Pseudopatients and their discontents: an historical perspective.
by Susannah Cahalan; New York: Grand Central Publishing, 2019
383 pages • $28.00 (hardcover)
Dr Kaplanis Clinical Associate Professor, Graduate School of Medicine, University of Wollongong, Australia.
In 1973 the psychiatric profession was deeply shaken by a paper published in Science that purported to show that psychiatric diagnosis was effectively useless.1 The author, Stanford psychologist David Rosenhan, described a unique experiment: Eight volunteer “pseudopatients” presented at psychiatric hospitals under fake names, complaining they heard voices. The intention was to examine psychiatric diagnosis: was it scientifically valid or merely a random, subjective, and erratic process?
The pseudopatients were in the hospital for a mean of 19 days. All but one were diagnosed with schizophrenia in remission. The statistics from the study showed contact with doctors averaged just 6.8 minutes a day; 71% of doctors averted their heads when addressed. While patients suspected study participants were fake-possibly because they spent their time writing notes-the staff did not.
Rosenhan’s message was devastating: “We cannot distinguish the sane from the insane in psychiatric hospitals,” adding, “If sanity and insanity exist, how shall we know them?” These findings, taken at face value, were very difficult to refute.
Pseudopatients and their discontents: an historical perspective
Arguably the most influential psychological paper published in the last half of the 20th century, it caused a sensation. By the 1980s, it was mentioned in most psychology textbooks. Rosenhan became a star in his field and the study still one of the most cited social science papers, as well as prescribed reading in psychology and social work courses.
The timing was good; the anti-psychiatry movement started in the 1960s by Szasz, Laing, Goffman, and others had kicked into high gear.2 The movie One Flew Over the Cuckoo’s Nest played its part in persuading the counterculture to see psychiatry as inherently oppressive. Michael Foucault, the French intellectual superstar, said that asylums represented the punitive arm of society (based on his rather dubious historical research).
The response of psychiatrists to the paper was predictable. They rushed to defend their profession, although with varying degrees of conviction. It led to a vortex of discussion about the practice of psychiatry and its future. It fed into the deinstitionalization movement, an agenda driven by governments, radicals, the counterculture, and others. Although not realized at the time, the results were catastrophic. Community services could never come close to meeting the needs of the discharged patients and the vacuum was filled by the streets and prisons.
The fallout from the research played a part in the organizing principles that led to the Diagnostic & Statistical Manual (DSM-III) in 1980. For the first time, diagnoses were categorized with listed symptoms, free of etiological presumptions, notably psychoanalysis.
This brought with it an unexpected and unprecedented benefit. In a triumph of medical marketing, the American Psychiatric Association (APA) created a brand that may be as well-known as Apple or Coca Cola, culminating in its latest incarnation in 2013 with DSM-5. The APA made many millions and can expect the rivers of gold to keep flowing. But the solution was not megabucks alone. Has DSM solved the problems, not least epistemological, that beset psychiatric diagnoses? Hardly, if the furore of DSM-5 is any indication.
The paper did not lack supporters and news coverage. In Australia, an attempt was made to replicate it. Radical Sydney psychologist Dr Robin Winkler did two studies, published simultaneously in the Medical Journal of Australia in September 1974.3,4 The first paper reported the findings of pseudopatients who had visited general practitioners claiming to suffer depression; in the second article the infiltrators had themselves admitted to psychiatric hospitals.
Winkler concluded that “a permanent body should be established with the co-operation of medical, research and consumer organisations, to report regularly on health care services as evaluated with pseudopatient observations.” Psychiatrist Neil McConaghy, who had a distinguished research record, was skeptical.5 While there could be some value in such studies, Winkler had failed to account for observer bias, if not other issues.
This brings us to Susannah Cahalan
She first wrote of her psychiatric experience in Brain on Fire, the account of how psychotic symptoms developed. She was seen by a number of psychiatrists and set to be admitted to a psychiatric hospital. A neurologist, however, discovered she had a rare autoimmune condition: anti-NMDA receptor encephalitis. She received treatment and recovered.
Anti-NMDA receptor encephalitis is extremely rare with 1.5 cases per million people (about 80% female) per year,6 and psychiatrists can hardly be blamed for not picking it up at the time. Cahalan, understandably, does not see this as an excuse and describes another undiagnosed case where the patient suffered brain damage.
The experience lingered; had she not had the neurological intervention, she would have suffered permanent brain damage. This did not dispose her favorably toward psychiatric science and led her to David Rosenhan. Rosenhan had died in 2012 but Cahalan was able to get access to his notes. Starting the journey in a spirit of optimism, she expected it to be a positive experience, but her hopes were soon dashed.
An early warning came from a colleague who called Rosenhan a “bullshitter.” Another, Eleanor Maccoby, was deeply suspicious saying it was “impossible to know what he had really done, or if he had done it,” according to Cahalan. The paper was his only significant work and he lived off it for the rest of his career.7 One thing that puzzled Cahalan was his failure to complete a book on the study, commissioned with a lucrative advance payment which he had to return. The explanation for this was to come later as a denouement to the investigation.
It was known that Rosenhan was the first member of the pseudopatient group. Was it ethical or appropriate for the person running the study to participate in it? He was admitted as “David Lurie” to Haverford State Hospital in Pennsylvania. Cahalan found the notes of his interview with psychiatrist Frank Bartlett. He noted that Rosenhan had been suicidal, finding the voices so upsetting he put copper pots over his ears to tune them out and having symptoms for months. Rosenhan also said that he was sensitive to radio waves and could hear what people were thinking.
