Flashcards in Diagnosis and Classification of depression Deck (16):
What is inter-rater reliability?
Whether 2 independent assessors give the same diagnosis
What can low inter-rater reliability levels lead to?
Faulty diagnosis and treatment
Who assessed the inter-rater reliability of a structured interview as an assessment of major depressive disorder?
Lobbestael et al (2001) and found a moderate agreement with coefficient .66
What's test-retest reliability?
The consistence of the diagnosis over time
Why is test-retest reliability important?
To create new and improved systems
Who studied test-retest reliability?
Beck et al (1996) studied the responses of 26 outpatients tested at 2 therapy sessions one week apart using BDI and found a coefficient of .93
the extent to which 2 conditions co-occur, e.g. anxiety is the single biggest clinical risk for the development of depression
Whats content validity?
whether the items are representative of that which is been measured, e.g. the BDI scale is considered high because it was constructed based on symptoms found in psychiatric patients
Whats concurrent validity?
the extent to which a test concurs with already existing systems. Beck demonstrated this with the BDI scale and Hamilton Depression Scale.
Evaluation: reliability for DSM diagnosis
Keller et al (1995) suggests that inter-rater reliability is fair to good and test-retest is fair. they recruited 524 depressed ps from 5 settings and interviewed each using DSM criteria, then again 6 months later.
Evaluation: Reasons for low relaibility
It was suggested by Keller that because 5/9 symptoms must be present for a diagnosis, one item disagreement could cause the difference between a diagnosis and not
Evaluation: who suggested the DSM is too lengthy?
Zimmerman et al (2010) stated doctors have difficulty identifying all 9 syptoms
Evaluation: support that the DSM is too complex?
Tiller et al (2001) studied 25000 GPs from Australia and New Zealand and found only 1/4 could list 5 of the symptoms.
Evaluation: what was Zimmermans simpler system?
Using only mood and cognitive symptoms, this had 95% agreement in diagnosis'.
Evaluation: validity for types of depression
McCullough et al (2003) compared 681 patients with various types of depression. they found few differences between their symptoms therefore suggesting distinctions may not be valid