d.2 Data Analysis d.2.1 General Overview One advantage of the COGA mechanism is that It brings together a large group of investigators with diverse interests and expertise from ten different sites. This expert investigative team, coupled with the extensive phenotypic data generated on one of the largest high-risk samples in existence, makes it possible to address a wide range of important questions regarding mechanisms of how genetic and environmental factors translate into the eventual risk for alcohol use disorders. Accordingly, in the Data Analysis section, we present an overview of the different techniques we will utilize to study pathways of risk toward alcohol use disorders, including alcohol dependence. Consistent with Specific Aim #2, analysis of the development of alcohol use behaviors including dependence, other substance dependence, and psychopathology in adolescents and young adults will involve comprehensive, multivariate models of risk assessment that Incorporate clinical Information, genotypic and environmental factors. Taking advantage ofthe full COGA phenotypic information, outcomes will include categorical (e.g., presence/absence of alcohol dependence), quantitative (e.g., age onset of first drink, regular drinking, alcohol problems;number of alcohol symptoms), and survival-time (e.g., onset of alcohol dependence) measures. A major focus of analysis will be the impact of genetic variables (i.e., genotypes or haplotypes) on selected outcomes. Covariates can also include individual-level (e.g., comorbidity, personality traits) and family-level (e.g., parental psychopathology) measures. The analysis will address the methodological considerations that arise from longitudinal studies Including missing data, timeinvariant factors (e.g., sex, genotype), time-dependent factors (e.g., age, comorbid psychopathology), attrition, and clustered data, as discussed in more detail below. Analysis of the longitudinal data collected from the repeated assessments of the adolescents and young adults will permit investigation of numerous important aspects in the development of alcohol dependence including the point prevalence of psychopathology at different stages of development;the ages at onset of both symptoms and disorders;the developmental sequences and sequential patterns of symptomatology;the stability and continuity of alcohol and substance use and abuse/dependence across developmental stages;the predictive ability of events in childhood and adolescence on outcomes in young adulthood;and aggregate trends over time within and between individuals. Because COGA is a longitudinal study, it incorporates the strengths of cross-sectional, longitudinal, and accelerated longitudinal studies that allows for analyses to disentangle age, period, and cohort effects on the development of alcohol dependence and related phenotypes (Farrington, 1991;Willetetal., 1998).
Showing the most recent 10 out of 466 publications
