The Book We Love to Hate

Why DSM-5 Makes Nobody Happy

Magazine Issue
March/April 2014
An illustration of a person chasing a book with a large hammer

The Diagnostic and Statistical Manual of Psychiatric Diagnoses (DSM) has never been exactly a page turner in any of its iterations (I, II, III, III-R, IV, IV-TR, and now 5), nor have its hundreds of authors over the decades ever been short-listed for the Nobel Prize in literature. The thing’s a slog to read, and it gets longer, sloggier, and pricier with every edition. DSM-II, at 134 pages, cost about $3 in 1968; DSM-5, the latest version, at 947 pages, is priced at $199 by the American Psychiatric Association (APA)—though you can now get it discounted on Ebay for $120.46, free shipping included.

And yet, from small, insignificant beginnings in 1952, when almost nobody read it, it’s become a kind of sacred literary monster—as well as a massive, hernia-inducing brick of a book—that’s read avidly, if not a bit fearfully, by millions of mental health professionals, insurance company employees, drug industry executives, physicians, social workers, government employees, educators, lawyers, law enforcement and public health officials, adoption agency staff, and a surprisingly large swath of the general public. Over the years, it’s become one of the most studied, most professionally cited, and most culturally influential mental health books ever published.

It’s also the most controversial, perhaps the most detested, and certainly the most debated of any mental health classification scheme—any disease nosology—ever devised. Still, the outpouring of rage against the DSM-5 among mental health professionals is startling in its volume and scope—a virtual in-house outbreak of oppositional defiant disorder, and surely the most intense, widespread, and sustained challenge to DSM’s legitimacy in its 62-year history.

The protesters on record include 15,000-plus signatures on the Coalition for DSM-5 Reform petition, sponsored by the APA’s Division 32 (Society for Humanistic Psychology), as well as 50-something other mental health organizations (many of them divisions within the APA), an ad hoc group called Boycott the DSM-5, and many thousands more who’ve voiced their unhappiness on both journalistic and professional websites.

Even more surprising is that the widely acknowledged chief spokesperson for the opposition is none other than Allen Frances—perhaps the last person you would expect to trash this latest, biggest DSM. An eminent psychiatrist, researcher, and writer on a host of issues (including personality and anxiety disorders, chronic depression, schizophrenia, AIDS, and psychotherapy), former chair of the psychiatry department at Duke University, and founder of two psychiatric journals, he himself helped prepare DSM-III and DSM-III-R, and, in what would be the capstone of an illustrious career, chaired the task force that published DSM-IV in 1994. Indeed, in a New York Times article in April of that year, author Daniel Goleman called him “perhaps the most powerful psychiatrist in America.” And yet today, he’s become a de facto spokesperson for the anti-DSM movement.

At this point, a short digression about the 5 in DSM-5 seems necessary: the producers of the new manual prefer the Arabic numeral 5 to the Roman V, because the latter, they claim, is “limiting.” In other words, it’s awkward to use when disseminating revisions electronically—i.e., DSM-5.1, DSM-5.2, and so on—which they plan to do, possibly going up to DSM-5.102 and beyond, for all we know. This means that every time there’s a tiny blip on the diagnostic scene, it can be duly entered into an online revision. No more waiting a decade or more! Besides, as one writer for pointed out, in a techie/digital world, Roman numbers just aren’t cool.

Interestingly, Frances—whose interview with the Networker appears on page 32—was drawn into the debate by another DSM shepherd: Robert Spitzer, the chair of the DSM-III Task Force. Spitzer first attacked the DSM-5 authors in 2008 for imposing a nondisclosure agreement on contributors, thereby allowing them to operate in confidence and not let in any outsiders—even him, the founding father of the entire modern, scientifically aspirational DSM industry. “When I first heard about this agreement, I just went bonkers,” Spitzer was quoted as saying in the New York Times in 2008. “Transparency is necessary if the document is to have credibility, and, in time, you’re going to have people complaining all over the place that they didn’t have the opportunity to challenge anything.”

