Like it or not, this is what a revolution looks like.
“It is as important to psychiatrists as the Constitution is to the US government or the Bible is to Christians,” wrote psychotherapist Gary Greenberg. All drug trials from the birth of the DSM-III forward were based on the manual’s criteria; insurance companies used it to decide how much coverage a person should receive; if a shrink or any kind of mental health professional wanted to get reimbursed for their time, they’d better know how to cite the DSM from memory. The DSM-III turned madness into different types of disorders that each responded to specific drug treatments, creating “rich pickings for the pharmaceutical industry.” And it didn’t stop with psychiatrists, extending to psychologists, social workers, and lawyers. It’s used in everything from criminal cases to custody battles, from courtrooms to the allocation of special needs resources in public schools.
One of Spitzer’s pet projects was to define mental disorder, a pursuit that he had been fixated on since the homosexuality debacle. The DSM-III laid that out at the very outset: A mental disorder “is conceptualized as a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is typically associated with either a painful symptom (distress) or impairment in one or more important areas of functioning (disability).” Not only did it associate mental illness with dysfunction, which was meant to protect us against making illnesses out of healthy eccentricity, but it also located the cause of mental illness inside the person (not with overbearing mothers or weak fathers, for example) in the same way that physical diseases, like cancer or heart disease, affect the body. So the manual used the term disorder—which implied a stronger biological connection—and threw away reaction, a relic of the psychodynamic era.
The DSM said outright that the continued distinction between physical and mental, between organic and functional, was “based on the tradition of separating these disorders,” while acknowledging that these distinctions were somewhat arbitrary. “Hence, this manual uses the term ‘physical disorder,’ recognizing that the boundaries for these two classes of disorders (‘mental’ and ‘physical’ disorders) change as our understanding of the pathophysiology of these disorders increases.”
To reflect this, the manual did not provide causes for the psychiatric disorders listed—the science just wasn’t there. The goal instead was to keep that part open-ended until the science caught up. It’s unclear if the clinicians who bought these books took note of these caveats, however, because everyone else saw the manual, combined with the promise of emerging neuroscience and genetics, as a recasting of psychoanalytically interpreted illnesses into full-blown brain illnesses.
No matter how little proof was there, psychiatry fully embraced the illness model—also known as the field’s remedicalization. Harvard psychiatrist Gerald Klerman called it “a victory” for science. It altered the way both doctors and patients saw the provenance of illness and their roles in it—instead of repressed egos and ids or frigid mothers, you had screwed-up brain chemicals or faulty (but not our fault) wiring. Psychiatrists like Nancy Andreasen saw this as a step forward for patients who “no longer must carry the burden of blame and guilt because they have become ill.” And that the world should “behave towards a patient just as they would if he had cancer or heart disease.”
All the while, the problem of Rosenhan and his pseudopatients nettled the manual’s creator. As Spitzer worked on drafts of the DSM, he often returned to Rosenhan’s study and asked himself: Would David Rosenhan and his pseudopatients get past this one?
“When we would write a criterion, for instance, we would often have the study in the back of our minds,” explained Spitzer’s wife, Janet Williams, who also worked on the DSM-III. “Criterionating, we used to call it. You had to write the criteria down and then think of every which way to question it, to improve it… We were always asking those things. This was when Rosenhan would inevitably come up.”
Spitzer was determined to make sure that the publicity nightmare that Rosenhan and his seven pseudopatients generated would never happen again. “Rosenhan’s pseudopatients would never have been diagnosed as schizophrenic if the interviewing psychiatrists had been using DSM-III,” wrote Tanya Marie Luhrmann.
“What Bob [Spitzer] did,” psychiatrist Allen Frances said in an interview, “was change the face of psychiatry, change the face of how people saw themselves. It wasn’t just a plus, but he did change the world, and that change was very much instigated by the Rosenhan project.” Without Rosenhan’s study, Frances told me, “Spitzer could never have done what he did with the DSM-III.”
