Still, as Kraepelin had demonstrated, it was possible to make observations carefully and group them systematically. Spitzer’s attraction to this method wasn’t so much a matter of conviction as predilection. “Ever since I was a child, I liked to sort things,” he said, recalling that at summer camp he graphed his attractions to different female campers. Of course, to sort out the girls successfully, you have to know which categories to put them in, and what makes them belong in one or the other. You have to believe that beauty and intellect and sense of humor are real properties, and that your way of discerning them is accurate and consistent. Like Kraepelin, Spitzer was certain that if he was careful enough in observing them, the outward manifestations would reveal the underlying mental condition.
To Spitzer, it wasn’t the unfulfilled promise of Kraepelin and Salmon, but rather psychoanalysis—with its claims that psychopathology was the human condition, that same-sex love was the result of damage inflicted in childhood by absent fathers and overbearing mothers, and that, in general, mental suffering was the result of the eternal war among ego, id, and superego—that had led psychiatry to near shipwreck. It was psychoanalysis that had led psychiatrists like Menninger to abandon the idea of sorting suffering into medical categories. It was psychoanalysis that had persuaded doctors to sort mental disorders according to the inner turmoil that had allegedly caused them—the Oedipal conflict, say, or a fixation on the anal stage of infantile sexual development. And it was psychoanalysis that claimed that when it came to our psychological lives, the line between illness and health could be drawn by determining if the problem was the result of intrapsychic conflict, of the lies we tell ourselves about ourselves, of the truths we dance around or repress and transmute into symptoms.
Spitzer hadn’t much liked the psychoanalytic training that had been required of him and most psychiatrists of his era, and he really didn’t like being an analyst. “I was uncomfortable with not knowing31 what to do with their [patients’] messiness,” he said. “I just didn’t know what the hell to do.” And it was obvious to him that Freud’s theory of mind was a poor substitute for pathological anatomy, and the complexes and resistances and defense mechanisms—the psychoanalyst’s stock-in-trade—were far too ungrounded in any kind of empirical reality to be useful. Proving the existence of ego, id, and superego was like proving the existence of the Holy Trinity. These notions were more metaphysics than physics, psychoanalysis more religion than science, and the crises of the 1960s and 1970s were the result.
Of course, this was exactly the problem: psychoanalysis had thrived in the theoretical vacuum left by the continued ignorance of how the brain works. So it wasn’t like there was a theory waiting in the wings to replace it. But, Spitzer reasoned, if his only options were a theory that couldn’t be proved (and that was leading his profession to disaster after disaster) or no theory at all, then the correct choice was obvious. It was time to abandon Freud’s pretense to understanding the origin and nature of mental illness, and to return to the one thing Kraepelin said psychiatrists could safely claim to know: what they could observe.
Spitzer was already working with a group of researchers at Washington University in St. Louis to resurrect Kraepelin. By 1972, the group had described32 fourteen different diagnostic groupings, established the criteria by which patients could be placed into one or another of them, and conducted research showing that the diagnoses were reliable. Six years later, the Washington University group issued the Research Diagnostic Criteria33 (RDC), twenty-one categories with checklists of the criteria by which each one could be known.
The RDC bore virtually no resemblance to the DSM, then in its second edition. Where the DSM-II listed illnesses like Depressive Neurosis, defined as a “disorder manifested by an excessive reaction of depression due to an internal conflict,” the RDC created Major Depressive Disorder, defined not in a paragraph full of Freudian jargon, but as a list of symptoms:
A.One or more distinct periods34 with dysphoric mood or pervasive loss of interest or pleasure. The disturbance is characterized by symptoms such as the following: depressed, sad, blue, hopeless, low, down in the dumps, “don’t care anymore,” or irritable . . .
B.At least five of the following symptoms are required to have appeared as part of the episode for definite and four for probable (for past episodes, because of memory difficulty, one less symptom is required).
