In his third year as a resident, Feighner was invited to attend meetings with Robins, Guze, and Winokur. He quickly absorbed their Kraepelinian perspective on diagnosis and decided to try to develop diagnostic criteria for depression based upon their ideas. He reviewed close to a thousand published articles about mood disorders and proposed specific symptoms for depression based upon the data in these papers. Impressed with their resident’s swift progress, the Washington University trinity formed a committee to help Feighner—and encouraged him to find criteria not just for depression but for all known mental illnesses.
The committee, which also included Washington University psychiatrists Robert Woodruff and Rod Munoz, met every week or two over a period of nine months. Feighner worked tirelessly, bringing every possible paper he could find on every disorder to the committee for review and using this research to propose criteria that were debated, refined, and endorsed by the group. In 1972, Feighner published their final system in the prestigious Archives of General Psychiatry as “Diagnostic criteria for use in psychiatric research,” though his system soon became immortalized as the Feighner Criteria. The paper concluded with a deliberate shot across the bow of psychoanalysis: “These symptoms represent a synthesis based on data, rather than opinion or tradition.”
The Feighner Criteria eventually became one of the most influential publications in the history of medicine and one of the most cited papers ever published in a psychiatric journal, receiving an average of 145 citations per year from the time of its publication through 1980; in contrast, the average article published in the Archives of General Psychiatry during the same period received only two citations per year. But when Feighner’s paper was first published, it had almost no meaningful impact on clinical practice. To most psychiatrists, the Washington University diagnostic system seemed like a pointless academic exercise, an esoteric research instrument with little relevance to treating the neurotics they saw in their clinical work. But a few psychiatrists did take notice. One was Robert Spitzer. Another was me.
Five years after the publication of his paper, John Feighner came to St. Vincent’s Hospital in New York where I was a second-year resident and gave a talk about his new diagnostic criteria. Feighner was physically unimpressive, but his brash manner and energetic intelligence made for a charismatic presence. His ideas resonated with my own growing disenchantment with psychoanalysis and spoke to the confusing clinical reality I was facing with my patients every day.
As was the custom, the St. Vincent’s residents had lunch with the speaker following his lecture. Over pizza and soda, we peppered Feighner with questions, and I recall being an overeager interrogator; I even followed him out of the building and all the way down the street while he hailed a taxi so I could continue talking to him as long as possible. He told me he had just moved to join the faculty of the newly established psychiatry department at the University of California in San Diego and opened a private psychiatric hospital in nearby Rancho Santa Fe that employed his new diagnostic methods, the first of its kind. This encounter with Feighner proved quite fortuitous for me.
A few months after I met Feighner, I received a call from an uncle who informed me that his daughter, my cousin Catherine, was having problems while attending a midwestern college. I was surprised, since I had grown up with her and knew her as intelligent, sensible, and grounded. But according to her father, she was out of control. She stayed out late partying, getting drunk, having risky sex, and engaging in numerous tumultuous relationships. But she would also tuck herself away in her room for days at a time, skipping classes and refusing to see anyone. My uncle didn’t know what to do.
I called Cathy’s roommate and the resident counselor in her dorm. From their concerned descriptions it seemed that she was probably suffering from some form of manic-depressive illness, today called bipolar disorder. Though her university offered mental health services, the staff consisted of psychologists and social workers who mainly provided counseling. The psychiatry department at the university, meanwhile, was run by psychoanalysts, as were all of the eminent psychiatric centers at the time (including the Menninger Clinic, Austen Riggs, Chestnut Lodge, Sheppard Pratt, and the Payne Whitney Clinic). I had begun to question the effectiveness of psychoanalytical treatments, and didn’t want to consign my cousin to misguided care at any of these Freudian institutions. But, then, how to help Cathy? An inspired idea suddenly occurred to me: I would call John Feighner.