It seemed to be a win for us all. Now we had a solid diagnostic system; we had medical language that replaced psychobabble; we had reliability so that doctors all over the world would make a consistent diagnosis.
It sounded, at least at first, like progress to me. I’ve met some of the holdover psychiatrists from the psychoanalytic era—one told me that he used to get an erection while standing at a podium in front of a new class of medical students and that he’d show it off by jutting out his hips and walking up and down the aisles. Another told me that I was fully healed from autoimmune encephalitis not because of advances in immunology or cutting-edge neuroscience, but because I “hadn’t experienced any real trauma before that moment.” As if a five-minute interaction can reveal something so deeply rooted.
If this arrogance is what the DSM-III replaced, good riddance.
21
THE SCID
In 2016 Spitzer’s wife, Janet, invited me to attend his memorial lecture at the New York State Psychiatric Institute, his long-term employer. On my way to the lecture, while wandering along a cul-de-sac formed by a group of identical academic buildings, I lost my way and asked two young men, who looked like medical interns or residents, where I could find the institute. They pointed me to a building at the end of the street and waved as I walked off.
Their helpful responses reminded me of Rosenhan’s mini-experiment in “On Being Sane in Insane Places.” In the experiment’s first iteration, research assistants posed as lost students at Stanford Medical School and were catered to with a pushy level of politeness. In the second iteration, Rosenhan had his pseudopatients ask staff for directions and then monitored the responses. Rosenhan included this interaction from his Haverford hospitalization in his published paper:
Pseudopatient: Pardon me, Dr.. Could you tell me when I am eligible for grounds privileges?
Physician: “Good morning, Dave. How are you today?” (moves off without waiting for a response.)
(It is worth noting that all I could find in Rosenhan’s notes were the students who had conducted the experiment in the medical school—there is, frustratingly, zero conclusive evidence beyond what he wrote in the study that Rosenhan or the other pseudopatients actually conducted this experiment inside the psychiatric hospitals.)
When I finally arrived at the memorial lecture, the auditorium was packed. Dr. Michael First, a close colleague of Spitzer’s, opened with an overview of Spitzer’s work. Guess who made the cut?
“The following year David Rosenhan published a controversial paper in Science describing how eight pseudopatients were admitted to psychiatric wards for an average of nineteen days despite behaving normal after a single initial claim of hearing a voice that said ‘thud,’” Dr. First said. On my recording you hear my laugh. Rosenhan had wormed his way into Spitzer’s bio. “Now Bob wrote a scathing critique of this study, and this is a quote—and I like this quote because this is a typical way of Bob in his artful way of using language of sort of putting the study down. He said, ‘A careful examination of the study’s methods, results and conclusions leads me to a diagnosis of ‘logic in remission.’”
The room erupted in laughter. It still killed.
Dr. First finished his short introduction and called on Dr. Ken Kendler, a researcher and professor of psychiatry at Virginia Commonwealth University who contributed to the DSM-III-R (the revision o
f DSM-III) and DSM-IV, and chaired the DSM-5’s Scientific Review Committee. (I’m giving you this background because it makes what comes next all the more surprising.) I expected his lecture to be a “rah-rah-rah!” celebration of psychiatry’s bible. I was mistaken.
Ken Kendler has the kind of mind that expects you to rise to its level, but for our purposes I’ll attempt to sum up. Basically, he told the audience that in the process of legitimizing the DSM, psychiatrists took it literally, ignoring all the gray unknowns. Psychiatrists believed in the “reification of psychiatric diagnoses.” Or, in my words, psychiatrists got high on their own supply and started to believe that there was more there there. “We were really proud of our criteria when these came out and that kind of added to the sense that we really wanted a glow around these [diagnoses], to say that these are ‘real things,’ we’ve really got it here, it’s all in the manual,” Dr. Kendler said. “Kind of like Moses coming down from Mount Sinai, except it was a Jewish guy called Bob Spitzer.”
