Shrinks (14 page)

It was the American military that finally came to psychiatry’s aid.

Psychotic Soldiers

As the American military recruited ever-increasing numbers of soldiers to fight in the Second World War, it encountered a puzzling problem. Each potential recruit was evaluated by a military doctor to determine if he was fit to serve. Military officials expected that the rates of rejection for medical reasons would be consistent from state to state, but when they reviewed the actual rejection rates around the country, they were surprised to find that the rejection rates varied wildly. A draft board in Wichita might have a 20 percent rejection rate, while a draft board in Baltimore might reject 60 percent of its applicants. When military officials looked more closely at the problem, they realized that this variability was not due to physical conditions, such as flat feet or heart murmurs, it was due to dramatic differences in the way each doctor judged recruits to be mentally ill.

The military had not considered the consequences of applying contemporary methods of psychiatric diagnosis to the evaluation of conscripted soldiers. If a military doctor found a potential recruit unfit for service, he needed to specify the precise diagnosis that rendered the draftee ineligible… but, of course, Freud-influenced psychiatrists were not used to establishing precise diagnoses. Each psychoanalytical psychiatrist employed his own idiosyncratic interpretation of buried conflicts and neuroses. Even non-Freudians could not refer to an obvious diagnostic system to use to justify their rejections. While many non-Freudians relied on the
Standard
, that manual had been developed to gather statistics on institutionalized patients; it was never designed to diagnose mental illnesses that might be found in the wider community, and it was certainly never intended to evaluate the ability of potential soldiers to function in combat.

Recruits who exhibited behavior that was perceived as problematic in a military setting—such as an inability to pay attention or hostility toward authority—were often shoehorned into a category like “Psychopathic Personality.” Some draft boards saw as many as 40 percent of volunteers rejected because of “psychosis.”

In hopes of establishing a consistent and comprehensive system for evaluating the mental health of potential recruits, the army convened a committee in 1941 headed by William Menninger, former president of the American Psychiatric Association and cofounder of the Menninger Clinic, to develop a set of clearly defined diagnoses for mental illness that could be used to determine whether a given candidate was fit to serve. (Ironically, William’s brother Karl, who cofounded the Menninger Clinic, wrote in his book
The Vital Balance
, “There is only one class of mental illness—namely, mental illness. Thus, diagnostic nomenclature is not only useless but restrictive and obstructive.”)

Menninger issued his new psychiatric classification system in 1943 as a twenty-eight-page War Department Technical Bulletin that became known as the
Medical 203
, after its bulletin number. It was immediately implemented as the official manual for diagnosing both recruits and soldiers in the American military. The
Medical 203
described about sixty disorders and represented a landmark in clinical psychiatry: It was the first diagnostic system that classified every known form of mental illness, including serious disorders found in patients in mental institutions and mild neuroses found in patients who could function effectively in society.

At last, here was a comprehensive roadmap for diagnosing mental illness—and yet the
Medical 203
was almost completely ignored by civilian psychiatrists. For shrinks seeing patients in their private practices, the prevailing sentiment was, “I didn’t need a pointless classification manual before the war, and I certainly don’t need one now.” Psychoanalysts continued to use their own creative diagnoses, while asylum psychiatrists and teaching centers continued to rely on the
Standard
or some local variant.

After the war ended, American psychiatry remained a patchwork of diagnostic systems. Imagine a medical world where military physicians defined heart attacks one way, universities defined heart attacks another way, hospitals defined them yet another way, while primary care physicians suggested that since everybody’s heart was sick to some degree, heart attacks didn’t really exist at all. American psychiatry was experiencing a crisis of reliability.

In a famous study in 1949, three psychiatrists independently interviewed the same thirty-five patients and independently came up with their own diagnosis for each patient. They ended up agreeing on the same diagnosis for a given patient (such as “manic-depressive illness”) only 20 percent of the time. (Consider how frustrated you might feel if oncologists only agreed that the mole on your arm was skin cancer 20 percent of the time.) The leaders of the American Psychiatric Association recognized that this unsettling lack of reliability would eventually undermine the public credibility of psychiatry. Despite the protestations of many psychoanalysts, in 1950 the APA formed a Committee on Nomenclature and Statistics tasked with the development of a diagnostic system that would standardize the classification of mental illness within civilian psychiatry once and for all. Unlike the
Standard
, this new system would include diagnoses relevant to private practice, the illnesses that shrinks saw (or believed they saw) every day in their offices.

The committee took the
Medical 203
as its starting point, lifting many passages of text directly from Menninger’s military bulletin. At the same time, the committee also sought to establish continuity with the
Standard
by borrowing the phrase “Statistical Manual” from its title. In 1952, the APA published the new system as the very first
Diagnostic and Statistical Manual of Mental Disorders
, today known as the
DSM-I
. It listed 106 mental disorders—an expansion from the 22 disorders in the
Standard
and the 60 disorders in the
Medical 203
. It relied heavily on psychoanalytical concepts, most obviously in the names of disorders, which were referred to as “reactions,” a term originating with the psychoanalyst Adolf Meyer, who oversaw the creation of the
DSM-I
while he was president of the APA and who believed that mental illness arose from maladaptive habits in response to the stressors of life. According to Meyer, mental illness should be diagnosed by identifying a patient’s unique stressors and the patient’s responses to them. Schizophrenia, for example, was a bundle of unruly reactions to the stresses and challenges of life. This perspective was codified in the
DSM-I
’s description of psychotic reactions: “a psychotic reaction may be defined as one in which the personality, in its struggle for adjustment to internal and external stresses, utilizes severe affective disturbance, profound autism and withdrawal from reality, and/or formation of delusions or hallucinations.”

