This study encompasses five lines of work focused on the National Comorbidity Survey (Adolescent), a national survey of the mental health of adolescents. The NCS-A has interviewed over 10,000 12- to 17-year-olds and administered questionnaires to parents of more than 6,600 of them, using an adolescent version of the WHO CIDI. A small K-SADS clinical reappraisal study was also conducted. The study’s aims are to:

• multiply impute K-SADS diagnoses for all 10,000 adolescents and perform parallel analyses of CIDI-A and predicted K-SADS diagnoses. The multiple imputations and the MI analysis programs used to analyze them will be made publicly available as part of the NCS-A public use dataset;
• analyze criterion- and diagnosis-level concordance of CIDI and KSADS responses, to identify CIDI-A items that might be imprecise or lack clarity and therefore might be suitable targets for rewording or deletion;
• improve short self-report and parent-report screening scales for any disorder and for severe emotional distress based on an analysis of an extensive battery of screening questions in the NCS-A;
• develop methodology for estimation of school-level prevalence of SED for mental health treatment planning, using data from short screening scales. Bayesian small-area estimation methods will be used here to borrow strength from school covariates and to combine individual- and school-level data to improve estimates for individuals.