I explained Cathy’s situation and worked out a plan for her admission to his new hospital halfway across the country, arranging for her to be directly under his care. Following her arrival, Feighner confirmed my provisional diagnosis of manic-depressive illness using his Feighner Criteria, treated her with lithium (a new and highly controversial drug), and within weeks had stabilized her condition. Cathy was discharged, resumed her classes, and graduated on time.
Today I argue against sending patients out of state for psychiatric treatment, since it’s usually possible to find competent care locally. But in 1977, at that early stage in my career, I did not have sufficient confidence in my own profession to risk the health of someone dear to me to psychiatry’s existing standard of care.
While Feighner made a big impression on me, his criteria were mostly greeted with yawns. According to historian Hannah Decker, the Kraepelinians at Washington University were not surprised by their lack of impact, believing they would be “lucky” to make “a dent” in a field ruled by psychoanalysis.
They turned out to be very lucky indeed.
A Book That Changed Everything
“The Washington University people were absolutely delighted I got the job because they were completely outside of the mainstream, but now I was going to use their diagnostic system for the DSM,” Spitzer says, smiling. Spitzer had been introduced to the Washington University group in 1971, two years before he was appointed chair of the DSM-III, while working on an NIMH study of depression. The head of the project suggested that Spitzer visit Washington University to check out the Kraepelin-influenced ideas about diagnosing depression that originated with Feighner and the Robins, Guze, Winokur triad. “When I got there and discovered they were actually establishing menus of symptoms for each disorder based on data from published research,” Spitzer recounts with obvious pleasure, “it was like I had finally awoken from a spell. Finally, a rational way to approach diagnosis other than the nebulous psychoanalytical definitions in the DSM-II.”
Armed with the Feighner Criteria and determined to counteract the claims of the antipsychiatry movement by establishing rock-solid reliability in diagnosis, Spitzer’s first job as chairman was to appoint the other members of the DSM-III Task Force. “Outside of the APA Board, nobody really cared much about the new DSM, so it was totally under my control,” Spitzer explains. “I didn’t have to clear my appointments with anyone—so about half my appointees were Feighner types.”
When the seven Task Force members assembled for the first time, each expected to be the odd person out, believing their desire for increased objectivity and precision in diagnosis would represent the minority view. To their surprise, they discovered that, as a group, they unanimously favored the “dust bowl empiricism” of Washington University: There was universal consensus that the DSM-II should be unabashedly jettisoned, while the DSM-III should use specifically defined, symptoms-based criteria instead of general descriptions. Task Force member Nancy Andreasen of the University of Iowa recalls, “We shared the feeling we were creating a small revolution in American psychiatry.”
Spitzer established twenty-five separate DSM-III subcommittees, each asked to produce detailed descriptions for one domain of mental illness, such as anxiety disorders, mood disorders, or sexual disorders. To fill these committees, Spitzer appointed psychiatrists who saw themselves primarily as scientists rather than clinicians and instructed them to scour published data relative to the establishment of possible diagnostic criteria—regardless of whether these data-based criteria aligned with
the traditional understanding of a disorder.
Spitzer threw himself into the creation of a new DSM with a fierce and focused energy. “I was working seven days a week, sometimes twelve hours a day,” he recollects. “Sometimes I would wake up Janet in the middle of the night asking for her opinion on a point, and then she’d get up and we’d work together.” Spitzer’s wife, Janet Williams, who has a doctorate in social work and is a leading expert in diagnostic assessment, confirms that the DSM-III was an all-consuming project for them both. “He answered every letter the Task Force received while he was working on the DSM-III, and responded to every critical article about it, no matter how obscure the journal—and remember, this was before computers,” Janet recounts. “Fortunately, we were very fast typists.” Jean Endicott, a psychologist who worked closely with Spitzer, remembers, “He would come in on Mondays having clearly worked on the DSM all weekend. If you sat by him on the plane, there was no question what you were going to be talking about.”
