Screening for At-Risk Drinking in a Population Reporting Symptoms of Depression: A Validation of the AUDIT, AUDIT-C, and AUDIT-3.

Abstract

Excessive alcohol use is common in patients presenting with symptoms of depression. The aim of this study was to evaluate how the Alcohol Use Disorders Identification Test (AUDIT) and its most commonly used abbreviated versions perform in detecting at-risk drinking among subjects reporting symptoms of depression. A subsample (n=390; 166 men, 224 women) of a general population survey, the National FINRISK 2007 Study, was used. Symptoms of depression were measured with the Beck Depression Inventory-Short Form and alcohol consumption with the Timeline Follow-back (TLFB). At-risk drinking was defined as ≥280g weekly or ≥60g on at least 1 occasion in the previous 28days for men, 140 and 40g, respectively, for women. The AUDIT, AUDIT-C, and AUDIT-3 were tested against the defined gold standard, that is, alcohol use calculated from the TLFB. An optimal cutoff was designated as having a sensitivity and specificity of over 0.75, with emphasis on specificity. The AUDIT and its abbreviations were compared with carbohydrate-deficient transferrin (CDT) and gamma-glutamyltransferase. At-risk drinking was common. The AUDIT and AUDIT-C performed quite consistently. Optimal cutoffs for men were ≥9 for the AUDIT and ≥6 for AUDIT-C. The optimal cut-offs for women with mild symptoms of depression were ≥5 for the AUDIT and ≥4 for AUDIT-C. Optimal cutoffs could not be determined for women with moderate symptoms of depression (specificity <0.75). A nearly optimal cutoff for women was ≥5 for the AUDIT. The AUDIT-3 failed to perform in women, but in men, a good level of sensitivity and specificity was reached at a cutoff of ≥2. With standard threshold values, the biochemical markers demonstrated very low sensitivity (9 to 28%), but excellent specificity (83 to 98%). Screening for at-risk drinking among patients presenting with symptoms of depression using the full AUDIT is recommended, although the AUDIT-C performed almost equally well. Cut-offs should be adjusted according to gender, but not according to the severity of depressive symptoms. The AUDIT and its abbreviations were superior to biochemical markers.