Spitzer certainly got it right about “people complaining all over the place.” However, there’s something about the attacks on DSM-5 that bring to mind the famous Yogi Berra comment—“it’s déjà vu all over again”—because what most critics object to are the same features, more or less, that have been true of every edition since DSM-III in 1980. Spitzer’s manual was the first to describe and categorize mental disorders as biomedical diseases, which was intended to—and did—reanoint psychiatry as a medical specialty, rather than a languishing psychoanalytic backwater, thus probably saving it from professional irrelevance and oblivion. As the tale goes, once diagnoses were medicalized, they could be medicated, meaning medical doctors could prescribe lots of expensive psychotropic drugs, guaranteeing vast profits to the drug industry. Ever since, according to DSM’s large and growing crowd of hecklers, the DSM, APA, and Big Pharma have been in flagrante delicto, all in bed together, committing indecent acts of diagnostic inflation and mutual back-scratching. It doesn’t help the manual’s reputation for untainted scientific objectivity that around 70 percent of DSM-5 Task Force members and 56 percent of Work Group members hold stock or serve on boards or as consultants to the pharmaceutical industry.

Diagnostic Inflation

Again, these complaints are old hat. Ever since number III hit the shelves, critics have argued that the proliferation of new and expanded disorders have led to wildly exaggerated prevalence rates for mental illness, capturing in DSM’s diagnostic dragnet people who otherwise would be considered mostly okay—maybe eccentric, or hot tempered, or worrywarts, or sad, but not sick. Spitzer himself came to worry about disorder inflation. Interviewed in 2007 by British filmmaker Adam Curtis for a documentary called The Trap, he said that in producing DSM-III, “we made estimates of prevalence of mental disorders totally descriptively without considering that many of these conditions might be normal reactions, which aren’t really disorders.” Asked if he had in effect medicalized ordinary human sadness and fear, Spitzer allowed regretfully that “I think we have to some extent. . . . I don’t know if it’s 20 percent, 30 percent. . . . But that’s a considerable amount if it’s 20 percent or 30 percent.

According to its critics, the problem of diagnosis and medication inflation has gotten much worse with DSM-5. They’re particularly disturbed by what they consider the flagrant reification of mental states—the sleight of hand whereby an unpleasant but vague and often ephemeral mental state can, by giving it a name, be magically transmuted into a biological, physical entity and then acted upon as if it were virtually a Newtonian object in space. Thus, depression or anxiety or attention deficit hyperactivity disorder, or almost any disorder, must be caused by an “imbalance of brain chemicals,” or a genetic anomaly, or some neurophysiological glitch in the works, which “explains” the disorder the way a virus “explains” the symptoms of flu, or an inflamed appendix “explains” a sharp pain in the lower right abdomen. This is bad, anti-DSMers say, because once firmly enclosed in the locked box of a diagnosis, no further exploration or questions ensue about what that person is actually experiencing, or in what context. And a diagnosis box begets a treatment box—one disorder, one remedy—most likely pharmaceutical.

DSM-5 is assailed for presumably surpassing previous editions in diagnostic imperialism, occupying ever larger swaths of what used to be considered normal human troubles. The critics are, for example, particularly disturbed by the removal of the bereavement exclusion from the diagnosis of major depressive disorder, thus—notwithstanding the hedges against overuse included in the wording—pathologizing a normal state of sadness, grief, and, yes, depression that attends a serious personal loss. Another new diagnosis that raises hackles is disruptive mood dysregulation disorder, which applies to children exhibiting persistent irritability and severe behavioral outbursts three or more times a week. Besides the limited amount of research substantiating this disorder (an endemic problem in DSM-5), it’s said to pathologize childish temper tantrums and sets kids up for unnecessary medications. Similarly, the new binge-eating disorder—eating to excess at least once a week—would seem to be what millions of Americans do every week, and perhaps nearly every day between Thanksgiving and New Year’s. Premenstrual dysphoric disorder, promoted from the DSM-IV Appendix to full-scale inclusion, incites fury in many women, who maintain that it stigmatizes women and provides yet another opportunity for the pharmaceutical companies to have their way with them.

The DSM-5 Task Force understandably makes much of the long and labored process that went into producing the new manual: reviews of decades of scientific research, 10 years of international conferences, and deliberation involving more than 900 experts from the United States and abroad, including clinicians and researchers in psychiatry, psychology, statistics, epidemiology, neurology, pediatrics, social work, and other disciplines and specialties, as well field trials involving 3,000 patients and roughly the same number of clinicians in different settings throughout the United States and Canada, not to mention thousands of online comments solicited from medical groups, mental health professionals, advocacy organizations, and whoever else wanted to add his or her two cents. If they’d been producing a graduate-school thesis about the history, sociology, philosophy, and practical operation of psychiatric diagnosis in the 21st century, they’d have received an A+. Unfortunately, they could still muster virtually no more scientific evidence for the diagnostic classification system than could the producers of DSMs III and IV—no genetic tests, no brain scans, no biochemical analyses, no cognitive measures—that corresponded with or backed up or pointed to any of the disorders included in it.

