No abstract

Background Adolescent and adult samples have shown that DSM-IV abuse and dependence criteria lie on a continuum of alcohol problem severity, but information on criteria functioning in college students is lacking. Prior factor analyses in a college sample (Beseler et al., 2010) indicated that a two-factor solution fit the data better than a single-factor solution after a binge drinking criterion was included. The second dimension may indicate a clustering of criteria related to excessive alcohol use in this college sample. Methods The present study was an analysis of data from an anonymous, online survey of undergraduates (N = 361) that included items pertaining to the DSM-IV alcohol use disorder (AUD) diagnostic criteria and binge drinking. Latent class analysis (LCA) was used to determine whether the criteria best fit a categorical model, with and without a binge drinking criterion. Results In a LCA including the AUD criteria only, a 3-class solution was the best fit. Binge drinking worsened the fit of the models. The largest class (class 1, n = 217) primarily endorsed tolerance (18.4%); none were alcohol dependent. The middle class (class 2, n = 114) endorsed primarily tolerance (81.6%) and drinking more than intended (74.6%); 34.2% met criteria for dependence. The smallest class (class 3, n = 30) endorsed all criteria with high probabilities (30% to 100%); all met criteria for dependence. Alcohol consumption patterns did not differ significantly between classes 2 and 3. Class 3 was characterized by higher levels on several variables thought to predict risk of alcohol-related problems (e.g., enhancement motives for drinking, impulsivity and aggression). Conclusions Two classes of heavy drinking college students were identified, one of which appeared to be at higher risk than the other. The highest risk group may be less likely to “mature out” of high-risk drinking after college.