Carbohydrate-deficient transferrin (CDT) is a useful biomarker to identify excessive alcohol consumption; however, few studies have validated the %CDT cut-off value in elderly men. This study estimated the optimal %CDT cut-off value that could identify excessive alcohol consumption in men aged ≥65 years.

This retrospective study included 120 men who visited the department of family medicine at Chungnam National University Hospital for health check-up between January 2010 and August 2013. At-risk drinking included heavy- and binge drinking. Heavy drinking was defined as more than seven standard drinks/wk, and binge drinking was defined as more than three standard drinks/d. The cut-off %CDT values for at-risk drinking were determined using receiver operating characteristic (ROC) curves.

Based on the ROC curves, the optimal %CDT cut-off values in ≥65-year-old men were 1.95% for at-risk drinking, 1.81% for heavy drinking, and 2.07% for binge drinking. The sensitivity, specificity, and positive and negative predictive values were 58.7%, 83.6%, 69.2%, and 76.2% for at-risk drinking, respectively. The AUROC were >0.7 for all three evaluated cut-offs.

Our results suggest that the %CDT cut-off value for at-risk drinking in elderly Korean men (≥65 years) should be readjusted to a lower value of 1.95%.