Objective: The aim of this study was to evaluate the diagnostic usefulness of the brief versions of the Alcohol Use Disorders Identification Test (AUDIT) for detecting hazardous drinkers and to compare it with that of the full-AUDIT in primary care settings.
Method: Five hundred patients were randomly selected in a primary care center. An interview on quantity-frequency was administered for assessment of weekly alcohol intake. The standard used for classification of hazardous drinkers was a weekly alcohol consumption of 280 g for men and 168 g for women. Cut-off points were 8 for the full-AUDIT, 1 for the AUDIT-3 (third item), 3 for the AUDIT-C (items 1, 2 and 3), 5 for the AUDIT-PC (items 1, 2, 4, 5 and 10) and 3 for the modified Fast Alcohol Screening Test (m-FAST; items 3, 5, 8 and 10). Sensitivity, specificity, positive and negative predictive values, and areas under the receiver operating characteristic (AUROC) curves were measured.
Results: Diagnostic usefulness of the questionnaires for detecting hazardous drinkers was for the full-
Audit: 81.4% sensitivity, 94.6% specificity and 0.97 AUROC curve; for the AUDIT-3: 83.1% sensitivity, 90.9% specificity and 0.89 AUROC curve; for the AUDIT-C: 100% sensitivity, 79.4% specificity and 0.97 AUROC curve; for the AUDIT-PC: 98.3% sensitivity 90.9% specificity and 0.97 AUROC curve; and for the m-FAST: 79.7% sensitivity, 93.7% specificity and 0.93 AUROC curve.
Conclusions: The AUDIT-C and AUDIT-PC show a higher sensitivity, lower specificity and a similar AUROC curve than the full-AUDIT, thus allowing their use as screening instruments that are as reliable as the original test for detecting hazardous drinkers. The AUDIT-3 and m-FAST, when compared with the full-AUDIT, performed less well, therefore limiting their use for this purpose.