Six months of abstinence from alcohol is a commonly used criterion for liver transplantation eligibility for patients with alcoholic cirrhosis. There is limited evidence to document the validity of this criterion with regard to risk of alcoholism relapse. Ninety-one patients with alcoholic cirrhosis were interviewed for relapse risk using the High Risk Alcoholism Relapse (HRAR) Scale. The HRAR model can be used to predict relapse risk independent of duration of sobriety and therefore can be used to examine the validity of the 6 months of abstinence criteria in this clinical population. The two methods demonstrated poor to fair agreement. Agreement was highest with a cutoff allowing a 5% 6-month relapse risk when 79% agreement (kappa = 0.56) was demonstrated between the two methods. Using the 6-month abstinence criterion alone disallows a significant number of candidates who have a low relapse risk based on their HRAR score. The validity of the 6-month abstinence criterion is supported somewhat by comparison with the HRAR model. However, use of the 6-month abstinence criterion alone forces a significant number of patients with a low relapse risk by HRAR to wait for transplant listing. A relapse risk model based on an estimate of alcoholism severity in addition to duration of sobriety may more accurately select patients who are most likely to benefit from liver transplantation.