[Sensitivity and specificity between the Composite International Diagnostic Interview Version 3.0 (World Mental Health, CIDI) and the Standardised Clinical Evaluation version I (SCID-I) in a mental health survey of the city of Medellin, 2012]. / Sensibilidad y especificidad entre la Entrevista Diagnóstica Internacional Compuesta versión 3.0 (World Mental Health, CIDI) con la Evaluación Clínica Estandarizada versión I (SCID-I) en la Encuesta de Salud Mental de la ciudad de Medellín, 2012.

INTRODUCTION: In order to address the mental health problems of the Colombian population it is necessary to have diagnostic tools (local and international) that are valid, easy to apply, and comparable. OBJECTIVE: To compare the sensitivity and specificity between the CIDI 3.0 and the SCID-I for major depressive disorder, bipolar I and II disorder, and substance dependence disorder. METHODOLOGY: Cross-sectional study comparing the life prevalence of three mental disorders in 100 subjects using the CIDI 3.0 and the SCID-I. The study was approved by the Institutional Ethics Committee. The two diagnostic interviews were performed that measured by sensitivity, specificity, positive predictive value and negative predictive value with confidence intervals of 95%. The SPSS version 21.0 software was used for data analysis. RESULTS: The median age was 43.5 years, with an interquartile interval of 30 years. The highest sensitivity (Se) and specificity (Sp) was observed for drug dependence diagnosis - with 80%, (95%CI, 34.94-100), and 98.46 (95%CI, 94.7-100), respectively. CONCLUSIONS: SCID-I and CIDI 3.0 showed different levels of sensitivity and specificity for the three disorders studied with: high for substance dependence disorder, moderate for bipolar disorder I and II, and low for major depressive disorder.