A growing body of investigations demonstrate that elevated levels of carbohydrate-deficient transferrin (CDT) effectively distinguishes alcoholics recently consuming large amounts of alcohol from light social drinkers or teetotalers. Nevertheless, important questions still remain concerning the value of CDT as a more generalized marker of alcohol consumption. Most important, the nature of the drinking pattern, including quantity and frequency, necessary to raise levels of CDT significantly remains unclear. Neither has research convincingly demonstrated that CDT is as accurate a marker for women, young adults, or non-Caucasian ethnic groups as for White, middle-aged men. Whereas CDT might serve as a useful outcome measure in trials of alcoholism treatment effectiveness, current research suggests that CDT is of limited value in identifying problematic drinking in general medical or community settings in which a broad continuum of drinkers is represented. Combining CDT with other biochemical or self-report screening measures may, however, improve sensitivity in these contexts. At present, the most accurate laboratory technique to detect CDT seems to be isoelectric focusing. Additional research, however, is needed to resolve the issue of whether CDT is best quantitated as a simple value or if its ratio to total transferrin or non-CDT results in higher predictive validity.