Recover password

This secret experiment tricked psychiatrists into diagnosing sane people as having schizophrenia

It was a secret experiment. There was a graduate student, a housewife, a painter, a pediatrician, a psychiatrist and three psychologists. Using fake names, they went out to 12 hospitals across the country and claimed to hear voices. Their mission was to see what would happen.

What they found rocked psychiatry.

David Rosenhan, a psychologist at Stanford University, published the results of the experiment in a 1973 issue of the journal Science. “On Being Sane in Insane Places” would become one of the most influential studies in the history of psychiatry.

According to Rosenhan, each of what he called the “pseudopatients” told hospital staff about hearing voices that used the words “empty,” “hollow” and “thud.” The pseudopatients claimed the voices were difficult to understand but sounded as if they came from the same sex as that of the fake patients. Other than making claims about voices and giving themselves phony names and false occupations, the pseudopatients – Rosenhan among them – made up nothing else. None of them had any significant history of mental illness.

All of them were admitted to psychiatric units, at which point they stopped reporting any psychiatric symptoms. Still, nearly every person in the experiment was diagnosed with schizophrenia. Their hospitalizations ranged from seven to 52 days. Doctors prescribed them more than 2,000 pills, including antipsychotics and antidepressants, which the pseudopatients largely discarded.

In the hospitals, staff often misinterpreted the pseudopatients’ behaviors to fit within the context of psychiatric treatment. For example, the pseudopatients took copious notes while studying the environment of the psychiatric ward. One nurse reportedly wrote in the chart, “Patient engages in writing behavior.”

Although none of the pseudopatients were unmasked by hospital staff, other patients on the psychiatric units became suspicious of them. Across several of these hospitalizations, 35 patients expressed doubts that the pseudopatients were actually mentally ill, according to the study.

Still, Rosenhan’s conclusions were stark: People feigning mental illness all gained admission to psychiatric units and, after they stopped faking symptoms, remained there for lengthy periods. He famously wrote, “It is clear that we cannot distinguish the sane from the insane in psychiatric hospitals.”

The study has been called a “disaster” and an “albatross” for psychiatry. Its findings generated intense debate over the validity of psychiatric diagnoses and practices, with critics heralding the study as proof of the inherent flaws in psychiatric care.

Many psychiatrists decried the study’s methods and conclusions. Robert Spitzer, who has been called the father of modern psychiatry, described the Rosenhan experiment as “pseudoscience presented as science.” Some readers pointed out that pseudopatients could just have easily fooled physicians by alleging symptoms of headaches or back pain.

In a letter to Science, Fred M. Hunter, a physician, criticized the idea that the experiment participants were kept in hospitals despite acting normally: “The pseudopatients did not behave normally. Had their behavior been normal, they would have walked to the nurses’ station and said, ‘Look, I am a normal person who tried to see if I could get into the hospital by behaving in a crazy way or saying crazy things. It worked and I was admitted to the hospital, but now I would like to be discharged from the hospital.’ ”

There is a term for patients who come into the hospital faking illness for a purposeful gain: It’s called malingering. Patients may malinger for different reasons, such as getting pain medications or a place to sleep in the hospital. And it can be difficult to distinguish these complaints from genuine suffering, whether in psychiatry or any other medical specialty.

In 2000, William Reid, a psychiatrist, published a journal article about malingering. He wrote that “most of the commonly held axioms about separating real from bogus patients don’t hold up under scrutiny. Liars don’t reliably fidget or blink more, avoid eye contact, or use less detail in their explanations.” Spotting a malingerer often requires extended observation, suspicion that a patient may be feigning illness, and collaboration across multiple providers.

Rosenhan’s critique dealt a blow to the validity of psychiatric diagnoses, and more recent studies have found that different psychiatrists will frequently come to different conclusions when asked to diagnose the same patients. Yet, research also suggests that psychiatry isn’t all that different from other medical specialties: In other words, doctors may be no better at diagnosing non-mental-health conditions such as strokes or osteoarthritis.