1.Poor appetite or weight loss or increased appetite or weight gain (change of 0.5 kg a week over several weeks or 4.5 kg a year when dieting)
2.Sleep difficulty or sleeping too much
3.Loss of energy, fatigability [sic], or tiredness
4.Psychomotor agitation or retardation (but not mere subjective feeling of restlessness or being slowed down)
5.Loss of interest or pleasure in usual activities, including social contact or sex (do not include if limited to a period when delusional or hallucinating) (The loss may or may not be pervasive.)
6.Feeling of self-reproach or excessive or inappropriate guilt (either may be delusional)
7.Complaints or evidence of diminished ability to think or concentrate, such as slowed thinking, or indecisiveness (do not include if associated with marked formal thought disorder)
8.Recurrent thoughts of death or suicide, or any suicidal behavior
C.Duration of dysphoric features at least one week, beginning with the first noticeable change in the subject’s usual condition (definite if lasted [sic] more than two weeks, probable if one to two weeks).
Purged of shaggy concepts and imprecise language, the RDC could tell clinicians exactly what to look for and what they had found if they saw it—a method that would leave much less room for disagreement.
The RDC may have renewed Kraepelin’s promissory note, but, so Spitzer claimed, it also provided the means for paying it off. Of course, Socrates’ observation still held: to gather together scattered particulars into groups was not to carve nature at its joints. But, Spitzer contended, all those observations, made by many researchers using the same language, would add up to a body of knowledge for each disorder, and the nosology would inexorably gain substance35. Two different teams researching depression, for example, would be able to claim that they were looking at the same phenomenon, and as findings converged and solidified, so would the case for the existence of depression as a disease. And if that convergence couldn’t or didn’t occur, then that would be an indication that there was something wrong with the category, that it had been ill conceived or poorly defined, or that it just plain didn’t exist. Reliability—the extent to which diagnostic criteria would yield agreement among clinicians—was not the same as validity, or the extent to which the diagnosis described an actual disease. But beefed-up reliability could at least make the profession seem less bogus.
“The use of operational criteria36 for psychiatric diagnosis is an idea whose time has come!” Spitzer wrote in 1978. Of course, he knew that twenty-one diagnoses wouldn’t be nearly enough to cover the entire terrain over which psychiatrists claimed dominion, or to ensure that all their patients would qualify for a diagnosis and thus insurance reimbursement. He invited his colleagues to name the disorders they thought they were treating, and he found many of them eager to get in on the ground floor of the new scheme. He and his committees winnowed the suggestions, developed the criteria by which the survivors would be known, and assembled them into a diagnostic manual the likes of which had never been seen. A 500-page hardbound tome, the DSM-III made the spiral-bound, 134-page DSM-II look like a mere pamphlet. And by nearly doubling the number of mental disorders, it also vastly expanded the manual’s scope, turning it into an entirely new psychopathology of everyday life.
Even so, the DSM-III was not universally popular among psychiatrists. Some thought its symptom lists, its plain language, and its workaday disorder names degraded the profession. “Clerks rather than experts can make this kind of classification,” one grumbled.<
br />
This, of course, was exactly the point. Imagination was what had led psychiatry to founder; a Kraepelinian nosology was the best way to prevent psychiatrists from steering their profession back onto the shoals of unreliability. Spitzer had fashioned a dictionary of disorder that allowed psychiatrists to identify our foibles without recourse to the troublesome Freudian mumbo jumbo or, for that matter, any other mumbo jumbo.
And the result was sensational. The DSM-III not only restored both internal and external confidence in psychiatry; it was also an international bestseller. “It made an unbelievable amount of money for the APA,” Spitzer said. “That was a huge surprise.” And looking back on it, Spitzer has no question where the book’s popularity came from. “DSM-III looks very scientific,” he told me. “If you open it up, it looks like they must know something.”
