The reliability of alcoholism history in patients with alcohol-related cirrhosis

Alcoholic liver disease is considered an indication for liver transplantation when a candidate is felt to have a high likelihood of abstinence following transplantation. Historical variables such as duration of sobriety, duration and quantity of drinking, and treatment history are commonly used to estimate alcoholism prognosis, yet their reliability and validity in patients with alcoholic cirrhosis has received limited study. Fifty subjects (9 women and 41 men) with alcoholic cirrhosis underwent an alcoholism history interview. Each subject had a collateral source (usually a spouse) who was interviewed by a second interviewer blind to the subject's alcoholism history. The two histories were compared for duration of abstinence in months and estimated alcoholism relapse risk was calculated using the High-risk Alcoholism Relapse scale (HRAR). Duration of sobriety correlated highly between subject and collateral source (Spearman r= 0.96, P = 0.0001) as did HRAR total score (Spearman r = 0.72, P = 0.0001). Categorical assignments also showed high correlations with duration of sobriety (kappa = 0.97) and HRAR category (kappa = 0.63). When disagreements were present, collateral sources tended to underestimate severity of alcoholism. We conclude that patients with alcoholic liver disease provide a reliable history for alcoholism variables when compared with a collateral source, and that, when disagreements are present, subjects tend to report a more acute or severe alcohol problem. The results support the clinical use of patient history information in making decisions about medical interventions for alcoholic liver disease.