These are severe symptoms that every psychiatrist would take seriously. This is what Bartlett did; his admission notes conveyed a more detailed and disturbing picture of mental illness than Rosenhan said the pseudopatients had presented. Cahalan’s damning response: “Dr Bartlett wasn’t a bad doctor who made a bad decision . . . He was a good doctor who made the best call given the information he received.”
As her doubts escalated, she questioned whether the other pseudopatients even existed: “It was becoming alarmingly clear that the facts were distorted intentionally-by Rosenhan himself,” she writes.
Cahalan tracked down Bill Underwood. Such was his emotional poise that Rosenhan questioned whether he could pass for a psychiatric patient, again showing the experimenter’s bias. As it happens, Underwood was admitted for 9 days with a diagnosis of paranoid schizophrenia. While Rosenhan stated that he carefully prepared his volunteers for the admissions, Underwood only recalled having brief guidance on how to hide pills in his cheek. The paper stated that Underwood had spent 7 days in a hospital with 8000 patients, but in fact he spent 8 days in a hospital with 1500 patients.
Harry Lando, pseudopatient No. 9, was cut from the study on the grounds that he “falsified aspects of his personal history.” He spent 19 days at an institution where the staff didn’t wear uniforms, he went to group therapy, and he had a beach day trip. Lando told Cahalan he was treated well and enjoyed the experience. This was not what Rosenhan wanted to hear. The notes state: “HE LIKES IT.” The result? Lando was excised from the study because his experience had been positive.
The identity of the other pseudopatients was a mystery, and Cahalan concluded that they were wholly fabricated, a finding that eliminates the likelihood there were just a few correctable mistakes in the study. In fact, another pseudopatient’s false details were inserted in his account. So much for including all study findings regardless of whether they supported the hypothesis.
Rosenhan could dismiss most objections by psychiatrists, but Robert Spitzer, as a leading figure behind DSM-III, was on a different level.8 Spitzer wrote, “Some foods taste delicious but leave a bad aftertaste.” The paper was “pseudoscience presented as science” and the conclusion a diagnosis of “logic in remission.” Rosenhan’s research methods were “unscientific”; the terms “sanity and insanity” were legal concepts, not psychiatric diagnoses; and the designation “in remission,” showed that the doctors were aware that the pseudopatients were different from the rest. Rosenhan failed to disclose his data and his sources, wilfully withholding information from readers.
These were penetrating and accurate criticisms which Cahalan followed in Rosenhan’s file. He became increasingly defensive as the exchange between the two progressed, enraged at being called out in this fashion. Cahalan put her finger on it: this was the reason he did not write the book commissioned on the study and had to pay back the lucrative advance fee. Rosenhan knew he was on thin the ice; being called out would be the end of his career.
One motivation for the experiment not considered was one of the oldest: turf war. Psychologists, especially at that time, were excluded from many activities on which psychiatrists had a “monopoly.” Discrediting their practice would expand the opportunities for psychologists and social workers.
Just as the paper was deeply damaging for psychiatry when it was published, it has now rebounded on psychology. Rosenhan’s effort, its deception unknown at the time, was the predecessor of the current replication crisis in psychology in which serious doubts have been raised about the work of Diederik Stapel, Philip Zimbardo, and Stanley Milgram.
Andrew Scull, hardly an admirer of psychiatry, sums up the situation.9 Rosenhan pulled off one of the greatest scientific frauds of the past 75 years, and its consequences still resonate today. One only has to look at the individuals crowding our streets and prisons, the result of deinstitutionalization, for which he should take much credit.
While some may react to the frequent insertions of her own experience, this is a minor criticism. Cahalan is to be congratulated on her meticulous investigation.
1. Rosenhan DL. On being sane in insane places. Science. 1973;179:250-258.
2. Laffey P. Histories of Psychiatry after Deinstitutionalisation: Australia and New Zealand. Health History. 2003;5:17-36.
3. Owen A, Winkler RC. General practitioners and psychosocial problems: An evaluation using pseudopatients. Med J Aust. 1974;2:393–398.
4. Winkler RC. Research into Mental Health Practice Using Pseudopatients. Med J Aust. 1974;2:399–403.
5. McConaghy N. Pseudopatients and Evaluation of Medical Practice. Med J Aust. 1974;2:383-385.
6. Kayser MS, Dalmau J. Anti-NMDA Receptor Encephalitis in Psychiatry. Curr Psychiatry Rev. 2011;7:189–193.
7. Guide to the David L. Rosenhan Papers. Online Archives of California. https://oac.cdlib.org/findaid/ark:/13030/c8tq6082/entire_text. Accessed March 4, 2020.
8. Spitzer RL. On pseudoscience in science, logic in remission, and psychiatric diagnosis: A critique of Rosenhan's “On being sane in insane places.” J Abnorm Psychol. 1975;84:442–452.
9. Scull A. How a fraudulent experiment set psychiatry back decades. The Spectator. January 25, 2020. https://www.spectator.co.uk/2020/01/how-david-rosenhans-fraudulent-thud-experiment-set-back-psychiatry-for-decades. Accessed March 4, 2020.