The present study increases the existing knowledge in four ways.

First, to our knowledge, this study is the first to examine GW786034 purchase the association between ADHD and both prevalence and age of onset of three different stages of alcohol use. Second, both the mediating and modifying role of CD in the association between ADHD and alcohol use (disorder) is examined. Third, using data of a general population study enables us to examine associations which are applicable to the population at large. Moreover, the use of an adult sample enables us to associate childhood ADHD with AUD at a much later age than most other studies in which the association between ADHD, CD, and alcohol use (disorder) was examined (Disney et al., 1999, Fergusson

et al., 2007, Flory et al., 2003 and Molina et al., 2002). This provides us the opportunity to study processes that emerge in adulthood. Fourth, not symptom counts but DSM-IV diagnoses of ADHD, CD, and AUD will be used. Data were derived from the baseline assessment of NEMESIS-2. Methods have been reported elsewhere (de Graaf et al., 2010). Briefly, NEMESIS-2 is based on a multistage, stratified, random sampling of households, with one respondent randomly selected in each household (response rate 65.1%). The Composite International Androgen Receptor Antagonist Diagnostic Interview (CIDI) version 3.0 was used to determine the presence of ADHD, CD, and AUD according to these DSM-IV criteria. The CIDI is a fully structured, lay administered interview developed by the World Health Organization. The CIDI is used worldwide, and has been shown to be a reliable and valid instrument (Haro et al., 2006). To increase accuracy of retrospective recall, ADHD and CD were only assessed among respondents aged 18–44 (conform Kessler et al., 2007). This resulted in a total sample of 3309 respondents. Respondents who answered positively to one of the screener questions for ADHD or for CD entered the relevant CIDI sections. In these sections symptoms of the disorder, impairment due to these symptoms, and age of onset were assessed. Computerized CIDI algorithms were used to generate diagnoses according

to full DSM-IV criteria. All participants entered the alcohol section which started with a question to measure alcohol initiation: ‘How old were you the very first time you ever drank an alcoholic beverage?’ Only participants who reported ever-use continued with the alcohol section, the next question assessed regular drinking: ‘How old were you when you first started drinking at least 12 drinks per year?’ Only participants who reported regular drinking continued with the next part of the alcohol module assessing symptoms of alcohol abuse and dependence, impairment due to these symptoms, and age of onset. Analyses were performed using Stata version 11.1 which enabled us to control for the complex sampling and recruitment procedure of the study.