[The modified psychiatric interview: a study of reliability].

In order to estimate the prevalence of mental disorders in epidemiologic studies in the community, our department has designed the Modified Psychiatric Interview (MPI), a semi-structured interview derived from the Standardized Psychiatric Interview by Goldberg et al. This instrument confirms the presence of "psychiatric cases" and is applied by interviewers with clinical experience. The purpose of this article is to evaluate the degree of agreement between clinical interviewers (using an interviewer-observer design), utilizing MPI in an out-patient population attending a general hospital, as part of an investigation on prevalence of mental disorders at a 2nd. level of medical care. The degree of agreement was calculated in the different sections of the instrument by Pearson's product-moment correlation, Kappa statistic and Yule's "Y" coefficient. The results showed a high level of reliability among the clinical interviewers in each section of the MPI.