The most cost effective strategy for antiviral therapy of chronic hepatitis C is the earliest identification and treatment of patients at risk of developing life-threatening complications such as hepatocellular carcinoma. Liver fibrosis represents the best predictor of unfavourable outcome. However, some patients with liver fibrosis already have a histological diagnosis of cirrhosis and there is a debate about whether alpha interferon is still effective in lowering the risk of disease progression in such patients. We identified some of the reasons that may explain seemingly contradictory results of studies addressing this issue. A major cause appears the beginning of follow-up at different starting points during the course of clinically compensated cirrhosis. Some investigators recruited patients because of anti-HCV positivity and elevated transaminases and found cirrhosis only at histology, whereas others recruited patients because cirrhosis had been diagnosed. Ultrasonographic signs of portal hypertension appear to be a useful tool to distinguish the two phases of the disease. Another important cause of reduced response rate to antiviral therapy is the presence of cofactors of liver disease and hepatocellular carcinoma such as present or past HBV infection. Early phase cirrhotics without cofactors appear to benefit most from therapy with a significant lower risk for hepatocellular carcinoma than untreated controls. The therapeutic decision in these patients could be the same as in patients with chronic hepatitis C without cirrhosis. In contrast, the efficacy of interferon remains questionable in HCV patients who already have ultrasonographic signs of portal hypertension and/or past or present HBV coinfection. Prospective, randomized clinical trials should be performed after stratification of these patients for stage and cofactors of liver disease.