One hundred and eleven patients with non-A, non-B/type C (NANB/C) chronic active hepatitis were randomly assigned to two groups to receive recombinant interferon alfa-2b treatment as follows: 3 million units (MU) interferon three times weekly for six months or 3 MU interferon three times weekly for the first six months, 2 MU for the next three months, and 1 MU for the last three months. At the end of treatment, the number who had responded completely was similar in both groups (47.2% in the six months group v 41.4% in the 12 months group). Cirrhosis was found to be the only predictive factor for response; a complete response was observed in 50.6% of patients without cirrhosis v 33.3% of those with cirrhosis (p = 0.04). After one year of treatment, the rate of relapse was lower in patients who had received longer treatment (33%) than in those who had received six months of interferon (60%). The difference between the groups was not statistically significant and the trend favouring longer treatment for sustained response requires further confirmation.