To determine whether a higher dosage of interferon (IFN) associated with ribavirin and/or prolonged time of administration may improve therapeutic efficacy, we conducted a 4-arm randomized trial on patients with chronic hepatitis C not responding to one or more previous treatment courses with IFN monotherapy. Group 1 (n = 139) received 3 million units (MU) IFN-alpha2b 3 times a week (t.i.w.) plus ribavirin 1,000 mg/d for 12 months; group 2 (n = 162) received 5 MU t.i.w. plus ribavirin for 12 months; group 3 (n = 142) received 3 MU t.i.w. plus ribavirin for 6 months; and group 4 (n = 151) received 5 MU t.i.w. plus ribavirin for 6 months. The primary end point was hepatitis C virus (HCV)-RNA clearance at the end of 6-month follow-up. HCV-RNA was negative in 15% of group 1, 23% of group 2, 11% of group 3, 16% of group 4 (group 2 vs. group 3, P =.04). Among patients with genotypes 1 and 4, sustained response was significantly higher in group 2 vs. group 3 (18% vs. 7%, P =.03; group 1 = 9%, group 4 = 12%, P = not significant [NS]). In patients with genotypes 2 and 3, sustained virologic response was not affected by the different regimens (group 1 = 32%, group 2 = 30%, group 3 = 30%, group 4 = 35%, P = NS). In conclusion, about 23% of nonresponders to IFN monotherapy may achieve a sustained response if re-treated by 5 MU t.i.w. IFN plus ribavirin 1,000 mg/d for 1 year. Patients with genotype 1 should receive a high dosage of IFN plus ribavirin for 12 months, whereas therapy for patients with genotype 2 or 3 should be less aggressive.