New methodologies for pharmacological treatment trials for alcohol dependence.

As the number of pharmacological agents for the treatment of alcoholism continues to proliferate, there is an increasing need to design studies that accurately measure the efficacy of these agents. Key study design issues include selecting subjects, calculating patient retention, determining medication compliance, maintaining adequate sample size to ensure that a clinically meaningful therapeutic effect will not be missed, and measuring the study outcome, including the amount of alcohol consumed and levels of craving. Because there is evidence that self-report of alcohol consumption and medication compliance can be unreliable in subjects, biological methods have been used. One type of direct measurement, carbohydrate-deficient transferrin levels, has been found to be a reasonably sensitive indicator of alcohol consumption that may provide meaningful outcome information. The ingestion of riboflavin with medication and its urinary measurement is an accepted method of compliance detection. Improved study design will lead to continued advances in pharmacotherapy for the treatment of patients with alcoholism.