Reliability and Validity in Diagnosis Flashcards
Inter-rater reliability
- 2 raters assess same pt with same tool
- correlate scores (kappa score)
- kappa score >0.8 is good
Inter-rater reliability: Cheniaux
2 psychiatrist, 100 fake case studies
- psychiatrist 1 2x more likely to diagnose
- ICD-10 2x more likely to result in diagnosis
Why low inter-rater reliability?
bizarre is subjective
Culture Bias: willingness of psychiatrist
COPELAND found US psychiatrists were more likely to diagnose than UK
Culture bias: diagnostic tool
developed in western culture - symptoms assumed to be maladaptive - imposed etic
Gender Bias
Men more likely to be diagnosed than women
Gender Bias: Cotten et al
Men over diagnosed, women under
- women more likely to remian functioning with Sz therefore less diagnosis
Gender Bias: Broverman et al
US clinicians equated mentally ‘adult’ healthy with male behaviour so women were viewed as less mentally healthy
Gender Bias: Loring + Powell
Gave clinicians fictional cases
- 56% male diagnosed
- 20% female diagnosed
- men higher diagnosis with Sz
Symptom overlap
- flat affect in unipolar depression
- delusion in bipolar depression
- pt Sz in ICD but bipolar in DSM
- questions validity of classification systems
Comorbidity
2 or more disorders in one person
Comorbidity: Buckley et al
People w/ Sz have..
- depression = 50%
- substance abuse = 47%
- PTSD = 29%
What caused what?
