Sinhalese AUDIT: a screening instrument to detect alcohol risk groups

Abstract

OBJECTIVES: To translate and validate the alcohol use disorders identification test (AUDIT) in Sri Lanka. DESIGN, SETTING AND METHODS: This validation study was conducted in community and hospital settings. The translation and adaptation were carried out according to the WHO guidelines which included forward translation, expert evaluation, backward translation, pre-testing and international consensus. The adaptation included beverage specific visual aids and a conversion chart to convert milliliters consumed in to grams. The instrument was rated in terms of internal consistency, test- re-test reliability, and construct and criterion validity. The cut-off values to differentiate hazardous drinking (HZD) and alcohol use disorders (AUD) from low risk drinking (LRD), and alcohol use disorders (AUD) from hazardous drinking (HZD) and low risk drinking (LRD) were determined. RESULTS: The Cronbach's alpha coefficient was 0.89. The area under ROC curve to differentiate HZD and AUD from LRD was 0.962 (95%CI=0.939-0.985) and the cut off of 7 was associated with a sensitivity of 94.3%, specificity of 90.3% and positive and negative likelihood ratios of 9.7 and 0.06 respectively. The area under ROC curve to differentiate AUD from HZD and LRD together was 0.967 (95%CI=0.949-0.985) and the cut off of 16 was associated with a sensitivity of 96.6%, specificity of 74.7% and positive and negative likelihood ratios of 3.8 and 0.05 respectively. CONCLUSION: The AUDIT is a reliable and valid instrument, which can differentiate HZD and AUD from LRD and AUD from HZD and LRD among married men both in community and hospital settings.