When Spitzer brought his tablets “down from the Mount” in the form of the DSM-III, the field embraced the manual with an almost religious devotion. “We ask people: Are you sad? Are you guilty? Is your appetite down? We’re struggling as a field. Symptoms and signs are all we fundamentally have,” said Dr. Kendler. Though the symptoms and signs are very real, the underlying causes remain as mysterious as they were a century ago.
The DSM-III did fundamentally change mental health care in this country—but many experts now question if the change was the right direction. “Rather than heading off into the brave new world of science, DSM-style psychiatry seemed in some ways to be heading out into the desert,” wrote Edward Shorter in his A History of Psychiatry. “The sheer endlessness of the syndrome parade caused an uneasy feeling that the process might be somehow out of control.”
It’s easy to forget that all of the major psychiatric diagnoses were designed and created by consensus. Creation was neither smooth nor orderly. A core group of less than ten people, most of whom were psychiatrists, “clustered around Spitzer, all of them talking as he banged out text on his typewriter. There were no computers, and revisions were made by manual cutting and pasting,” wrote Hannah Decker in The Making of the DSM. Angry disagreements abounded. Feelings were hurt. All the while Spitzer typed away furiously, a demon on his typewriter getting it all down, devoting seventy to eighty hours a week to the project. “There would be these meetings of the so-called experts or advisers, and people would be standing and sitting and moving around,” one psychiatrist who worked on the manual told the New Yorker. “People would talk on top of each other. But Bob would be too busy typing notes to chair the meeting in an orderly way.” Psychologist Theodore Millon, a DSM-III task force member, described the scene: “There was very little systematic research, and much of the research that existed was really a hodgepodge—scattered, inconsistent, and ambiguous. I think the majority of us recognized that the amount of good, solid science upon which we were making our decisions was pretty modest.”
Even reliability, trumpeted as one of the major wins of the new manual, was oversold. In 1988, 290 psychiatrists evaluated two case studies and were asked to offer a diagnosis based on DSM criteria. The researchers, however, had devised a way to test the clinicians’ own diagnostic biases: They created multiple patient case studies out of the two set examples by altering two factors: race and gender. Even when presented with identical symptoms, clinicians tended to identify black men as more severely ill than any other group. (This continues to be true today: One 2004 study showed that black men and women were four times more likely to receive a schizophrenia diagnosis than white patients in state hospitals.)
The issue with reliability is that consensus does not necessarily translate to legitimacy. “In days of yore, most physicians might agree that a patient was demonically possessed. They had good reliability, but poor validity,” noted Michael Alan Taylor in Hippocrates Cried.
Rosenhan never spoke publicly about his thoughts about the DSM. Given his private correspondence with Spitzer, I’m sure he suspected that his paper at least shaped parts of the manual. Would he be proud of the wide-reaching effects of his experiment or would he be dispirited by how his study was exploited to push the field’s agenda to save itself?
The next edition, the DSM-IV, was overseen by Allen Frances in 1994. “It followed dutifully in Spitzer’s footsteps, though it included new diagnoses and broadened and weakened the criteria that had to be met for any particular diagnosis to be assigned,” according to sociologist Andrew Scull.
As we saw, diagnostic boundaries for mental illness have collapsed and expanded over time. When Rosenhan was hospitalized, the schizophrenia diagnosis cast a far wider net than today. How shall we know them? Make that bucket too wide and these words become meaningless; make it too narrow and you miss people who desperately need help. Dr. Keith Conners, considered the “godfather of medication treatment for A.D.H.D.” who helped establish standards for diagnosis of the condition, expressed dismay at the growing numbers of kids (15 percent of high schoolers) with the label. “The numbers make it look like an epidemic. Well, it’s not. It’s preposterous,” he told the New York Times in 2013. “This is a concoction to justify the giving out of the medication at unprecedented and unjustified levels.”
