Abstract

Background: In this study, we developed the Screening Test for At-risk Drinking in the Elderly (STAD-E), a new abbreviated version of the Alcohol Use Disorders Identification Test (AUDIT) intended for use in the elderly populations. This test comprises three questions that reflect the structure of the AUDIT 10 questionnaire and the characteristics of the Korean population and was developed using nationally representative data from the Korea National Health and Nutrition Examination Survey (KNHANES).

Methods: We performed exploratory factor analysis for each question from the AUDIT questionnaire that was responded by elderly (≥ 65 years) participants of KNHANES IV-V to derive our abbreviated test based on the structure of each AUDIT item. For validation, we analyzed the sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) of the new abbreviated test using KNHANES VI datasets (excepting KNHANES VI-2).

Results: We selected question (Q) 1, Q3, and Q7 on the three-factor structure for the elderly population. The cut-off values of STAD-E were 4 for elderly males and 3 for elderly females. During the validation test, STAD-E yielded significantly greater AUROC values than AUDIT-QF and similar values to AUDIT-C.

Conclusions: Unlike previous abbreviated tests, STAD-E reflects the item structure of AUDIT and the alcohol consumption patterns in an elderly population. Therefore, it can be used as a simple and reliable screening test for at-risk drinking in clinical settings.