We have conducted a randomized study to compare the efficacy and tolerance of human interferon (IFN) beta vs recombinant IFN-alpha 2b in patients with chronic active hepatitis C. Forty patients were included: 21 received IFN-alpha (group A) and 19 IFN-beta (group B). IFN was administered intramuscularly at a dose of 6 MU three times a week (tiw) for 2 months (induction phase), followed by 3 MU tiw for 4 months. Clinical, epidemiological and pathological features were similar in the two groups. Normal alanine aminotransferase (ALT) values at the end of treatment was regarded as a response to therapy and the response rate was 57% (12/21) in group A and 5.2% (1/19) in group B (P < 0.01). Both types of IFN induced a significant decrease in mean ALT values by the end of the induction phase (P < 0.01). When the dose was reduced to 3 MU, a marked, but not significant increase in ALT, was seen in group B, whereas no increase was seen in group A. IFN-beta was better tolerated and haematological adverse effects (platelet and leucocyte decrease) were less pronounced with IFN-beta. Hence, human IFN-beta was less effective than IFN-alpha in treating chronic hepatitis C virus (HCV). Doses of IFN-beta of 3 MU intramuscular (IM) tiw were clearly insufficient and it remains to be established whether higher doses of intramuscularly IFN-beta can be useful.