Multivariate imputation by chained equations (MICE) has emerged as a principled method of dealing with missing data. Despite properties that make MICE particularly useful for large imputation procedures and advances in software development that now make it accessible to many researchers, many psychiatric researchers have not been trained in these methods and few practical resources exist to guide researchers in the implementation of this technique. This paper provides an introduction to the MICE method with a focus on practical aspects and challenges in using this method. A brief review of software programs available to implement MICE and then analyze multiply imputed data is also provided.

General population data from the National Comorbidity Survey are presented on co-occurring DSM-III-R addictive and mental disorders. Co-occurrence is highly prevalent in the general population and usually due to the association of a primary mental disorder with a secondary addictive disorder. It is associated with a significantly increased probability of treatment, although the finding that fewer than half of cases with 12-month co-occurrence received any treatment in the year prior to interview suggests the need for greater outreach efforts.

Increased availability of research-supported, school-based prevention programs, coupled with the growing national policy emphasis on use of evidence-based practices, has contributed to a shift in research priorities from efficacy to implementation and dissemination. A critical issue in moving research to practice is ensuring high-quality implementation of both the intervention model and the support system for sustaining it. The paper describes a three-level framework for considering the implementation quality of school-based interventions. Future directions for research on implementation are discussed.

No abstract

No abstract