@article{ref1, title="Alcohol use inventory: screening and assessment of first-time driving-while-impaired offenders: ii. typology and predictive validity", journal="Alcohol and alcoholism", year="2001", author="Chang, I. and Lapham, Sandra C. and C'de Baca, Janet and Davis, J. W.", volume="36", number="2", pages="122-130", abstract="This study evaluated the use of Alcohol Use Inventory (AUI) for driving-while-impaired (DWI) screening, by determining whether DWI offenders (n = 1644), grouped according to their reported alcohol involvement on the AUI, would have different rates of recidivism in a 5-year follow-up. Cluster analysis using the six second-order scales produced six groups (clusters 1-6) described as the Low-Profile (50%), Alcohol-Preoccupation (14%), Enhanced (22%), Enhanced-Disrupt (9%), Anxious-Disrupt (3%), and High-Profile (1%) types. They were characterized by different sociodemographic profiles. Members of cluster 4 were associated with the highest DWI recidivism rate (40%), committing one or more further DWI, and clusters 5 and 6 were associated with the highest rate of committing two or more DWIs. Rates of subsequent traffic convictions and crashes were, however, not statistically different among the clusters. Predictors of DWI recidivism included male gender, young age, less-educated, high blood-alcohol concentration at arrest, and clusters of 3 and 4. Different typologies indicated that the needs for treatment might be different. Evaluators should keep in mind the strength of AUI, use risk factors identified in the study, and take measures of test-taking defensiveness to enhance overall predictive validity.", language="", issn="0735-0414", doi="", url="http://dx.doi.org/" }