, to ask how nosology relates to the practice of psychiatry.” I gave Frances a hypothetical case, a psychotic person he has diagnosed with Bipolar Disorder. “How does that diagnosis help him to proceed?” I asked. It was the question former APA president Paul Fink once answered by saying, “I got paid.”

  “Like to help,” Frances answered. “But the question makes no sense to me. Suggest you read a textbook of psychiatry.”

  At $80.99, Kaplan & Sadock’s Concise Textbook of Cl
inical Psychiatry seems like a real bargain, compared with the $410.99 Kaplan & Sadock’s Comprehensive Textbook of Psychiatry. At 700 pages, it isn’t exactly concise, but it is a lot shorter than the 4,884-page full-size version, not to mention 200 pages shorter than the DSM-IV. So maybe Benjamin and Virginia Sadock, authors of the Concise Textbook, are just trying to save space, but when they tell students that “DSM-IV-TR attempts to describe13 the manifestations of the mental disorders,” they don’t mention that there’s no reality to those mental disorders or warn students of the dangers of reification.

  They do, however, tell the young doctors that it is a “major challenge” to separate the bipolars from the depressives, and they discuss the “difficulty of distinguishing a manic episode14 from schizophrenia.” They explain that “depressive symptoms are present15 in almost all psychiatric disorders,” that “every sign or symptom seen in schizophrenia16 occurs in other psychiatric and neurological disorders,” and that “the distinction between generalized anxiety disorder and normal anxiety is emphasized by the use of the words ‘excessive’ and ‘difficult to control.’” But, the book reassures the students, these difficulties can be overcome through careful clinical observation. Even if the categories don’t exist, in other words, people can nonetheless be sorted into them.

  “Once a diagnosis has been established17,” Sadock and Sadock write, “a pharmacological treatment strategy can be formulated.” That may involve psychosocial treatment, but as just about any psychiatrist will tell you, the days in which psychiatrists underwent psychoanalysis as part of their training are long gone, as are the days in which psychiatrists routinely practiced talk therapy, so that treatment is most likely not going to be provided by them. When it comes to the technique that remains their sole bailiwick—pharmacotherapy—“no one drug is predictably effective18.” For bipolar patients, the doctor has at her disposal lithium, anticonvulsants such as Depakote, tranquilizers such as Ativan, and antipsychotics such as Haldol and Zyprexa. “Often,” advise Sadock and Sadock, “it is necessary19 to try several so-called ‘mood stabilizers’ before an optimal treatment is found.”