Rosenham (1973) On being sane in insane places

To investigate the validity and reliability of the psychiatric diagnosis system

Study 1:

To see if normal and sane people, who presented themselves to a psychiatric hospital demonstrating one symptom, would be diagnosed as insane and admitted to a psychiatric hospital.

Study 2:

To see if hospital staff would misidentify genuinely insane patients as being sane

Field experiment: Admission of pseudo-patients to the hospitals  

• IV: Which of the 12 hospitals they tried to get admitted to
• DV: Whether or not they were admitted

Participant observation: Of staff and patients in the hospitals

Self-report:

1. Young lady asked 6 controlled questions to staff at Stanford University

2. Staff asked to rate the likelihood that patient was a pseudo using 10 point scale (Study 2)

Both quantitative and qualitative data was gathered

• 8 sane people
• 5 men, 3 women
• Various ages and occupations (most based around psychology)
• Rosenham was a pseudopatient
• Used fake names and occupations (when necessary) but told the truth otherwise which favours them not being admitted as they had no history of insanity
• Staff did not know they were pseudos apart from the hospital administrator and chief psychologist knowing about Rosenham

Pseudo patients became participant observers and the psychiatrists, doctors, nurses, patients and other members of staff in the 12 hospitals became the participants.

In the control experiment with the young lady, the staff at Stanford University became participants

Being admitted to hospital:

• Reported hearing voices which were unclear, unfamiliar, of the same sex and "empty", "hollow" and "thud"

Once in hospital:

• Immediately started to behave normally and told staff they no longer were experiencing symptoms
• Did not swallow medication
• Interacted with staff and patients
• Participated in-ward activities
• Began covertly recording observations but then wrote down observations openly
• They had the responsibility of persuading hospital staff that they were sane and could, therefore, be discharged
• In 4 of the hospitals, pseudo-patients approached staff and asked about being discharged, being eligible for grounds privileges, being presented at staff meeting but they never approached an individual staff member more than once a day

Control Study:

• The young lady approached individual faculty members at Stanford University and asked them 6 standardised questions including:
• -Looking for a doctor
• -Looking for a psychiatrist

Study 2:

• Staff at a research and training hospital (aware of study 1 results) were warned that one or more pseudopatients would present themselves over the next 3 months and try to get admitted but none actually did
• Staff were asked to rate all the patients using a 10 point scale on how likely they were to be a pseudopatient/ not genuinely insane

• All pseudopatients were admitted to the hospitals and diagnosed as schizophrenic (apart from 1: manic-depressive psychosis)
• Their normal behaviours were misinterpreted as being part of their illness
• Queuing for lunch early-oral-acquisitive nature of syndrome
• If they got upset or distressed it was seen as due to their illness and not the situation
• Their non-psychotic past was misinterpreted by doctors due to illness label
• Medication was flushed down the toilet by other patients, not just the pseudos
• Staff tended to keep themselves away from patients except for admin or practical duties, they stayed inside their glass quarters, the "cage"
• Doctors were even more remote
• Personal hygiene and waste evacuation was often monitored, W.Cs often had no doors
• Their personal files could be accessed by anyone at anytime
• Examinations conducted in semi-public rooms
• Several instances of physical abuse from staff when no other members of staff were around
• In the psychiatric hospital, most of the staff only briefly talked to them, averted their head, gave no response at all and most of them gave no eye contact

• Stayed in hospital between 8-52 days (average=19 days)
• 35/118 of the insane patients voiced their suspicions in the first 3 hospitalisations recognising they were fake
• Daytime nurses emerged on average 11.5 times per shift and evening nurses 9.4 times per shift
• Doctors only emerged on the ward 6.7 times on average a day
• The Stanford staff talked to lady 100% of the time whereas doctors and nurses only talked to the pseudopatients 4% and 0.5% (respectively) of the time
• When the student asked to see a psychiatrist, the rate dropped from 100% to 78%
• Out of the 193 genuine patients, 41 were judged with high confidence, to be a pseudo by at least 1 staff member.
• 23 were suspected by at least one psychiatrist and 19 by one psychiatrist and one other staff member

• Psychiatric diagnosis is extremely inaccurate
• The DSM IV had poor reliability
• Patients experience powerlessness and depersonalisation
• The sane can't be distinguished from the insane in psychiatric hospitals 
• Being labelled causes all subsequent behaviour to be misinterpreted and distorted as does being in a psychiatric hospital
• The diagnosis of mental illness is influenced by the situation an individual finds themselves in
• Hospitals seem to be special environments where patient are treated in ways that perpetuate any problems rather than support patients
• Mental health workers are insensitive to the feelings and emotions of patients
• Psychiatrists are unable to reliably identify sane pseudopatient (type 1 error-false positive) and also fail to reliably detect insanity (type 2 error-false negative)

• High ecological validity
• Rich qualitative data from unstructured observation
• All used the same symptoms to be admitted and all behaved normally once in hospital
• Range of hospitals used
  
• Covert=high validity and decrease in demand characteristics
• Confidentiality and prepared lawyers if needed for pseudos
• Small sample-easily skewed
  
• Didn't follow standardised procedures (data excluded from 1 pseudo for not following procedures)
• Possible loss of objectivity and empathised with patients so distorted their view of staff
• Pseudos were sane so doesn't show anything about how insane people are diagnosed (lowers validity)
• Low ecological validity as psychiatrist don't expect people to lie about mental illness in order to be admitted
• Deception, no consent and no right to withdraw from hospital staff or patients
  
• Pseudos had no protection and experienced stress and witnessed physical abuse

Usefulness: Huge impact on mental healthcare worldwide and caused hospitals to review their admission procedure and how they trained their staff to interact with patients. Major influence on reforming the DSM and making clearer guidelines

Nurture: The environment and the way in which we are nurtured affects our mental health and the treatment we receive from others. Once someone is labelled as mentally ill, all their behaviour is distorted and misinterpreted as being part of the illness

Situational: Situational factors can cause us to misinterpret behaviour and can affect the diagnosis of mental illness. The depersonalisation and powerlessness of patients within the hospital can also be explained by the fact that is an institution 

Free Will: The pseudos voluntarily tried to be admitted but they were not allowed to leave unless the staff approved it

Determinism: Having the label of a mental illness remains with you forever and carried a stigma which can affect how others treat you

Science: Fairly unscientific procedure with only a few standardised aspects so has low reliability