• • •
The psychiatrists who wrote and used the DSM-III indeed knew something. They knew that certain symptoms tended to group together and that psychiatrists could reliably identify the people who belonged in those groups. But Spitzer acknowledges that the book did not solve the validity problem. He doesn’t even think it was supposed to. Indeed, he told me, the APA hired him to achieve only the smallest of bureaucratic goals—to bring the DSM into harmony with the World Health Organization’s International Classification of Diseases, known as the ICD. When I pressed him, he allowed that, of course, increasing reliability was also on his—and the APA’s—mind. But validity? “No, no,” he said, “not at all.” That’s not what they wanted, and that’s not what he meant to do.
Even now, in fact, the man who crafted the deal to delete homosexuality isn’t sure that homosexuality is not a valid disorder. “It has a distinct course, there’s no doubt about that,” he said, adding that there are gender differences in prevalence and evidence that it is a familial trait. But, he cautioned, “to decide whether it’s a disorder or a normal variant, you’d have to decide whether homosexuality represents a dysfunction. People who think it is a disorder would argue from an evolutionary viewpoint that we are hardwired for heterosexual attraction.”
Some of those people would also argue that a lack of heterosexual attraction is a disease to be cured, and in 2003, after interviewing two hundred people who claimed to have been “cured” of their sexual orientation by “reparative therapy,” Spitzer determined that “highly motivated” patients could indeed change their preference, a conclusion he published in the Archives of Sexual Behavior37. The journal accompanied the article with fiercely critical commentary, and in the popular press gay activists pilloried the man who had freed them from psychiatry. Spitzer defended his work then, but more recently, he told me, he had come to regret the paper and was considering “writing something in which I would say the critiques are largely or in many ways true.”
In May 2012, Spitzer did exactly that. His assumption that “participants’ reports of change were credible and not self-deception or outright lying” had been incorrect, he wrote in a letter to the Archives38. His subjects had told him they were no longer attracted to people of the same sex, but “there was no way to determine if [their] accounts of change were valid.” Spitzer apologized to the “gay community” for making “unproven claims of the efficacy of reparative therapy” and to “any gay person who wasted time and energy undergoing [it].”
Spitzer’s recantation, and his ongoing uncertainty about homosexuality, reflect his questing and curious mind, which seeks empirical answers to difficult questions and is always open to new evidence. It also reflects his honesty about the limitations of his paradigm, which can elicit detailed accounts of what people are experiencing, but can’t say exactly what, if anything, to make of those accounts. Descriptive psychiatry can’t determine whether or not a person’s story about his sexual orientation is a true one. More important, it can’t tell us whether the list of symptoms, no matter how reliable, constitutes a disease. It can gather scattered particulars into a category called gay, but it can’t say whether those amount to the natural formation known as disease. It can’t carve nature at its joints.
Spitzer is also honest about the fact that the decisions he made to admit or exclude a diagnosis from the DSM-III were not entirely scientific. “The categories that were added were concepts that clinicians in those days thought were important,” he said, and their criteria consisted of what “clinicians said was a good way of defining them.” Spitzer was the nosological diplomat among clinicians squaring off over pet concepts. He was perfectly suited to this role because, in addition to his predilection for sorting, he said, “I love controversy. I love it!”
Spitzer no doubt had in mind the controversy he had presided over as his committees fought about diagnoses and criteria—but it was only a matter of time before a different kind of controversy set in. It was exactly the kind of controversy that Thomas Salmon had worried about, and the culprit was none other than Bob Spitzer’s DSM.
Chapter 3
Allen Frances often sings praise as a prelude to criticism, and sometimes in exact proportion. So when he tells me how brilliant Bob Spitzer is, how valuable he has been to psychiatry, and then adds, “I don’t want anything in your book that would hurt him the slightest bit,” which he does repeatedly, it is pretty clear that he’s winding up to plunk him. Not that he doesn’t deeply respect Spitzer—that’s obvious as soon as you see the two of them together—or doesn’t mean the praise. But he definitely has a beef with the man he replaced at the DSM’s helm.
