reliability and validity of diagnosis Flashcards
what is reliability of diagnosis?
how consistent a measure is
two clinicians should come to the same conclusion of the diagnosis of a disorder
what is validity of diagnosis?
how true a measure is in what it claims to measure
why is it difficult to assess the reliability of diagnosis?
- unstrcutured interviews that are used to diagnose patients are unreliable as some clinicians will choose to focus on specific areas rather than others
- cultural factprs may affect the diagnosis
- not all data is gathered due to willingness and memory etc
- some disorders ar harder to diagnose than others due to reasons such as comorbidity
how are dsm and icd becoming reliable?
they are consistently being updated and changed
dsm 5 has a similar coding system now to the icd
dsm 5 has more of a consideration towards the cultural factors that may affect diagnosis
how should dsm and icd be used in order to increase reliability?
should not be used as a checklist
, interviews should take place in order to fully asses someone and then incorporate them
what studies can be used to demonstrate the reliability of diagnosis?
nicholls
goldstein
cooper
beck et al
what did cooper et al find?
that reliability of diagnosis was low
new york and london psychiatrist
london twice as likely to diagnose mania
new york twice as likely to diagnose shiz
what did beck et al find?
diagnosis agreement of 153 patients was only v54% due to inconsistencies in how data was gathered and vague criteria
why did beck say that reliability was low?
due to inconsistencies in techniques of gathering data and vague criteria
what did nicholls find?
that when diagnosing 81 patients aged 7-16 who has eating disorders only 50% could be diagnosed using criteria
reliabilty was found to be 64% but only because the raters agreed that they could not diagnose patients
why did nicholls find high reliability?
as they could agree that they couldst diagnose patients (64%)
hat did goldstein find?
he found that the dsm III was reliable
case studies of previous patients who had a history of schizophrenia had a high level of agreement with previous diagnosis
how have diagnosis systems improved on reliability?
due to multi axial approach, more standardised techniques, more specific diagnosis criteria
what did rosenham find about validity?
that it was low
all sane people who were diagnosed with schiz
one diagnosed with bipolar even though all showed the same symptoms
what is descriptive validity?
that symptoms of one disorder should be different from another category,k distinguish
what is an example of not having descriptive validity?
comorbidity of disorder si high is specific ones such as anxiety
ocd and depression
what is predictive validity?
id the system identifies a condition that will respond in a particular way to treatment
should precept which treatment will work the best
what is the study for predictive validity?
heather - found that there is only a 50% chance of predicting which treatment will work the best
what is aetological validity?
requires all patients who have the same disorder to have developed it the same way
what is aerological validity hard to gain?
for many disorders there is not a set explanation of how it is caused e.g. biologically or socially
generally thought to be unaitable
what is concurrent validity?
extent to which two systems agree on a disorder e.g. icd and dsm
what is the study for concurrent validity?
andrews
found that when using the dsm iv and icd 10 on 1500 people found 68% agreement on the diagnosis of substance abuse, depression and anxiety
what are the types of validity?
concurrent
ateological
predictive
descriptive
what is the study for descriptive validity?
eysenck
said that 2/3 of people diagnosed with anxiety and also been diagnosed with one or more additional anxiety disorders
