Many of the aforementioned studies used standard measures of alcohol use and/or problems including the Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria, the Alcohol Dependence Scale (ADS), the Alcohol Expectancy Scale (AES), and the Alcohol Use Disorders Identification Test (AUDIT). Measures of alcohol problems vary by scientific field, setting (clinical vs. research), historical trend (DSM-III vs. DSM-IV), and availability. However, there is evidence to suggest that genetic association results may vary as a function of the alcohol measure used in the analysis. In 2002, Connor and colleagues tested the association between DRD2 and a variety of alcohol phenotypes, finding association with certain alcohol phenotypes (alcohol quantity, alcohol consumed per week, alcohol dependence scale score) and not others (frequency of alcohol use). This is an example of how even when using an identical sample and method in genetic association analyses the measure of the phenotype can affect the results.
