RESUMO
The Composite International Diagnostic Interview [CIDI] is a comprehensive, standardized diagnostic interview for the assessment of psychiatric disorders. There have been few studies on the validity of the CIDI. The objective of present study was to assess the validity of a Farsi translation of the complete CIDI and its psychosis/mania module in five referral clinical psychiatric settings. Two hundred and three as well as 104 consecutive admissions were interviewed using the complete and the psychosis/mania module, respectively. Within two days of the CIDI interview, two last year residents of psychiatry or psychiatrist who were blind to the CIDI diagnosis completed the Clinical diagnostic checklists [based on DSM-IV and ICD-10 criteria] simultaneously and reached the consensus diagnosis. Data analysis was performed using SPSS 11 to determine the validity of CIDI. The sensitivity and specificity for the diagnosis of schizophrenia was 0.12 and 0.96 using DSM-IV criteria. According to ICD-10 criteria, the results were the same with 0.19% sensitivity and 0.96% specificity. The sensitivity for the diagnosis of bipolar I disorder was low [0.21 using DSM-IV criteria and 0.17% using ICD-10] and specificity, high [0.90 compared to DSM-IV and 0.89 compared to ICD-10 criteria]. The results were rather similar for the psychosis/mania module of CIDI. This study suggests that the Farsi translation of both the complete CIDI and the psychosis/mania module of CIDI have good specificity, but poor sensitivity for the diagnosis of schizophrenia and of bipolar I disorder