Related Experiment Video
Updated: Aug 11, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Development of an Australian version of the Alcohol-Related Problems Survey: a comprehensive computerised screening
Stephen J Bright1, Arlene Fink, John C Beck
1Peninsula Health, Frankston, Victoria, Australia; Curtin University, Perth, Western Australia, Australia.
Aim:
The Alcohol-Related Problems Survey (ARPS) reliably classifies drinking as non-hazardous, hazardous or harmful using scoring algorithms that consider quantity and frequency of alcohol use alone and in combination with health conditions, medication-use and functional status. Because it has been developed using a 14-g US standard drink, it is not valid in Australia where a standard drink contains 10 g of ethanol.
Method:
We recalibrated the ARPS scoring algorithms for a 10-g Australian standard drink and updated the list of medications. The Australian ARPS (A-ARPS) was then administered to 50 non-treatment-seeking participants in waves of five.
Results:
The A-ARPS recalibrated scoring algorithms reliably classified all 50 individuals. Sixty-six per cent were classified as hazardous or harmful drinkers. Many were taking medications that interact with alcohol or had medical conditions that can be exacerbated by alcohol consumption.
Conclusion:
The A-ARPS is available for use in Australia. Its utilisation could reduce the incidence of alcohol-related harms.
