Two Brief Measures of Alcohol Use Produce Different Results: AUDIT‐C and Quick Drinking Screen

Author(s)

Brian Letourneau, Nova Southeastern University
Linda C. Sobell, Nova Southeastern UniversityFollow
Mark B. Sobell Ph.D., Center for Psychological StudiesFollow
Sangeeta Agrawal, Gallup Consulting, Omaha, Nebraska
Christopher J. Gioia, Nova Southeastern UniversityFollow

Document Type

Article

Publication Date

3-1-2017

Publication Title

Alcoholism: Clinical and Experimental Research

Volume

41

Issue/Number

5

First Page

1035

ISSN

1530-0277

Last Page

1043

Abstract/Excerpt

Background

Several psychometrically sound measures of alcohol use have been developed to assess drinking. The Alcohol Use Disorders Identification Test (AUDIT) and its shorter counterpart the AUDIT‐C, which contains the first 3 AUDIT questions, were developed by the World Health Organization and have become the preferred brief measures for screening and evaluating problem severity. This study compared the first 3 questions on the AUDIT with another psychometrically sound brief measure of alcohol use, the Quick Drinking Screen (QDS).

Methods

Data were obtained from a randomized controlled trial of a mail‐based intervention promoting self‐change with 472 alcohol abusers (n = 280, no prior alcohol treatment; n = 192, prior alcohol treatment). Participants’ retrospective self‐reports of alcohol consumption were collected using the QDS and the 3 AUDIT‐C questions and compared. Although both measures contain similar questions (2 quantity‐frequency and 1 binge drinking), they differ in 2 important ways: (i) temporal interval over which data are collected, and (ii) formatting of response options (i.e., a continuous number vs. categorical).

Results

Intraclass correlations for drinking variables were moderate to moderately high. A repeated‐measures MANOVA using treatment condition and gender as variables revealed significant differences in absolute values between the 2 drinking measures with the QDS showing greater consumption on almost all variables. Participants’ numerical answers on the QDS were compared to their categorical answers to the similar alcohol use questions on the AUDIT‐C. The comparison revealed that participants’ answers on the AUDIT‐C were associated with extreme variability compared to their QDS answers. This variability suggests the AUDIT‐C would be unreliable as a quantitative measure of alcohol consumption.

Conclusions

The differences between the 3 alcohol use questions on the AUDIT‐C and the same questions on the QDS may reflect the imprecision of the AUDIT‐C's drinking response categories. Results suggest that the QDS can be used to identify risky drinking and to provide a more informative characterization of a drinker's alcohol consumption than that provided by the AUDIT‐C.

DOI

https://doi.org/10.1111/acer.13364

NSUWorks Citation

Letourneau, B., Sobell, L. C., Sobell, M. B., Agrawal, S., Gioia, C. J. (2017). Two Brief Measures of Alcohol Use Produce Different Results: AUDIT‐C and Quick Drinking Screen. Alcoholism: Clinical and Experimental Research, 41(5), 1035-1043.
Available at: https://nsuworks.nova.edu/cps_facarticles/1583