Comparing short versions of the AUDIT in a community-based survey of young people
1 Centre for Population Health, Burnet Institute, 85 Commercial Rd, Melbourne, VIC, 3004, Australia
2 Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, VIC, Australia
3 The Nossal Institute for Global Health, The University of Melbourne, Melbourne, VIC, Australia
BMC Public Health 2013, 13:301 doi:10.1186/1471-2458-13-301Published: 4 April 2013
The 10-item Alcohol Use Disorders Identification Test (AUDIT-10) is commonly used to monitor harmful alcohol consumption among high-risk groups, including young people. However, time and space constraints have generated interest for shortened versions. Commonly used variations are the AUDIT-C (three questions) and the Fast Alcohol Screening Test (FAST) (four questions), but their utility in screening young people in non-clinical settings has received little attention.
We examined the performance of established and novel shortened versions of the AUDIT in relation to the full AUDIT-10 in a community-based survey of young people (16–29 years) attending a music festival in Melbourne, Australia (January 2010).
Among those reporting drinking alcohol in the previous 12 months, the following statistics were systematically assessed for all possible combinations of three or four AUDIT items and established AUDIT variations: Cronbach’s alpha (internal consistency), variance explained (R2) and Pearson’s correlation coefficient (concurrent validity). For our purposes, novel shortened AUDIT versions considered were required to represent all three AUDIT domains and include item 9 on alcohol-related injury.
We recruited 640 participants (68% female) reporting drinking in the previous 12 months. Median AUDIT-10 score was 10 in males and 9 in females, and 127 (20%) were classified as having at least high-level alcohol problems according to WHO classification.
The FAST scored consistently high across statistical measures; it explained 85.6% of variance in AUDIT-10, correlation with AUDIT-10 was 0.92, and Cronbach’s alpha was 0.66. A number of novel four-item AUDIT variations scored similarly high. Comparatively, the AUDIT-C scored substantially lower on all measures except internal consistency.
Numerous abbreviated variations of the AUDIT may be a suitable alternative to the AUDIT-10 for classifying high-level alcohol problems in a community-based population of young Australians. Four-item AUDIT variations scored more consistently high across all evaluated statistics compared to three-item combinations. Novel AUDIT versions may be more effective than many established shortened versions as an alternative screening tool to the AUDIT-10 to measure hazardous or harmful alcohol consumption in this population.