Exploratory Validation Study of the Individual AUDIT-C Items among Older People.

Abstract

The AUDIT-C is a brief and commonly used alcohol screening tool, with few data available on the relative validity of the three individual items in older adult populations. The aim of this study was to explore the validity of the AUDIT-C items in identifying unhealthy drinking among older people, with a view to developing a single-item screener. A sample of 143 older adults (mean age=71) were recruited from non-clinical settings in the UK. AUDIT-C scores were compared to an unhealthy drinking reference category of consuming more than the UK recommended weekly units of alcohol. Standard analyses were conducted for men and women, and for those prescribed medications for long term conditions. The AUDIT-C items performed well in identifying unhealthy drinking in this sample of older people, with generally high sensitivity, specificity and area under the ROC curve. No significant differences were found in the validity of the three items, though the combined sensitivity and specificity scores and ROC values for item 3 were consistently slightly lower than for items 1 and 2. The findings were similar for men and women, and for participants prescribed medications for long-term conditions. AUDIT-C items 1 and 2 performed as well as item 3 in identifying unhealthy drinking among older people in this study. Both are reasonable single-item screener candidates, especially given relative ease of administration, with further validation study needed to examine psychometrics and how alcohol screening for older people can best be implemented in clinical settings.