Because of the expected complexity of factors contributing to alcoholism risk, COGA required a large sample size to allow detection of the genetic “signal” through the “noise.” Of particular concern was the likely variability within the sample of both the number and type of genetic and environmental factors contributing to alcoholism risk; therefore, the contribution of any one factor would only account for a small fraction of the variation in risk. The investigators chose a family study design to allow the use of multiple methods of genetic analysis. Systematic recruitment from outpatient and inpatient alcoholism treatment facilities and assessment of families initially was carried out at six sites across the United States, with a seventh site more recently. The study also included a large sample of control families that were randomly selected from the community. For the analyses, the researchers chose a split-sample design—two groups of subjects (i.e., an initial sample and a replication sample) were analyzed independently; this approach allows investigators to examine the reproducibility of the initial study findings.