When the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders came out in 2013, it crash-landed to terrible press. The DSM, behind schedule and belabored by criticisms from inside (and outside) its own ranks, aimed to implement a “dimensional aspect” or a continuum of mental disorders rather than the strict categories that defined the previous volumes. At least three books in 2013 slammed the manual before it was even published—Gary Greenberg’s The Book of Woe, Michael Alan Taylor’s Hippocrates Cried, and Allen Frances’s Saving Normal.
Saving Normal, which Frances described as “part mea culpa, part j’accuse, part cri de coeur,” all anti-DSM-5,1 was the most vociferous given his former position as the head of the DSM-IV task force and his close relationship with DSM godfather Spitzer. It was, of all people, Spitzer himself who had recruited Frances from out of retirement to join him in warning the public that the new manual would likely “produce a very dangerous product.” The release of the manual was stalled twice—thanks, at least in part, to these two heavyweights. Frances wrote open letters to the APA, op-eds, and tweets. He admitted to the public that he had failed “to predict or prevent three new false epidemics of mental disorder in children—autism, attention deficit, and childhood bipolar disorder.” Diagnoses of childhood bipolar disorder had increased fortyfold in the eight years between 1994 and 2002; there had been a fifty-seven-fold increase in children’s autism spectrum diagnoses between the 1970s and today; and attention-deficit/hyperactivity disorder, once a rarity, now affected an estimated 8 percent of children between the ages of two and seventeen. Frances’s point that our definitions have drastic, real-life implications was a valid one—were we reaching people who had long been ignored or were we overdiagnosing and overmedicating children? Frances warned that the DSM-5 would further “mislabel normal people” and create “a society of pill poppers” (in a time when already one in six adults were using at least one drug for psychiatric problems). Some APA psychiatrists reacted by arguing that Frances had not only a reputation to save but also money to lose, because the new manual would reduce the royalties he was collecting on his own creation, the earlier version of the book.
Still other greats in the field piled on. Dr. Steven Hyman, director of the Stanley Center for Psychiatric Research at the Broad Institute at MIT and Harvard, called it “an absolute scientific nightmare.” Dr. Thomas Insel, the former director of the National Institute of Mental Health, said that the manual had a “lack of validity” and was “at best a dictionary.” Here’s the deal: The science wasn’t there when Spitzer and company wrote the manual (and they tried to acknowledge this by leaving the manual open for revisions). Despite all the effort in the three decad
es since, it still isn’t there.
Many research psychiatrists I’ve interviewed liken DSM diagnoses to our understanding of headaches—we have symptoms with no knowledge of the underlying cause. You can, for example, think you have merely a headache when you in fact have a brain tumor. Pop an Advil and your headache might go away, but you’ve still got a metastasizing mass in your skull. Without a way to find that tumor, how do we tell the difference?
The most concerning part, from my perspective, is that the DSM approach rendered the practice so rigid, so fixed, that the patient, the person, the human, was lost. As I would learn, this doesn’t just affect the relationship between doctor and patient, but can increase misdiagnosis.
I had tested this out myself with Dr. Michael First, the man who introduced Spitzer and mentioned Rosenhan at the memorial lecture.
“I’m nervous,” I said as I turned on the tape recorder in First’s office. “Why am I nervous? Have you been SCIDed yourself?”
“Nope,” Dr. First said.
Dr. First is not exactly warm and fuzzy—he’s hyperclinical and a straight shooter, two things that have made him key in the creation of the last three incarnations of the DSM—but the chunky metal ring I spotted on his finger during our interview betrays what I interpret as his softer, Woodstock-hippie vibe. He is often called upon to consult in high-profile criminal cases, recently that of the murder of six-year-old Etan Patz, which ended in a hung jury (the defendant was found guilty in a second trial). But his main contribution to the DSM world is the SCID—the Structured Clinical Interview for DSM—a prewritten set of interview questions designed to make a psychiatric diagnosis based on DSM criteria. I had asked if he would be open to SCIDing me about my experience with psychosis, pretending that he didn’t know the diagnosis. Dr. First seemed open to a challenge—even if the odds were stacked against him.
The Great Pretender Page 20