Validity and reliability of the Greek version of the semistructured Schedule Clinical Interview for personality disorders (SCID-II).

The semistructured Schedule of Clinical Interview for Personality Disorders (SCID-II) is a useful tool for measuring personality disorders according to DSM criteria. Personality traits and their assessment are culturally sensitive. Because of this, it is important for clinicians and researchers to have a clearer view of the performance of such instruments in their own culture. Despite the fact that the SCID-II interview has been translated to the Greek language, the psychometric properties of this version have yet to be tested. To address this need, we conducted this study to assess the validity and reliability of the SCID-II interview in its DSM-III-R version in the Greek population. A total of 32 patients, 13 men and 19 women, were involved in this study. Sixteen patients were interviewed three times by three different interviewers. The first two interviewers used the Greek version of the SCID-II, and the third interviewer used the International Personality Disorder Examination (IPDE), which was used as the gold standard. Of the remaining 16 patients, 14 were interviewed with the SCID-II by two interviewers, and 2 were unable to complete the interview and were excluded from the study. A total of 69 interviews were performed. The internal consistency of the interview was acceptable, with a Cronbach's alpha coefficient of 0.623. The SCID-II also demonstrated good reliability. Cohen's Kappa score ranged between 0.375 for histrionic disorder and 1.000 for defeatism and antisocial personality disorder. Pearson's correlation coefficient was also very strong for both the individual criteria score and the overall diagnosis between the interviewers. There was an exception for the not otherwise specified personality disorder, where there was no agreement in any of the reliability measures between the interviewers. The interview validity was high when measured against the gold standard. The specificity of the SCID-II ranged from 79-100%, with the expectation of not otherwise specified personality disorder being 66%. The overall sensitivity was moderate and ranged from 0-100%. The Greek version of the SCID-II is a reliable, valid and easy-to-use instrument that can be adopted by various mental health professionals for clinical as well as research purposes.