RESUMO
To assess the validity of diagnoses obtained with the Iranian version of the Structured Clinical Interview for DSM-IV [SCID-I]. This study was undertaken in two stages: [a] translation of SCID-I into Persian [Iranian language], [b] assessing the validity of the Persian version in a sample of Iranian patients. We recruited 299 psychiatric patients- including inpatients and ambulatory cases- from 3 teaching hospitals. A trained SCID interviewer administered the SCID and then two psychiatrists developed a consensus diagnosis, using data from multiple sources. The degree of agreement between SCID interviews and psychiatrists' diagnosis ranged from "moderate" for obsessive-compulsive and major depressive disorders to "good" for bipolar disorder and schizophrenia. With the psychiatrists' diagnosis used as the gold standard, the SCID-based diagnosis showed high specificity and moderate to high sensitivity for most psychiatric diseases. The results of this study indicate that the Iranian version of the SCID is a valid instrument for diagnosis in clinical settings