This painful fact was brought home in April 2013, just before the release of DSM-5, when Thomas Insel, director of the National Institute of Mental Health (NIMH), which had long backed DSM, pulled the plug. He announced that since the manual lacked scientific validity, NIMH would no longer fund research based on its categories, which were increasingly out of sync with whatever hard data had emerged from studies of the brain and how it worked. In a way, Insel’s statement was a long-expected coup de grace on the scientific pretensions of the whole DSM enterprise. Begun in 1999, the last year of the Decade of the Brain (a 10-year government initiative to advance brain science), DSM-5 represented the high hopes that by the time the manual was ready, it’d reflect enough neuroscientific breakthroughs to change profoundly the way psychiatric diagnoses were made.

Well, not so much, as it turned out. In an interview on National Public Radio’s “All Things Considered,” psychiatrist Michael First, editor of DSM-IV-TR and editorial and code consultant for DSM-5, explained that the new manual couldn’t be significantly more grounded in science than the last edition because the science just wasn’t there. The research hadn’t allowed for a “paradigm shift” in brain science, so “the DSM is not a paradigm shift either. . . . We were hoping and imagining that research would advance at a pace that laboratory tests would have come out. And here we are 20 years later, and we still, unfortunately, rely primarily on symptoms to make our diagnoses.” First went on to recall that in writing a guidebook to DSM-IV, “I stuck my neck out. We said that by the time DSM-5 comes out, Alzheimer’s will be the first diagnosis that has a laboratory test. Actually, you know, I sort of laid it out there. And I [was] wrong. . . . A lot of it really has to do with the brain, [which] is very, very complicated, and it really hasn’t yielded its secrets yet.”

Defending DSM-5

At this point in the DSM contretemps, a little vagrant wavelet of sympathy might surface for the poor, besieged producers of DSM-5, whose worst sin seems to be that, like Don Quixote, they just dreamed the impossible dream: to establish DSM on firmer, more scientific grounds, to clean up some of DSM-IV’s shortcomings (the inadequate treatment of childhood and adolescent disorders, for example), and to help clinicians more effectively and accurately treat their clients. As for the accusation that they’re money-grubbing shills for Big Pharma and the APA, not even Frances believes that. “I know the people working on DSM-5 and know this charge to be both unfair and untrue,” he wrote in a Psychology Today blog. “Indeed, they have made some very bad decisions, but they did so with pure hearts and not because they wanted to help the drug companies. Theirs is an intellectual, not financial, conflict of interest that results from the natural tendency of highly specialized experts to overvalue their pet ideas, to want to expand their own areas of research interest, and to be oblivious to the distortions that occur in translating DSM-5 to real-life clinical practice (particularly in primary care where 80% of psychiatric drugs are prescribed).”

Darrel Regier, vice chair of the DSM-5 Task Force—interviewed by the Networker on page 32—appears a mild, unassuming researcher, slightly bemused that the release of what was intended to be a more accurate, nuanced, and rigorously researched manual has raised such an uproar. Director of the APA’s research division, he spent 25 years at NIMH heading research divisions in epidemiology, prevention, clinical research, and health-services research. Like other DSM producers and defenders, he knows perfectly well that psychiatric diagnosis is often inherently ambiguous and that the science behind the classification system isn’t all it might be. “We are faced with an interesting situation of having what are well-known deficiencies in the current diagnostic system,” he said with some understatement in a Medscape Medical News interview. A lot of what was going into DSM-5, he admitted, “has not been tested as well as we would like. The current problem is that because the DSM has so dominated research practice for the last 30 years, nobody would even think or get funding to test different diagnostic criteria until they’re really adopted by the DSM.”

In other words, build it and they’ll come—but not before it’s embedded in the brick and mortar of the DSM edifice. Regier regards DSM-5 not as The Law, and certainly not any kind of bible of psychiatry, but as a work in progress, “a set of scientific hypotheses that are intended to be tested and disproved if the evidence isn’t found to support them.” And, again, that’s why it’s DSM-5, not -V: the manual will be a “living document,” which can be regularly revised, “in much the same way as is done with software updates.”

The authors of DSM-5 do lay claim to a genuine innovation: describing many mental disorders on a spectrum from almost-normal to obviously disturbed, determined according to quantitative measures of relative severity or mildness. The idea makes intuitive sense: most mental states, ordered or disordered, exist along a continuum and vary according to how long they last, how powerful or debilitating or animating or life-altering they are, and how much they bleed into other such states; for example, it’s possible to feel both happy and sad, mad and anxious at the same time. DSMs III and IV had seemed increasingly arbitrary, enclosing ambiguous and changeable experiences into rigid categorical cells, either present or absent: either you had major depressive disorder, or you did not, period.

