Reproducibility and differential item functioning of the alcohol dependence syndrome construct across four alcohol treatment studies: An integrative data analysis

Background: The validity of the alcohol dependence syndrome has been supported. The question of whether different measures of the construct are comparable across studies and patient subgroups has not been examined. This study examined the alcohol dependence construct across four diverse large-scale treatment samples using integrative data analysis (IDA). Method: We utilized existing data (n = 4393) from the COMBINE Study, Project MATCH, the Relapse Replication and Extension Project (RREP), and the United Kingdom Alcohol Treatment Trial (UKATT). We focused on four measures of alcohol dependence: the Alcohol Dependence Scale (COMBINE and RREP), Alcohol Use Inventory (MATCH), the Leeds Dependence Questionnaire (UKATT), and the Diagnostic and Statistical Manual of Mental Disorders (COMBINE and MATCH). Moderated nonlinear factor analysis was used to create a measure of alcohol dependence severity that was moderated by study membership, gender, age, and marital status. Results: A commensurate measure of alcohol dependence severity was successfully created using 20 items available in four studies. We identified differential item functioning by study membership, age, gender, and/or marital status for 12 of the 20 items, indicating specific patient subgroups who responded differently to items based on their underlying dependence severity. Conclusions: Alcohol dependence severity is a single unidimensional construct that is comparable across studies. The use of IDA provided a strong test of the validity of the alcohol dependence syndrome and clues as to how some items used to measure dependence severity may be more or less central to the construct for some patients.