Spitzer soon proposed an idea that—if adopted—would fundamentally and irrevocably alter the medical definition of mental illness. He suggested dropping the one criterion that psychoanalysts had long considered essential when diagnosing a patient’s illness: the cause of the illness, or what physicians term etiology. Ever since Freud, psychoanalysts believed that mental illness was caused by unconscious conflicts. Identify the conflicts and you would identify the illness, ran the venerable Freudian doctrine. Spitzer rejected this approach. He shared the Washington University group’s view that there was no evidence to support the cause of any mental illness (other than addictions). He wanted to expunge all references to etiology that weren’t backed up by hard data. The rest of the Task Force unanimously agreed.
To replace causes, Spitzer laid down two new essential criteria for any diagnosis: (1) the symptoms must be distressing to the individual or the symptoms must impair the individual’s ability to function (this was the “subjective distress” criteria he had first proposed while fighting to depathologize homosexuality), and (2) the symptoms must be enduring (so if you were gloomy for a day after your pet hamster died, this would not constitute depression).
This was a definition of mental illness radically different from anything before. Not only was it far removed from the psychoanalytical view that a patient’s mental illness could be hidden from the patient herself, but it also amended Emil Kraepelin’s definition, which made no reference to subjective distress and considered short-lived conditions to be illnesses, too.
Spitzer laid down a two-step process for diagnosing patients that was as simple as it was shockingly new: first, determine the presence (or absence) of specific symptoms and for how long they had been active; then, compare these observed symptoms to the fixed set of criteria for each disorder. If the symptoms matched the criteria, then a diagnosis was justified. That’s it. No ferreting around in a patient’s unconscious for clues to a diagnosis, no interpreting the latent symbolism of dreams—just identifying concrete behaviors, thoughts, and physiological manifestations.
The DSM-III Task Force learned very quickly that in order to remain faithful to the published data, it was often necessary to create rather complex sets of criteria. In the DSM-II, for example, schizophrenia was dealt with in a series of impressionistic descriptions, including this definition of paranoid schizophrenia:
This type of schizophrenia is characterized primarily by the presence of persecutory or grandiose delusions, often associated with hallucinations. Excessive religiosity is sometimes seen. The patient’s attitude is frequently hostile and aggressive, and his behavior tends to be consistent with the delusions.
By contrast, the DSM-III provided several sets and subsets of conditions that were required for a diagnosis of schizophrenia. Here, for example, is condition C:
C. At least three of the following manifestations must be present for a diagnosis of “definite” schizophrenia, and two for a diagnosis of “probable” schizophrenia. (1) Single. (2) Poor premorbid social adjustment or work history. (3) Family history of schizophrenia. (4) Absence of alcoholism or drug abuse within one year of onset of psychosis. (5) Onset of illness prior to age 40.
Critics quickly sneered at the complicated “Select one from criteria set A, select two from criteria set B” instructions, calling it the “Chinese menu” approach to diagnosis, after the multitiered menus that were common in Chinese restaurants at the time. Spitzer and the Task Force countered that this increased complexity in the diagnostic criteria matched the evidence-based reality of mental disorders far better than the ambiguous generalities of the DSM-II.
But there was one notable problem with the Task Force’s utopian vision of better psychiatry through science: For many disorders, the science hadn’t actually been done yet. How could Spitzer determine which symptoms constituted a disorder when so few psychiatrists outside of Washington University and a handful of other institutions were conducting rigorous research on symptoms? What the Task Force needed were cross-sectional and longitudinal studies of patients’ symptoms and how these patterns of symptoms persisted over time, how they ran in families, how they responded to treatment, and how they reacted to life events. While Spitzer insisted that the diagnoses be based on published data, such data were often in very short supply.
If there was not an extensive body of literature on a particular diagnosis, then the Task Force followed an orderly procedure. First they reached out to researchers for unpublished data or gray literature (technical reports, white papers, or other research not published in a peer-reviewed format). Next, they reached out to experts with experience with the tentative diagnosis. Finally, the entire Task Force would debate the putative criteria until they reached consensus. Spitzer told me, “We tried to make the criteria represent the best thinking of people who had the most expertise in the area. The guiding principle was that the criteria needed to be logical and rational.” The DSM-III added many new disorders, including attention-deficit disorder, autism, anorexia nervosa, bulimia, panic disorder, and post-traumatic stress disorder.
