The value of CAGE, CUGE, and AUDIT in screening for alcohol abuse and dependence among college freshmen

Background: This study attempted to (1) determine the prevalence of alcohol problems in college freshmen, (2) assess the performance of both the CAGE and the Alcohol Use Disorders Identification Test (AUDIT) questionnaires in this population, and (3) assess the possibility of improving the CAGE and/ or AUDIT. Methods: A sample of 3564 consecutive college freshmen, with a mean age of 18 years, at the Catholic University of Leuven, (Belgium) completed, during a cross-sectional study, a questionnaire assessing drinking behavior and i dentifying students at risk as defined by DSM-IV criteria. The questionnair e also included the CAGE questionnaire and the AUDIT. Calculations of sensi tivity, specificity, negative predictive value, positive predictive value, likelihood ratios, and receiver operating characteristic curves for differe nt scores of the CAGE and the AUDIT were performed, using DSM-IV criteria a s the reference standard. Results: The area under the receiver operating characteristic curve of the CAGE and the AUDIT was 0.76 and 0.85, respectively. The cutoff score of 1 f or the CAGE was associated with a sensitivity of 42%, a specificity of 87%, a positive predictive value of 36%, and a negative predictive value of 90% . A score of 6 or more for the AUDIT gave a sensitivity of 80%, a specifici ty of 78%, a positive predictive value of 37%, and a negative predictive va lue of 77%. These results were related with a prevalence of 14.1% of alcoho l problems. Replacing one question of the CAGE by "often driving under the influence" resulted in the CUGE (acronym for "cut down, under influence, gu ilty feelings, and eye opener"), with an area under the curve of 0.96, a po sitive likelihood ratio of 8.7, and a negative likelihood ratio of 0.04. Conclusions: Prevalence of alcohol problems in college students is confirme d to be high. When screening for alcohol problems in a college freshmen pop ulation, one question seems extremely important. The newly constructed CUGE questionnaire may improve screening efforts in students, compared with exi sting questionnaires.