Validation of the screening test for at-risk drinking in an emergency department using a tablet computer

Abstract

Background: The Alcohol Use Disorder Identification Test (AUDIT) is widely used and validated in primary care settings for alcohol screening, yet practical challenges in conducting it in crowded clinics exist. Recently, a new abbreviated version of the AUDIT, was presented: the Screening Test for At-risk Drinking (STAD). This study aimed to evaluate the performance of STAD compared to other abbreviated versions of AUDIT for patients visiting the emergency department (ED). Methods: This cross-sectional survey was conducted with 543 patients in the urban tertiary academic hospital ED in South Korea We diagnosed at-risk drinking using the entire AUDIT score. The optimal cut-off values, sensitivity, specificity, and the area under the receiver operating characteristics (AUROC) of the STAD were analyzed. We compared the AUROC with AUDIT-C and AUDIT-QF, which are previously abbreviated versions of AUDIT. Results: For males, the optimal cut-off value in the STAD test was 3 points with 83.1% sensitivity (95% CI: 75.3–89.2) and 95.9% specificity (95% CI: 91.2–98.5). For females, the optimal cut-off value was 2 points with 95.9% sensitivity (95% CI: 88.5–99.1) and 89.1% specificity (95% CI: 83.9–93.0). The AUROC curves for STAD were 0.964 (95% CI: 0.934–0.983) for males and 0.980 (95% CI: 0.965–0.993) for females. Conclusions: The STAD is a simple and useful test to screen at-risk drinking in the ED, and its assisted applications will enable faster and more efficient screening and management of at-risk drinking. © 2021 Elsevier B.V.