Purpose: Aims to establish the critical score and screening accuracy of the CAGE Questionnaire in three treatment settings--primary health care, walk-in (triage) clinic and the emergency room.
Design/methodology/approach: Taguchi methods are applied to three screens of the CAGE questionnaire.
Findings: Analysis of the sensitivity and specificity data of three CAGE screens by leveling factor (p'), signal-to-noise ratios (S/N, SS/N) and their dependent relation resulted in critical CAGE scores of 1, 1 and 2; and high screening accuracy levels of 98.44, 97.20 and 94.92 percent, respectively. The illustrated method yielded excellent (> or = 95 percent) screening accuracy values for primary health care, emergency room and walk-in clinic patients.
Originality/value: To reduce misclassification rates of alcohol abuse, screening systems should concentrate first on developing ways to standardize protocols. Further work is needed to establish high screening accuracy in other clinical settings, and particularly in those at risk of alcohol abuse in the general population.