Patients with overt alcoholic liver disease who had participated in a multicenter therapeutic trial and subgroups of controls (i.e., alcoholic patients without liver disease and patients with neither alcoholism nor liver disease) were tested for hepatitis B virus and hepatitis C virus antibodies to determine the prevalence of these antibodies to determine the prevalence of these antibodies and any clinical association in the progression and outcome of alcoholic liver disease. Antibodies to hepatitis B (anti-HBs and/or anti-HBc) were found in 29.2% of patients with alcoholic liver disease, in 26.1% of hospitalized alcoholic patients without liver disease and in 24.2% of hospitalized nonalcoholic patients without liver disease; frequencies were not significantly different from one another. HBsAg was not evaluated because HBsAg+ patients had been excluded from the original trial. The presence of these antibody markers correlated with ethnic origin of and immunoglobulin levels in the patients. In contrast, antibody to hepatitis C, as detected by enzyme immunoassay, was positive in 27.1%, 4.8% and 3.0% of the three groups, respectively, the first differing significantly from the other two. Antibody to hepatitis C virus positivity correlated significantly with clinical severity of the disease and with the presence of histological features that imply chronic viral infection (periportal inflammation, cirrhosis), despite the fact that the supplementary assay for antibody to hepatitis C virus, using recombinant immunoblot assay, reduced the positive rate by 79%.(ABSTRACT TRUNCATED AT 250 WORDS)