The problem isn’t the DSM-III. The move to descriptive diagnosis was, Frances believes, necessary and beneficial for psychiatrists and patients alike. But almost immediately after that book came out in 1980, the APA decided to revise it. The new book would be called DSM-III-R, to reflect the fact that it was not a new edition, but a minor revision to correct textual errors and tweak criteria that were proving unwieldy. Spitzer was hired to direct the effort, but, according to Frances, “Bob couldn’t resist playing with it. He couldn’t resist the committee meetings, all the new diagnoses, all the excitement” as experts, once again given the opportunity to enshrine their pet ideas, advocated for new labels or criteria.
“In the morning, everyone would be screaming ideas1,” Frances recalled. “Bob [Spitzer] and Janet [Williams, Spitzer’s wife and a member of the revision committees] would be on a blackboard, trying to put it into some kind of order. Then we’d have lunch, usually a big lunch.” While the others ate, Spitzer and Williams would refine the morning’s arguing into diagnostic criteria. When the group reconvened, Frances said, “we’d be sleepy and much more subdued,” making it that much easier “for the most powerful person in the room to rule.” The wrangling continued after the sessions, as doctors collared Spitzer and lobbied for their proposals.
The backroom dealing was bad enough, Frances thought, but even worse, the fighting was in many ways pointless. “The things that looked so different to the people involved never amounted to a hill of beans,” Frances said. “Should the threshold for a diagnosis be four or five symptoms, should the criterion be this item or that? The answers are almost always arbitrary.” So in 1987, when Harold Pincus, then the APA’s director of research, offered him the job of running the DSM-IV revision, Frances told him he’d accept only if the process was entirely different. “The last thing I wanted was to be in rooms full of people pontificating without evidence about things that didn’t matter,” he said. As much as he might have coveted the job, “I never wanted to be in one of those meetings again.”
Frances didn’t always feel this way. In fact, at one time he was among the pontificators. “I knew the instinct,” he said—the one that leads a doctor to think that he’s seeing a cluster of symptoms that no one has put together before, and thus has discovered a new disorder. “You think you’re smarter than everyone else, and that what you’re seeing in this patient should be in the DSM.”
Frances himself had once followed the instinct, nominating
Masochistic Personality Disorder for DSM-III, aimed at people who “employ self-sacrificing and self-defeating behavior2 in service of maintaining relationships or self-esteem.” His proposal came under withering attack from feminists who considered it a way of blaming abusive relationships on women’s psychopathology. It also, Frances said, turned out to be a “dumb idea3, because all the behaviors in a diagnostic manual of mental disorders are by definition self-defeating. The concept really adds nothing.” By 1980, he had abandoned the proposal, having learned an important lesson about diagnostic manuals. “Realizing in retrospect how dumb my own off-the-cuff suggestion was made me more alive to the fallibility of the many other off-the-cuff suggestions DSMs necessarily attract.”
As much as he wished to steer clear of the dumb arguing, however, Frances was even more intent on avoiding the pitfall that had snared Spitzer when he ran the DSM-III-R effort. “It’s much better to have a common language than a Babel of different languages,” he said, and the DSM-III had achieved that. But whatever stability and respectability this success had brought to the field was threatened by DSM-III-R. When that book came out in 1987, a year late, over budget, and even longer than the original, clinicians once again had to master new diagnoses, researchers had to piece together the new disorders with the old literature, and psychiatry, which had just barely settled down, was once again in turmoil.
The DSM-III had systematized the description of mental disorders, put labels on clusters of symptoms, but, as the DSM-III-R process had proved, those clusters could be arranged and rearranged indefinitely. Descriptive psychiatry was no small achievement, but the categories, the boundaries between them, and the criteria within them—these were not discoveries of nature at work, at least not in the same way that the identification of streptococcus and influenza, their characteristics, and the boundaries between them were. They were approximations, and even if they were based on careful observations, they were forever in debt to expert opinion.
The Book of Woe: The DSM and the Unmaking of Psychiatry Page 5