한국판 진단적 면담검사의 개발과 그 타당도 검사 ᅳ강화도 정신과 역학 연구 ( I ) ᅳ

Development of Korean Version of the NIMH -Diagnostic Interview Schedule and Its Validity Test - Kangwha Psychiatric Epidemiologic Survey (I) —

Because the NIMH- Diagnostic Interview Sched-ule(DIS), a fully structured and computer - analyzable diagnostic instrument, was developed for the use of large - scale epidemiologic studies such as the EC A program in the United States, it can be most suitably applied to our Kangwha Psychiatric Epidemiologic Survey, that is the first step of our ongoing Kangwha Mental Health Project for the development of psychiatric service delivery system & community mental health modality in Korea. The authors presented translation procedure of the DIS -IIIA version and subsequent editing of Korean version, and discussed several problems occurring in the procedures as transcultural point of view. The concurrent validity of Korean version has been also studied by calculating agreement (kappa value) between DIS diagnoses by lay-interviewers and clinical diagnoses by psychiatrists ac cording to the line of DSM- III with exclusion criteria. The subjects included in this validity study were 155 selected from 5 different psychiatric clinic of both inpatient and outpatient services. The DIS interviews were accomplished by 17 lay-interviewers who had been trained for two weeks according to NIMH-St. Louis group workshop schedule. Independent clinical diagnoses for each subject were obtained by psychiatrists and their intake diagnostic conference procedures. The results were as follows ; 1) Various sections of DIS &#8212; IIIA , especially somatization and major depression section were inevitably modified as the result of transcultural considerations. 2) Kappa values of agreement are rather low in neurotic disorders such as panic (0.2), generalized anxiety disorder (0.22), and obsessive compulsive disordei< 0.49) without statistical significance, but rather high in the remaining disorders showing between 0.63 and 0.81. 3) The Kappa value of somatization disorder is 0.66, higher than expected before this study, and it can be possibly increased up to 0.92 by adding sup plementary 21 items of culture-bound symptom complaints.