For a medical specialty that had splintered into an anarchy of institution-specific definitions of illness, at last there was a single unifying document that could be used in any psychiatric setting, whether in an Arkansas state mental institution, an analyst’s office on the Upper East Side of Manhattan, or a medical unit on the front lines in Korea. The
DSM-I
represented a necessary first step in the unification and standardization of psychiatric medicine.

But it was also a precarious first step, since none of the diagnoses in the
DSM-I
were based upon scientific evidence or empirical research. They reflected the consensus of a committee that mostly consisted of practicing psychoanalysts, rather than researchers. It would not be long before the egregious shortcomings of the
DSM
would be exposed for the entire world to see.

On Being Sane in Insane Places

By the time I started medical school in 1970, the second edition of the
DSM
was in use.
DSM-II
was a rather thin spiral-bound paperback that cost $3.50. It had been published in 1968 to little fanfare, contained 182 disorders (almost double the number in
DSM-I
), and was every bit as vague and inconsistent as its predecessor.
DSM-II
had dropped the term
reactions
but retained the term
neuroses
. I only learned these facts later; while I was in medical school, I hardly laid eyes on the
DSM-II
—and neither did most psychiatry and psychology trainees.

Instead, I invested in an expensive black tome titled
The Comprehensive Textbook of Psychiatry
, a far more common reference book. This volume contained a potpourri of information from anthropology, sociology, and psychology—all mixed together with a heavy dose of psychoanalytic theory, of course. It still contained chapters on sleep therapy, insulin coma therapy, and lobotomies, while only 130 of its 1,600 pages contained references to the brain or neuroscience.

Most of what we learned in medical school did not come from books but from our instructors, each of whom purveyed his own interpretation of psychiatric diagnosis. One day, after we interviewed a young man who was manifestly psychotic, my professor began discussing the patient’s characteristics by way of formulating his diagnosis. In doing so he declared that the patient had the characteristic “smell of schizophrenia.” At first I thought he was using scent as a metaphor, like you might refer to the “sweet smell of success.” Eventually I realized that, like a psychiatric bloodhound, he believed his refined nose and olfaction could detect the apparently rather earthy aroma of the schizophrenic.

Other professors improvised their own methods of diagnosis like jazz musicians riffing on a melody, and encouraged us to follow their lead. Although this approach certainly respected the individual concerns and experiences of each patient—and liberated a clinician’s creativity—it did not foster diagnostic consistency. Confusing an impressionable young psychiatrist still further was a bevy of diagnostic frameworks that had splintered from Freudian theory: Adlerian, Jungian, Sullivanian, Kleinian, Kohutian, and many others, each coming from creative thinkers who were persuasive orators and charismatic personalities. It was as if the professional influence of each new diagnostic model radiated directly out of the dash and verve of its creator’s persona, rather than arising from any scientific discovery or body of evidence. When it came to clinical sway in the 1970s, the
DSM
was completely eclipsed by its cultish competitors.

Most shrinks, of course, didn’t view this as a problem. So what if there was an anarchy of philosophies about mental illness—that means I’m free to choose the one that best suits my own style! There was very little in the way of accountability and even less concern that the profession lacked anything that remotely resembled a set of “best practices.” This complacent attitude would be shattered by a study that slammed into psychiatry with the force of a battering ram.

In 1973 a sensational exposé appeared in the normally staid columns of the prestigious journal
Science
. A few pages after papers with technical titles like “Earliest Radiocarbon Dates for Domesticated Animals” and “Gene Flow and Population Differentiation” came a real attention-grabber: “On Being Sane in Insane Places.” The author was David Rosenhan. He was a little-known Stanford-trained lawyer who had recently obtained a psychology degree but lacked any clinical experience. His very first sentence made it clear that he intended to tackle one of the most basic questions for any medicine that laid claim to the mind: “If sanity and insanity exist, how shall we know them?”

Rosenhan proposed an experiment to determine how American psychiatry answered this question. Suppose perfectly normal people with no history of mental illness were admitted to a mental hospital. Would they be discovered to be sane? If so, how? Rosenhan did not merely offer this up as a thought experiment—he proceeded to share the results of an extraordinary study he had conducted over the previous year.

Unbeknownst to the hospital staffs, Rosenhan had engineered the secret admission of eight sane people to twelve different mental hospitals in five separate states on the East and West Coasts (some of his confederates were admitted to multiple hospitals). These “pseudopatients” used fake identities that varied their age and profession. At each hospital, they telephoned ahead for an appointment, and when they arrived they complained of hearing voices that uttered three words: “empty,” “hollow,” and “thud.”

In every instance, the pseudopatient was voluntarily admitted to the hospital. Once the confederates reached the psychiatry ward, they were instructed to say to the staff that they were no longer hearing voices (though they never shared that they had faked their symptoms to get admitted). They proceeded to act normally, presumably not evincing any symptoms of illness. Their behavior on the wards was variously recorded by nurses as “friendly,” “cooperative,” and “exhibiting no abnormal indications.”