The accuracy of the CAGE, the Brief Michigan Alcoholism Screening Test, and the Alcohol Use Disorders Identification Test in screening trauma center patients for alcoholism.

Abstract

To evaluate the accuracy of questionnaire screening instruments to identify lifetime alcohol dependence among trauma center patients. The study was conducted at a Level I trauma center between September 1994 and November 1996. Patients meeting eligibility requirements (> or = 18 years old, admission from injury scene, > or = 2 days of hospitalization, intact cognition) were evaluated for alcohol abuse and dependence. Screening instruments consisted of the CAGE, the Brief Michigan Alcoholism Screening Test, and the Alcohol Use Disorders Identification Test. Screening results were compared with lifetime alcohol dependence diagnoses made using the in-depth Psychoactive Substance Use Disorders section of the Structured Clinical Interview. Accuracy was quantified as sensitivity, specificity, positive/negative predictive values, and receiver operating characteristic curves (used to calculate area under the curve). Of the 1,118 patients studied, lifetime alcohol dependence was diagnosed by Structured Clinical Interview in 397 (35.5%), and abuse was diagnosed in 90 (8.1%) others. The CAGE was the best predictor of lifetime alcohol dependence, i.e., had the largest area under the curve (93%) and the highest sensitivity (84%), specificity (90%), positive predictive value (82%), and negative predictive value (91%). Among patients testing positive for alcohol, 63% had a lifetime alcohol dependence diagnosis. The CAGE is an efficient screening test to detect alcohol dependence in trauma center populations. It should be used in combination with alcohol testing to identify patients at risk of alcohol use problems.