The British Journal of Psychiatry
Clinical observation of assessment using the Composite International Diagnostic Interview (CIDI). An analysis of the CIDI Field Trials--Wave II at the St Louis site.
A Janca, L N Robins, L B Cottler, T S Early


Two clinicians scored the ICD-10 Research Criteria Checklist either while observing or after administering CIDI interviews to a sample of 20 subjects. Overall diagnostic concordance between clinical and CIDI assessments was found to be good (overall kappa = 0.77). Assessment of the specific diagnoses could be done only for the three most commonly represented in the studied sample: anxiety/phobic disorders (kappa = 0.73), depressive disorders (kappa = 0.78), and psychoactive substance use disorders (kappa = 0.83). While the lack of independence of the two assessments and the small, non-randomly selected sample might have exaggerated the concordance, this study shows that the CIDI provides all the data needed to score diagnoses in the ICD-10 nomenclature, as indicated by the small number of questions clinicians needed to ask following completion of the CIDI.