RESUMEN
OBJECTIVES: This study aims to examine the psychometric properties of the Persian version of the Structured Clinical Interview for DSM-5 for personality disorders (SCID-5-PD) among patients referred to psychiatric centres in Iran. METHODS: Between March 2017 and June 2019, 287 outpatients and inpatients aged 16 to 75 years who were referred to three psychiatric centres in Tehran, Iran were invited to participate. Patients were interviewed using the Persian version of the SCID-5-PD by two PhD students in clinical psychology who were blinded to patient records. Face validity and content validity of the Persian version of the SCID-5-PD were assessed by five specialists with ≥2 years of clinical experience. The agreement between the diagnoses made with the Persian version of the SCID-5-PD by the two PhD students in clinical psychology and the gold standard diagnoses made with DSM-5 by psychiatrists was determined, as were sensitivity, specificity, and positive and negative likelihood ratios. 109 (43.6%) patients were interviewed again after an interval of 7 to 10 days for inter-rater reliability and test-retest reliability. RESULTS: A total of 250 patients aged 17 to 74 (mean, 32.56) years were included. Face validity and content validity of the Persian version of SCID-5-PD were acceptable. The agreement between the Persian version of SCID-5-PD and DSM-5 (gold standard) was acceptable (kappa >0.4) for the diagnoses of obsessive-compulsive, paranoid, schizotypal, schizoid, histrionic, narcissistic, borderline, and antisocial personality disorders, whereas the agreement was unacceptable (kappa <0.4) for the diagnoses of avoidant and dependent personality disorders. Sensitivity for all diagnoses was high, except for avoidant (0.66) and dependent (0.66) personality disorders. Specificity for all diagnoses was high, except for avoidant personality disorder (0.66). The positive and negative likelihood ratios showed that the SCID-5-PD was accurate for diagnosing all personality disorders, except for schizoid personality disorder. Inter-rater reliability was good for all personality disorders, except for schizotypal personality disorder (0.531). Test-retest reliability was good for all personality disorders. CONCLUSION: The Persian version of the SCID-5-PD can be used to evaluate those who seek psychotherapy for all personality disorders, except for avoidant, dependent, schizoid, and schizotypal personality disorders.
