AP - Diagnosis validity reliability Flashcards
Validity - Heterogeneity
• Heterogeneity of clinical presentation
○ One disorder can manifest itself differently in different patients.
§ Patients need to display a “quota” of symptoms under a mental illness for diagnosis.
□ 5 for MDD
Validity - symptomology reliance
• Classification based of symptomology rather than etiology
○ For physicaly disease we use etiology than symptomology.
○ Knowledge of etiology too ambigous
○ Have to rely on symptoms
§ Overlap in dignoses
Validity - comorbidity
• Comorbidity ○ Co-occurrences of diagnosis § E.g GAD normally occurs with depression. ○ Very extensive in the DSM-3 ○ Hard to categorise § Same etiology? § Same disorder?
validity - stability
• Stability of symptoms
○ Exclude the chance the symptoms were a one time occurrence from other factors.
validity - cut off point
• Cut-off point
○ When is an action considered a symptom?
§ Clinically significant
§ Clinically insignificant
validity - treatment
• Treatment
○ Treatment corresponds to mental disorders
○ Important to identify the problem correctly
Rosenham - aim
○ Investigate if psychiatrists could tell the difference between sane and insane people.
Rosenham- sample
8 mentally healthy subjects
Rosenham- method
§ The sample were to serve as pseudo-patients and seek admission to psychiatric hospitals.
§ They were to present themselves to 12 different hospitals in 5 US states.
§ They were to follow standardized instructions
□ Made an appointment
□ Complain of hearing voices saying the words:
® “Empty”
® “hollow”
® “thud”
□ Only symptoms they made up.
□ Would be made to provide false personal information.
□ Would have to convince staff of being sane.
□ Upon admission to the ward all symptoms were ceased by the participants.
□ Interacted normally with staff.
□ Hospital records described them as friend and cooperative.
□ All instruction were followed.
® Medication was not swallowed.
□ Almost imeddiately after admission they sought to be discharged from the hospital.
□ All but one participant was diagnosed with schizophrenia.
® Each were discharged with schizphrenia in remission
Rosenham- results
○ Length of stay varied from 7 days to 52 days.
§ Mean of 19 days
○ No pseudo-patient was caught by hospital staff.
○ Many other patients identified the pseudo-patients as journalists or professors
○ Normal behaviour was overrepresented as an illness.
§ Writing notes was seen as “oral-acquisitive syndrome”
○ Pseudo-patients after admission were largely ignored by staff.
§ 71% ignored and moved on from questions.
§ 4% stopped and talked.
§ Nurses stayed in ward 90% of the time
Time spent by staff on each patient < 7 minutes
Rosenham - follow up
○ Informed the hospital pseudo-patients will attempt to seek admission.
○ Asked to use a 10 point scale to rate likelihood of pseudo-patient.
○ Results
§ 43 of 193 patients judged were confidently rated as pseudo-patient.
No pseudo-patients had been sent to the hospital at all.
Rosenham- conclusion
○ Psychiatrists could not distinguish normal people form mentally ill.
○ Stickiness of a Diagnostic label
§ Everything you do is associated with your diagnosis.
§ Powerful schema creators
“stereotype threat”
Rosenham - evaluation
○ DSM 2 was in use ○ Participant observtions may be biased. ○ Cost-Benfit § Admitting healthy > releasing ill ○ Major deception
beck et al (1965)
• Agreement on specific diagnosis for 153 patients between two psychiatrists was only 54%.
Kendall et al (1974)
• Studied 2000 patients who were admitted in 1964 then in 1969.
• Schizophrenia was re-diagnosed as a form of depression than the other way around.
• The study showed that diagnosis was not very consistent over time.
• There are two explanations:
○ The patients changed the pattern of their abnormal behavior
§ Depression and Schizophrenia are two separate and different disorders caused by different factors.
○ Psychiatrists were not applying diagnostic criteria in a standardized and coherent way.
• Began to shift the DSM from explanation to observation.
