Reliability and validity of diagnosis/classification Flashcards
reliability includes (2)
inter-rater reliability, cultural differences
inter-rater reliability is measured using a…
kappa score
Reiger et al found an inter-rater reliability score of…
0.46
culture bias: Copeland gave a patient description to _ UK psychiatrists and _ US psychiatrists. _% of US said the patient had Sz, compared to _% from UK
194 UK, 134 US
69% US, 2% UK
Luhrmann et al found culture differences between…
voices heard
culture bias: Luhrmann interviewed… and found…
20 Ghana, 20 Indian, 20 US schizophrenics,
African and Indians reported positive experiences with voices, Americans had negative experiences
validity includes (3)
gender bias, symptom overlap, comorbidity
gender bias comes from _, a criticism of the _ is that some categories _ one gender and not the other
stereotypes, DSM, pathologise
gender bias: who found psychiatrists equated healthy adult behaviour to mentally healthy male behaviour
Broverman et al
symptom overlap: who found people with DID have more Sz symptoms than schizophrenics
Ellason and Ross
symptom overlap: Read found many people with Sz also…
have sufficient symptoms of other disorders so they could receive another diagnosis
what is comorbidity
when 2/more conditions occur in a patient
Buckley: found _% of patients have comorbid _ and _% have comorbid -
50% have comorbid depression and 47% have comorbid substance abuse
comorbidity: _% of the population have Sz, _% have OCD, they are two very _ conditions however, who? found _% of Sz patients fulfill OCD criteria and _% show symtoms
1% have Sz, 2-3% have OCD, two very different conditions, Swets et al found 12% Sz p’s fulfilled OCD criteria and 25% show symptoms
Validity Evaluation: research support for gender bias - who? randomly selected - male+females psychiatrists to read 2 patient vignettes. he found if it said ‘male’ _% said Sz, if it said ‘female’ _% said Sz. also gender bias was not evident with _ psychs. so diagnosis is influenced by…
Loring and Powell randomly selected 290 male+females psychs to read 2 patient vignettes. he found if it said male, 56% said Sz, if it said female 20% said Sz. also gender bias wasn’t evident with female psychs. so diagnosis is influenced by gender of patient and psych
Validity Evaluation: consequences of comorbidity - who? looked at nearly _ million hospital discharge records to calculate comorbidity rates. psychiatric and behaviour related diagnosis accounted for _% of comorbidity. he found many comorbid p’s who had Sz and a _ problem. p’s received lower standard _ care which then affects _ of Sz.
Weber et al looked at nearly 6 million hospital discharge records to calculate comorbidity rates. psychiatric and behaviour related diagnosis accounted for 45% of comorbidity. he found many comorbid p’s who had Sz and a medical problem. p’s received lower standard medical care which then affects prognosis of Sz.
Validity Evaluation: differences in prognosis - no evidence to say p’s _ prognosis. it varies e.g. _% recover previous level of functioning, _% achieve lasting improvement and _% show improvement with small relapses. therefore it has very little _ validity.
no evidence to say p’s share prognosis. it varies e.g. 20% recover previous level of functioning, 10% achieve lasting improvement and 30% show improvement with small relapses. therefore it has very little predictive validity.
Reliability Evaluation: what did Whaley find inter-rater reliability for Sz to be
0.11 - very low
Reliability Evaluation: why may symptoms be unreliable
only 1 symptom has to be required if ‘delusions are bizarre’, but how do we know what is and isn’t bizarre?
Reliability Evaluation: unreliable symptoms - when 50 psychs were asked to differentiate between non-bizarre and bizarre symptoms the inter-rater reliability was…
0.40 - low, therefore it is a very subjective way of diagnosing Sz, reducing reliability
Reliability Evaluation: cultural differences - what did Barne find
cultural and racial differences in diagnosis
Reliability Evaluation: cultural differences - how many Sz p’s did Brekke and Barrio study and what were the two groups
184, 2 non-white minority groups (African-Americans and Latinos) 1 white American majority group
Reliability Evaluation: cultural differences - what did Brekke and Barrio find
majority members were more symptomatic then minority members who received better prognosis
Reliability Evaluation: cultural differences - what is the ethnic culture hypothesis and how does it link to Brekke and Barrio’s findings
minority experience less distress in mental disorders due to protective characteristics and social structures in their cultures, supported by their findings
