Rosenhan: Sane in insane places
Aim: To test reliability and validity of diagnosis by seeing whether people without a mental disorder would be admitted to a psychiatric hospital; and if so could this decision be reversed
Procedure: 8 pseudo-patients (including Rosenhan) went into 12 hospitals across the USA. The pseudo-patients called the hospitals saying they were hearing voices ‘ thud, hollow, empty’. All other details, apart from their name and in some cases occupation, were kept the same. Once admitted, the pseudo-patients said the voices had stopped. They measured the length of time patients were admitted to hospital, and some pseudo-patients kept a diary of their experience.
Findings: All pseudo-patients were admitted. Time spent in the institute was 7 – 52 days – 19 days as the average.
7 were diagnosed with schizophrenia, one with bipolar
35/118 genuine patients were able to identify them as ‘sane’ e.g. ‘you’re a journalist’
Over 2100 pills were given out – only 2 were taken
Nurses in office 90% of the time, 7 mins per day per patient
Behaviour interpreted based on the label of mental disorder Waiting outside the cafe for lunch = ORAL ACQUISITIVE SYNDROME; Writing in diaries = PATIENT ENGAGES IN PATHOLOGICAL WRITING BEHAVIOUR; Walking the corridor out of boredom = NERVOUSNESS
In a follow up study with one hospital they were told pseudo patients would be sent and to make judgements of confidence on patients being a pseudo patient - NO pseudo patients were actually sent BUT hospitals identified 41 pseudo patients
Conclusion: The diagnoses of mental illness is influenced by the context in which the behaviours are observed and the expectation of others. Once a patient has been labelled with a mental illness, it is hard to remove the label
Evaluation:
Generalisability: 12 hospitals used so can be generalised, not just one hospital labelling patients; However this was conducted in the 1970’s and only in the USA – may be cultural differences
Reliability: 8 people in 12 hospitals using the same procedure e.g. hearing ‘thud, hollow, empty’ – replicable and reliable. The number of days is also an objective measure, however it could be considered that as the pseudo patients were also the ones making notes they were being subjective
Application: Shows the effects of labelling on diagnosis but this was in the 1970’s with more emphasis on understanding mental disorders and community care
Validity: patients were being themselves with the exception of hearing voices which stopped once admitted; genuine patients noticed they were ‘normal’ so this suggests there was validity. However as staff were told they were hearing voices, a common sign of schizophrenia, it could be expected they would be admitted and be given this diagnosis. They also made the phone call asking to be admitted! Also with exp 2 as hospitals believed pseudo-patients would be arriving this may have lead them to be looking for signs (demand characteristics) based on this lie
Ethics: staff would have been left feeling embarrassed that they were wrong in their diagnosis, they would not have consented to the study. However staff and hospitals were not named and it would not have been just one individual making the incorrect diagnosis
Aim: To test reliability and validity of diagnosis by seeing whether people without a mental disorder would be admitted to a psychiatric hospital; and if so could this decision be reversed
Procedure: 8 pseudo-patients (including Rosenhan) went into 12 hospitals across the USA. The pseudo-patients called the hospitals saying they were hearing voices ‘ thud, hollow, empty’. All other details, apart from their name and in some cases occupation, were kept the same. Once admitted, the pseudo-patients said the voices had stopped. They measured the length of time patients were admitted to hospital, and some pseudo-patients kept a diary of their experience.
Findings: All pseudo-patients were admitted. Time spent in the institute was 7 – 52 days – 19 days as the average.
7 were diagnosed with schizophrenia, one with bipolar
35/118 genuine patients were able to identify them as ‘sane’ e.g. ‘you’re a journalist’
Over 2100 pills were given out – only 2 were taken
Nurses in office 90% of the time, 7 mins per day per patient
Behaviour interpreted based on the label of mental disorder Waiting outside the cafe for lunch = ORAL ACQUISITIVE SYNDROME; Writing in diaries = PATIENT ENGAGES IN PATHOLOGICAL WRITING BEHAVIOUR; Walking the corridor out of boredom = NERVOUSNESS
In a follow up study with one hospital they were told pseudo patients would be sent and to make judgements of confidence on patients being a pseudo patient - NO pseudo patients were actually sent BUT hospitals identified 41 pseudo patients
Conclusion: The diagnoses of mental illness is influenced by the context in which the behaviours are observed and the expectation of others. Once a patient has been labelled with a mental illness, it is hard to remove the label
Evaluation:
Generalisability: 12 hospitals used so can be generalised, not just one hospital labelling patients; However this was conducted in the 1970’s and only in the USA – may be cultural differences
Reliability: 8 people in 12 hospitals using the same procedure e.g. hearing ‘thud, hollow, empty’ – replicable and reliable. The number of days is also an objective measure, however it could be considered that as the pseudo patients were also the ones making notes they were being subjective
Application: Shows the effects of labelling on diagnosis but this was in the 1970’s with more emphasis on understanding mental disorders and community care
Validity: patients were being themselves with the exception of hearing voices which stopped once admitted; genuine patients noticed they were ‘normal’ so this suggests there was validity. However as staff were told they were hearing voices, a common sign of schizophrenia, it could be expected they would be admitted and be given this diagnosis. They also made the phone call asking to be admitted! Also with exp 2 as hospitals believed pseudo-patients would be arriving this may have lead them to be looking for signs (demand characteristics) based on this lie
Ethics: staff would have been left feeling embarrassed that they were wrong in their diagnosis, they would not have consented to the study. However staff and hospitals were not named and it would not have been just one individual making the incorrect diagnosis