Screening for Problem Drinking in College Freshmen

Abstract

Several brief questionnaires have been developed for use in screening for alcoholism. College students manifest a different spectrum of drinking behaviors from adult alcoholics, and these questionnaires have not been validated for use in this group. This study examines the ability of the CAGE, Brief MAST, and Trauma Score to identify problem drinkers among college students. Problem drinkers were identified as individuals who both drank more heavily than 75% of their peers and had experienced a negative social or physical consequence as a result of their drinking. They were compared with a group of drinkers with neither characteristic. The CAGE was best able to separate these groups, with a sensitivity of 57% and a specificity of 85%. The Trauma Score and Brief MAST had optimal sensitivities of 45% and 34% and specificities of 79% and 46%, respectively. When used as a screening test for problem drinking in a population similar to ours, the CAGE will fail to identify 43% of problem drinkers due to its low sensitivity and will falsely label 15% of all normal drinkers as problem drinkers. None of these widely used alcohol questionnaires is optimal for use in screening for problem drinking in college students.