Screening adolescents for problem drinking: Performance of brief screens against DSM-IV alcohol diagnoses

Objective: The performance of three brief screens, the CAGE, TWEAK and Alco hol Use Disorders Identification Test (AUDIT), was evaluated against a DSM- IV diagnosis of alcohol abuse or dependence in an adolescent sample. Method : Adolescents (13-19 years old) who presented to an emergency department fo r treatment of an injury, and who tested negative for blood alcohol concent ration at time of admission, were administered a structured diagnostic inte rview and modified versions of the CAGE. TWEAK and AUDIT. Results: Of the 4 15 adolescents for whom complete data were available, 18% met criteria for a DSM-IV alcohol use disorder according to the Diagnostic Interview Schedul e for Children (version 2.3). Teens who reported alcohol use in the last ye ar (n = 261, 58% male. 71% white) were included in analyses that compared t he performance of the three screening instruments. Receiver Operating Chara cteristic analysis indicated that the AUDIT demonstrated the best performan ce across the range of its cut-scores, with optimal performance at a cut-sc ore of 4. The TWEAK performed optimally at a cut-score of 2 and the CAGE at a cut-score of 1. Conclusions: Routine alcohol screening among adolescents seen in a hospital setting is indicated. Two important directions for futu re research include the identification of adolescent-specific alcohol scree ning items, and the validation of an adolescent-specific definition of prob lem drinking that addresses limitations of DSM-IV alcohol diagnoses when ap plied to adolescents.