Two-decades of genetic studies have left the relationship between DRD2/ANKK1 and alcoholism indeterminate. Many reasons have been put forth to explain the mixed association results. Among them, poor DNA extraction techniques, population stratification, and failure to properly screen controls for drug and alcohol disorders. Previous reviews of this literature have detailed the variability and limitations of these studies (Goldman, 1998). A 2000 review by Noble (Noble, 2000) focused on sample size, types of alcoholics analyzed, and the nature of comparative controls employed in a variety of previously published studies. He reviewed several samples each of which used varying measures of alcoholism (The Michigan Alcoholism Screening Test, the presence or absence of medical complications of alcoholism, alcohol consumption, Severity of Alcohol Dependence Questionnaire (SADQ), and the DSM-III-R criteria). In this paper, we focus on the variability in the measure of the phenotype used across this literature in an effort to understand how this variability may effect the conclusions one would draw about the evidence for association with DRD2/ANKK1.
