The semi-structured affective temperament interview (TEMPS-I) - Reliability and psychometric properties in 1010 14-26-year-old students

Background: The purpose of this study was to evaluate the reliability and psychometric properties of the Semistructured Affective Temperament Interview, and determine cut-offs for each temperament. Method: 1010 Italian students aged between 14 and 26 were evaluated by means of the Akiskal and Mallya criteria in a Semistructured Interview for depressive, cyclothymic, hyperthymic, and irritable temperaments. Results: This instrument has very good reliability and internal consistency. The percentage of subjects with a z-score higher than the second positive standard deviation (+2 SD) on the scales of depressive and cyclothymic temperaments are 3.6% and 6.3% (reaching scores of 7/7 and 9/10), respectively. Hyperthymic traits, on the other hand, are widespread in our sample: most subjects are included within the second positive standard deviation (+2 SD), and 8.2% of these reach a 7/7 score; therefore, the problem of defining a cut-off for this temperament is still open. By contrast, the irritable temperament is rare, conforming to a non-gaussian distribution, with 2.2% of cases above the second positive standard deviation (+2 SD). Limitation: The data are based on subject report without collateral information and external validation. Conclusion: This study contributes to more accurate definition of cut-offs for individual temperament scales. The standardization of the interview thus makes it possible to compare three out of four temperamental scales, showing the dominant temperamental characteristics for each subject. Prospective studies are needed to demonstrate the stability of these traits over time. (C) 1998 Elsevier Science B.V.