BACKGROUND: The Alcohol Use Disorders Identification Test (AUDIT) is a 10-item alcohol screener that has been recommended for use in Aboriginal primary health care settings. The time it takes respondents to complete AUDIT, however, has proven to be a barrier to its routine delivery. Two shorter versions, AUDIT-C and AUDIT-3, have been used as screening instruments in primary health care. This paper aims to identify the AUDIT-C and AUDIT-3 cutoff scores that most closely identify individuals classified as being at-risk drinkers, high-risk drinkers, or likely alcohol dependent by the 10-item AUDIT. METHODS: Two cross-sectional surveys were conducted from June 2009 to May 2010 and from July 2010 to June 2011. Aboriginal Australian participants (N=156) were recruited through an Aboriginal Community Controlled Health Service, and a community-based drug and alcohol treatment agency in rural New South Wales (NSW), and through community-based Aboriginal groups in Sydney NSW. Sensitivity, specificity, and positive and negative predictive values of each score on the AUDIT-C and AUDIT-3 were calculated, relative to cutoff scores on the 10-item AUDIT for at-risk, high-risk, and likely dependent drinkers. Receiver operating characteristic (ROC) curve analyses were conducted to measure the detection characteristics of AUDIT-C and AUDIT-3 for the three categories of risk. RESULTS: The areas under the receiver operating characteristic (AUROC) curves were high for drinkers classified as being at-risk, high-risk, and likely dependent. CONCLUSIONS: Recommended cutoff scores for Aboriginal Australians are as follows: at-risk drinkers AUDIT-C>5, AUDIT-3>1; high-risk drinkers AUDIT-C>6, AUDIT-3>2; and likely dependent drinkers AUDIT-C>9, AUDIT-3>3. Adequate sensitivity and specificity were achieved for recommended cutoff scores. AUROC curves were above 0.90.
Calabria, B., Clifford, A., Shakeshaft, A. P., Conigrave, K. M., Simpson, L., Bliss, D., & Allan, J. (2014). Identifying Aboriginal-specific AUDIT-C and AUDIT-3 cutoff scores for at-risk, high-risk, and likely dependent drinkers using measures of agreement with the 10-item Alcohol Use Disorders Identification Test. Addiction Science & Clinical Practice, 9, 17. https://doi.org/10.1186/1940-0640-9-17