But to critics, the new dimensional aspect exhibits, upon closer inspection, its own kind of fishiness: it resembles an updated replay of the psychodynamic principles of DSMs I and II, in which normal and abnormal didn’t absolutely exclude each other, with one potentially dangerous difference. “Conceiving of mental disorder in a spectrum harkens back to the psychoanalytic tradition that the profession had spent decades shedding,” writes Joel Paris in “The Ideology behind DSM-5,” a chapter in Making the DSM-5, which he edited. But unlike the psychoanalytic tradition, “this new dimensionality is rooted in quantification. . . .

The conviction that mental disorder is a point on a continuum flows directly from a neuroscience-based dimensional model.” Within this odd merging of psychoanalytic and neuroscience concepts, however, “there is no essential difference between normality and psychopathology. The danger of the DSM-5 ideology is that it extends the scope of mental disorder to a point where almost anyone can be diagnosed with one and treated accordingly.” And it supposedly adds the imprimatur of biological science to this diagnostic spread.

Pity the gallant producers of DSM-5, just trying—again—to make psychiatric diagnosis more scientifically respectable and more helpful to clinicians, to make what’s in effect a big dictionary of abstractions decently comport with real-world phenomena. Of course, whether you believe they failed or made strides in the right direction (DSM-5 does, in fact, have its own cheering section out there), everyone can agree that the great paradigm shift didn’t happen. If you believe Thomas Kuhn—who introduced the concept of paradigm shift in his, well, paradigm-shifting book The Structure of Scientific Revolutions—science proceeds not so much in incremental, linear steps toward ever greater knowledge and understanding, as by unexpected and fundamental changes in basic assumptions about the world and what constitutes reality. Such shifts are triggered by a crisis of confidence, when old theories and worldviews increasingly can’t be squared with new discoveries and observations; the more inherently complex the field, the greater the problem with the fit of an outmoded theory. Eventually, after a period of sturm und drang—challenges by visionaries and inventors, resistance and rejection by the old guard, appeal to a younger generation—the great new thing succeeds and fundamentally alters the way everybody looks at reality.

This, the argument goes, will—must!—happen not just to the DSM, the mere tip of the iceberg of outmoded thinking, but to psychiatry itself. According to the dissenters, psychiatry is stuck in its infatuation with scientific positivism, the belief that the only really true truth is found in objective, physical science. In fact, the argument most often made against DSM is that it’s not scientific enough, whereas the model Thomas Insel of NIMH holds up as a replacement is a new diagnostic template—Research Domain Criteria—based largely on the bald assumption that “mental disorders are biological disorders involving brain circuits that implicate specific domains of cognition, emotion, or behavior.” And that settles that.

But—even if you could analyze the profoundly complex mental suffering of human beings in terms of the even more profoundly complex workings of the brain with its billions of neurons and trillions of synapses, and then encapsulate that knowledge into some sort of diagnostic system—the completion of such a project is decades, maybe centuries, away. If the old saw is true that a system can never fully understand itself, and the human brain can never fully “get” the human brain, the completion date gets pushed back to never. Besides the inherent difficulties of such a task, it does seem to underplay a few other small variables in human life, including class, economics, race, social, cultural, and political context, education, geography, systems of belief and ideology, and physical health, not to mention the sheer unpredictable cussedness that lurks within us all, individually and collectively. After all, probably the best psychotherapists do not, and will not, worry much about determining a correct diagnosis, once the reimbursement question has been settled. Instead, they focus on patients and their problems, carefully observing, questioning, reassuring, trying to understand and help this or that unique and complicated person, regardless of what official diagnostic criteria he or she meets.

All this being said, at least for the time being, it looks as if therapists will have to bumble along as they always have, make of DSM-5 what they can, use it when they must, and make plenty of room on their bookshelves (or computers) for it and all its assorted gear: desk reference, pocket guide, clinical case examples, updates, and so forth. Meanwhile, it can’t hurt to be prepared for that morning some time in the future when we look out our window and see there on the eastern horizon the glorious rising of the new paradigm in the sky!



Illustration © James Yang / Illustration Source

Mary Sykes Wylie

Mary Sykes Wylie, PhD, is a former senior editor of the Psychotherapy Networker.