There was one overt nonscientific factor that influenced the new diagnostic criteria: ensuring that insurance companies would pay for treatments. Spitzer knew that insurance companies were already cutting back on mental health care benefits as a result of the antipsychiatry movement. To combat this, the DSM-III stressed that its criteria were not the ultimate diagnostic word but that “clinical judgment is of paramount importance in making a diagnosis.” They believed that this disclaimer would give psychiatrists protection against an insurance company intent on showing that a patient did not exactly conform to the criteria listed. In actuality, time has shown that insurance companies do not tend to challenge psychiatrists’ diagnoses—instead, they often challenge the choice and duration of treatment for a diagnosis.
The DSM-III represented a revolutionary approach to mental illness, neither psychodynamic nor biological, but able to incorporate new research from any theoretical camp. By rejecting causes (including neurosis) as diagnostic criteria, the DSM-III also represented a complete repudiation of psychoanalytic theory. Before the DSM-III, the Feighner Criteria had almost exclusively been used for academic research, rather than clinical practice. Now the DSM-III would render the Feighner Criteria the clinical law of the land. But first, there was one major hurdle to surmount, and it was a doozy.
The DSM-III would only be published by the APA if its members voted to approve it. In 1979, a strong and vocal majority of these members were psychoanalysts. How could Spitzer persuade them to endorse a book that ran counter to their approach and might spell their own doom?
The Showdown
Throughout his tenure, Spitzer transparently and continuously communicated the Task Force’s progress on the DSM-III via a steady stream of personal letters, meeting minutes, reports, bulletins, publications, and talks. Each time he made a public presentation or published an update on the DSM-III, he encountered pushback. At first, the criticism was relatively mild, sinc
e most psychiatrists had no vested interest in a new diagnostic manual. Gradually, as more and more was revealed about the contents of the DSM-III, the blowback intensified.
The turning point came in June 1976 at a special meeting in St. Louis (sponsored by the University of Missouri, not Washington University) with an audience of one hundred leaders in psychiatry and psychology. The DSM-III in Midstream, as the conference was called, marked the first time that many prominent psychoanalysts heard about Spitzer’s new vision for diagnosis. This was when the cat was finally let out of the bag. The meeting exploded in controversy. Attendees denounced what they viewed as a sterile system that purged the DSM of its intellectual substance, claiming Spitzer was turning the art of diagnosis into a mechanical exercise. Spitzer was frequently accosted in the corridors by psychoanalysts who demanded to know if he was intentionally setting out to destroy psychiatry, and by psychologists who demanded to know if he was deliberately attempting to marginalize their profession.
When it was over, influential groups mobilized to oppose Spitzer; he responded by throwing himself with redoubled energy into the task of responding to the opposition. Two of the most formidable opponents were the American Psychological Association, the largest professional organization of psychologists (sometimes referred to as “the big APA” since there are far more psychologists than psychiatrists in the U.S.), and the American Psychoanalytic Association (APsaA), still the largest professional organization of Freudian psychiatrists. One of the original goals of the DSM-III was to firmly establish that mental illness was a genuine medical condition in order to push back against the antipsychiatry movement’s contention that mental illness was merely a cultural label. But psychologists—therapists with a PhD instead of an MD—had benefited greatly from the antipsychiatry argument. If mental illness was a social phenomenon, as Szasz, Goffman, and Laing charged, then one didn’t need a medical degree to treat it: Anyone could justifiably use psychotherapy to guide a patient through her problems. If the American Psychiatric Association formally declared that mental illness was a medical disorder, psychologists stood to have their recent professional gains rolled back.
Shrinks Page 13