A COMPARISON OF 2 STRUCTURED DIAGNOSTIC INTERVIEWS - CIDI AND SCAN

The relationship between and the inter-rater reliability of the Compos ite International Diagnostic Interview (CIDI) and the Schedules for Cl inical Assessment in Neuropsychiatry (SCAN) for anxiety and depressive disorders were explored. The CIDI and the SCAN were administered by t rained interviewers in counterbalanced order. A subsample of interview s was observed to determine the inter-rater reliability of the instrum ents. Subjects were 101 patients accepted for treatment at an Anxiety Disorders Clinic; 29 of the 101 patients participated in the inter-rat er reliability study. Concordance between the instruments as measured by canonical correlation analysis was moderate for current (r=0.69, p= 0.05) and for lifetime (r=0.66, p=0.05) diagnoses. Inter-rater reliabi lity of the CIDI was perfect (overall intraclass kappa = 1.00), and of the SCAN was good (overall intraclass kappa = 0.67). It is concluded that although the two instruments made similar diagnostic distinctions , the clinical judgement involved in administering the SCAN resulted i n the more moderate levels of agreement between the interviewer and-ob server than those found for the CIDI.