reliability and validity in diagnosis Flashcards
what is the US classification book
DSM-5
what is the UK classification book
ICD-10
what are the 2 issues with reliability
- gender bias
- co-morbidity
what are the 2 issues with validity
- culture bias
- symptom overlap
definition of reliability and in terms of diagnosis
the extent to which a finding is consistent - when psychiatrists can agree on the same diagnosis when independantly assessing patients (inter-rater reliability) - reliable = reach same decision
Beck - research evidence for inter-rater reliability
Beck 1962 found the inter-rater reliability between 2 experienced psychiatrists for a sample of 154 patients was only 54%
Spitzer and Fleiss research evidence for inter-rater reliability
Spitzer and Fleiss 1974 conducted a meta-analysis of 6 studies which examined 18 disorders and found that although agreement in individual studies ranged from 10-90% the average was 52%
what did Soderberg find with the more updated DSM
(2005) a concordance rate of 81% = more reliable over time
what did Jakobsen find when using ICD-10
(2005) found a concordance rate of 98% in a study where 100 Danish patients were assessed using ICD-10 diagnostic criteria = high reliability
Cheniaux research and what does this mean
Cheniaux (2009) - Main problem is that there isn’t one single classification system used by all, (DSM-5 in USA and ICD-10 in Europe) - the same behaviour could be disgnosed differently
research = 2 psychiatrists independantly diagnosed 100 patients using DSM and ICD criteria
* psychiatrist A = 26 for DSM and 44 for ICD
* psychiatrist B = 13 for DSM and 24 for ICD
= may not recieve appropriate treatment
strength of research into the reliability of diagnosis
valuable contribution - the later version (DSM-III) improved the criteria and became less vague - included stardardised clinical interviews - suggests reliability has improved with revisions of the diagnostic criteria
However although they have became more similar they do still have different ideas about the diagnostic criteria of SZ = someone with SZ may only be diagnosed in UK not US
definition of validity relating to diagnosis
the extent to which we are measuring what we are intending to measure - with SZ we have to consider the validity of diagnostic tools e.g. do different assessment systems arrive at the same diagnosis for the same patient
research evidence into validity of diagnosis by the depression and bipolar support alliance
a survey conducted by the depression and bipolar support alliance (2002) found that 70% of bipolar respondants reported that their illness had been misdiagnosed at least once usually with SZ - suggesting diagnosis isn’t accurate
stats with variation of SZ symptoms
70% experience hallucinations and delusions but 30% don’t
what is the key issue with SZ (symptoms)
it doesn’t have any pathognomonic symptoms (no characteristics unique to SZ)
Rosenhan study aim
to investigate how situational factors affect a diagnosis of SZ
Rosenhan study method
(1973) 8 confederates acted as pseudopatients going to 12 diffeent hospitals - the participants were the hospital staff who didn’t know about the experiment - the pseudopatients rang the hospitals asking for an appointment and complained about hearing voices saying empty, hollow etc, they gave false names, occupations etc
Rosenhan study results
staff diagnosed 11 patients with SZ and one with manic depression - staff never detected their sanity - the average hospital stay was 19 days - while in hospital 35 real patients detected sanity e.g. saying you’re not crazy
Rosenhan study conclusion
psychiatric staff can’t always distinguish sanity from insanity - any diagnostic method making such errors cant be reliable or valid = situational factors do affect
Mason validity research
Mason et al (1997) - aimed to assess the accuracy of 4 different classification systems in predicting outcome of SZ - 4 different diagnostic criteria were used with a group of 99 patients with SZ and their progress was monitored after 13 years - it was found the more recent versions had high predictive validity = it’s improved over time
gender bias in terms of the validity of diagnosis
some of the critics of the DSM diagnostic criteria argue that some diagnostic categories are biased towards pathologising one gender rather than the other - Broverman et al found that clinicians in the US equated healthy adult behaviour with healthy ‘male’ behaviour = androcentrism = a tendency for women to be percieved as less menatlly healthy - also some research has indicated that a psychiatrists gender may affect their ability to diagnose
what has been found with males and females experiences with SZ
- males experience more negative symptoms and have higher rates of substance abuse
- women have better quality interpersonal functioning, better recovery rates and lower relapse rates
Loring and Powell research evidence for gender bias
Loring and Powell (1988) randomly selected 290 male and female psychiatrists and asked them to read 2 case articles and to make assessments using standard diagnostic criteria - found that 56% of male patients were diagnosed but only 20% of females were
However this gender bias was only evidence amongst male psychiatrists
Harrison et als research into culture bias of diagnosis
(1984) those of afro-caribbean origin were over-diagnosed with SZ by white doctors in Bristol due to the differences of their ethnic background
Copeland et al research into culture bias of diagnosis
(1971) gave a description of a patient to 134 US and 194 British psychiatrists - 69% of US diagnosed SZ but only 2% of British = symptyoms of ethnic minorities are misinterpreted - someone may have same symptoms but be diagnosed differently due to ethnic background
ethnocentrism definition
the cultural values of one country are assumed to have universal app and therefore are used to make judgements about behaviour about people from a different culture = can lead to misdiagnosis