Mental-health care was considerably different at the time of the Rosenhan experiment. The pseudopatients appeared to easily find available psychiatric hospital beds. There is no mention of the costs of care or insurance in the paper. Rosenhan described horrific conditions, with staff members beating patients and shouting profanities as if by routine.

In the decades since the study was published, shortages of psychiatric beds have become a national crisis, with lengthy waits for inpatient care. Psychiatric treatment is now unaffordable for many, even those with insurance. On the brighter side, state legislatures and Congress have passed numerous laws to protect patients receiving inpatient psychiatric care.

The Rosenhan study gave the impression that patients could go to a hospital, claim to hear voices and stroll into any psychiatric unit. But this is far from how mental-health care is practiced these days.

Here’s how a hospital today might approach a patient who arrives in an emergency department complaining of hallucinations. One or more nurses would take vital signs, complete a brief exam and gather some of the patient’s history. At least one emergency physician would repeat the process, interviewing and examining the patient again. Lab tests might include electrolyte panels, blood counts, alcohol level, thyroid levels and urine studies to screen for drugs or infection. The emergency physicians might order a CT scan of the head or other imaging, depending on the patient’s history.

If all of that work-up is unremarkable, the emergency department team may consider a psychiatry consultation if the hospital has psychiatrists on staff.

In addition to interviewing and examining the patient, a psychiatrist would review the patient’s chart and any available electronic records from other facilities for background information. If family or friends are available, psychiatrists try to speak with them. And psychiatric teams try to get in touch with previous providers or anyone else who might add insight to the situation. From start to finish, these evaluations can take hours.

Today, patients usually do not get admitted to psychiatric units just for saying, “I’m hearing voices.” To be hospitalized, patients need to have symptoms of a psychiatric disorder – such as hearing voices, suffering from depression and feeling suicidal – that are so profound as to cause safety concerns or significant impairments in daily living, such as dysfunction at work or home. If the patient were admitted to a psychiatric unit and then suddenly stopped having symptoms, it would be difficult to justify keeping that person on the unit. Insurance would stop paying for the hospitalization. Every day, physicians have to document why someone needs be treated in a hospital rather than in an outpatient setting.

Rosenhan’s findings may come across to hospital leaders as something that “could never have happened in my hospital,” and Rosenhan addressed this criticism in his original paper. The staff at one teaching hospital apparently heard about his study and believed that they would not make similar mistakes with pseudopatients. According to Rosenhan, he challenged them to spot pseudopatients that he would send to their hospital. The staff later claimed with a high degree of confidence to have identified 41 pseudopatients – then Rosenhan revealed that he had not sent any at all.

But some studies that attempted to replicate Rosenhan’s findings have, in fact, reinforced the notion that the psychiatry of today is not the same psychiatry as before. A small 2001 study, for instance, followed seven people with well-documented histories of schizophrenia who were actually in crisis and presented themselves to mental-health intake offices; six of the seven were denied treatment, often because of a lack of resources. In a 2004 book, psychologist Lauren Slater claimed to have gone to nine emergency departments and complained of hearing voices, as in the Rosenhan experiment. Although she reports having been prescribed various medications, she says she was not admitted to a single facility.

Nearly half a century after its publication, the Rosenhan experiment has left a lasting impression on psychiatry.

Researchers continue to propose methods of redoing the study – for example, by asking psychiatrists what they would do with similar patients or by sending actors in randomized trials to hospitals to feign hallucinations. While such studies may shed light on how mental-health care has changed since the 1970s, it’s already well established that catching impostors is a tall order in virtually any field of medicine.

Using fake patients was a radical way to expose the limitations of psychiatric treatment. But it’s a shame that doing so has sown lingering doubts about mental-health care.

Morris is a resident physician in psychiatry at the Stanford University School of Medicine.