Because interferon therapy exhibits low efficacy for cirrhotic patients infected with hepatitis C virus, this prospective study was conducted to determine effective interferon regimens tailored to treatment response by monitoring HCV RNA status. A total of 157 cirrhotic patients were enrolled to receive 9 million units (MU) of interferon three times a week. The HCV RNA values were drawn 8 weeks apart and the patients were randomized to a further 16 or 32 weeks of treatment after two sequential findings of negativity for HCV RNA. A total of 73 out of 157 patients (46%) proceeded to randomization to different durations of treatment, 37 short-course and 36 long-course (duration: 38 ± 8 and 49 ± 13 weeks; total amount of interferon: 940 ± 240 and 1130 ± 390 MU, respectively). The remaining 84 patients without two sequential negative serum HCV RNA determinations received 44.8 ± 27.4 weeks of interferon (IFN) therapy with total amount of 993 ± 633 MU. Of these 157 patients, sustained virological and biochemical response was shown in 32 (20%) and 37 patients (24%), respectively. Sustained virological and biochemical response rate in the randomized patients was significantly higher than in nonrandomized patients (41% vs. 2%, and 38% vs. 11%; each P < .01). Of the 73 randomized patients, the rate of sustained virological response in patients with long-course treatment (50%) was significantly higher than that of patients with short-course treatment (32%) (P = .026: log-rank test), and in patients with early disappearance of HCV RNA especially within 8 weeks, in patients with low virus load (≤106.3 copies/mL) and with HCV 2a. Multivariate analysis revealed that HCV RNA level and subtypes were the most important factors contributing to sustained virological response. Interferon is effective even in cirrhotic patients with low viral load and HCV 2a, but requires a longer course of administration.