Optimizing the Use of the US-Audit for Alcohol Screening in College Students

Introduction: Drinking is a serious public health problem in college. Recent estimates show adults 18-22 years old have higher rates of high-risk drinking than any other group, and college students in this age range have higher rates of high-risk drinking than non-college, same-age peers. Rates of alcohol use disorders during this life stage are about double other stages. Efficient identification and referral to treatment in college, especially in primary care and other health care settings, is increasingly popular. Using validated screening tools to identify at-risk students is vital. The Alcohol Use Disorders Identification Test (AUDIT) has been validated in adult samples and is often used with college students, but no studies have established optimal cutoff scores for the recent U.S. version (US-AUDIT).

Methods: 250 students (65% men) completed the US-AUDIT and a self-report analog measure for mental disorders. The US-AUDIT has 10 items (3 questions on alcohol consumption, 7 questions on alcohol-related harm and alcohol use disorder symptoms). This measure typically can be completed in 2-3 minutes. Items were modified from the original AUDIT for the U.S. (e.g., using U.S. standard drinks for binge drinking [5 for men, 4 for women]). For young adults, the recommended cutoffs are 8 or more for the 10-item US-AUDIT, and 7 for women/8 for men with the US-AUDIT1-3 (the first 3 items). Likely psychiatric disorder was assessed with 11 self-report questions mirroring DSM-5 symptoms of alcohol use disorder over a 12-month period; likely alcohol use disorder was defined as at least 2 positive responses.

Results: 50% of students had a likely alcohol use disorder. Receiver operating characteristic (ROC) curve analysis showed that the area under the ROC (AUROC) of the US-AUDIT was .80 (95% CI [.75, .86]). The AUROC of the US-AUDIT1-3 was .75 (95% CI [.68, .81]). The ideal cutoff of the 10-item US-AUDIT using the Youden method was 13 for men (sensitivity = .69, specificity = .81) and 8 for women (sensitivity = .83, specificity = .80). The ideal cutoff for the US-AUDIT1-3 was 10 for men (sensitivity = .61, specificity = .71) and 5 for women (sensitivity = .88, specificity = .71).

Conclusions: The best cutoff scores for the US-AUDIT and US-AUDIT1-3 for identifying alcohol use disorder in college students were different than recommendations, but in the same range as studies with the original AUDIT. Different cutoffs for men and women should be used, which may reflect differences in patterns of drinking. These empirical guidelines for alcohol screening with the new US-AUDIT may enhance identification of at-risk drinkers in college settings, or for college students in primary care and other health care settings.