Does This Patient Have Alcohol Use Disorder?: The Rational Clinical Examination Systematic Review

The accuracy of screening tests for alcohol use disorder (defined as a problematic pattern of alcohol use leading to clinically significant impairment or distress) requires reassessment to align with the latest definition in the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-5).

To assess the diagnostic accuracy of screening tools in identifying individuals with alcohol use disorder as defined in the DSM-5.

The databases of MEDLINE and Embase were searched (January 2013-February 2023) for original studies on the diagnostic accuracy of brief screening tools to identify alcohol use disorder according to the DSM-5 definition. Because diagnosis of alcohol use disorder does not include excessive alcohol use as a criterion, studies of screening tools that identify excessive or high-risk drinking among younger (aged 9-18 years), older (aged ≥65 years), and pregnant persons also were retained.

Sensitivity, specificity, and likelihood ratios (LRs) were calculated. When appropriate, a meta-analysis was performed to calculate a summary LR.

Of 4303 identified studies, 35 were retained (N = 79 633). There were 11 691 individuals with alcohol use disorder or a history of excessive drinking. Across all age categories, a score of 8 or greater on the Alcohol Use Disorders Identification Test (AUDIT) increased the likelihood of alcohol use disorder (LR, 6.5 [95% CI, 3.9-11]). A positive screening result using AUDIT identified alcohol use disorder better among females (LR, 6.9 [95% CI, 3.9-12]) than among males (LR, 3.8 [95% CI, 2.6-5.5]) (P = .003). An AUDIT score of less than 8 reduced the likelihood of alcohol use disorder similarly for both males and females (LR, 0.33 [95% CI, 0.20-0.52]). The abbreviated AUDIT-Consumption (AUDIT-C) has sex-specific cutoff scores of 4 or greater for males and 3 or greater for females, but was less useful for identifying alcohol use disorder (males: LR, 1.8 [95% CI, 1.5-2.2]; females: LR, 2.0 [95% CI, 1.8-2.3]). The AUDIT-C appeared useful for identifying measures of excessive alcohol use in younger people (aged 9-18 years) and in those older than 60 years of age. For those younger than 18 years of age, the National Institute on Alcohol Abuse and Alcoholism age-specific drinking thresholds were helpful for assessing the likelihood of alcohol use disorder at the lowest risk threshold (LR, 0.15 [95% CI, 0.11-0.21]), at the moderate risk threshold (LR, 3.4 [95% CI, 2.8-4.1]), and at the highest risk threshold (LR, 15 [95% CI, 12-19]). Among persons who were pregnant and screened within 48 hours after delivery, an AUDIT score of 4 or greater identified those more likely to have alcohol use disorder (LR, 6.4 [95% CI, 5.1-8.0]), whereas scores of less than 2 for the Tolerance, Worried, Eye-Opener, Amnesia and Cut-Down screening tool and the Tolerance, Annoyed, Cut-Down and Eye-Opener screening tool identified alcohol use disorder similarly (LR, 0.05 [95% CI, 0.01-0.20]).

The AUDIT screening tool is useful to identify alcohol use disorder in adults and in individuals within 48 hours postpartum. The National Institute on Alcohol Abuse and Alcoholism youth screening tool is helpful to identify children and adolescents with alcohol use disorder. The AUDIT-C appears useful for identifying various measures of excessive alcohol use in young people and in older adults.