Psychometric synthesis of the Alcohol Use Disorders Identification Test (AUDIT)

Abstract

Hazardous or harmful alcohol use is very prevalent among the US population. With 15.1 million adults in the United States currently identified as having an alcohol use disorder, identifying hazardous and harmful drinking earlier is necessary for early intervention strategies. The Alcohol Use Disorders Identification Test's (AUDIT) strong psychometric properties lead it to be a screening tool of choice across countries, cultures, and subgroups. A psychometric analysis aggregates all available evidence of reliability and validity to support the use of an instrument for screening, diagnosis, and treatment planning. AUDIT internal consistency for the 10-item total scale was α = 0.817 (n = 247,963). The two-factor model had the most structure validity support and aggregated convergent validity comparisons with other alcohol use assessments ranged from r = 0.578 to r = 0.701. The optimal cutoff score for diagnostic validity was 6/7, rather than the 8 recommended by the AUDIT authors. Clinically appropriate use and interpretation of the AUDIT as well as study limitations and suggestions for future research were discussed.