Validity of the bi-axial dependence concept: a test in the US general population.

According to the "bi-axial" concept of alcohol dependence, the Alcohol Dependence Syndrome (ADS) constitutes an axis or dimension of alcohol difficulties, while other alcohol-related problems (social, legal, etc.) constitute one or more separate dimensions. The validity of the bi-axial distinction was investigated in a stratified probability sample of 3212 US current drinkers who were interviewed in their households. Indicators of the Alcohol Dependence Syndrome and potentially distinct alcohol-related problems were covered in a structured interview administered by carefully trained interviewers. This interview provided extensive coverage of drinking patterns and problems. Aspects of the ADS covered included narrowing, salience, tolerance, withdrawal, withdrawal relief/avoidance and compulsion/control. Other alcohol problems included difficulties with work, health, the law, general social difficulties and problems in marriage/home life. Confirmatory and exploratory factor analyses were used to determine whether a single factor (dimension) or two or more factors fit the data best. Using all methods, we found that one general factor explained the structure of the data better than a two-factor model or other models for males, females, blacks and whites. Thus, the utility of this approach to distinguishing between types of alcohol problems was challenged, raising some questions about abuse/dependence distinctions in various nomenclatures.