The effect of ursodeoxycholic acid on liver function tests and on bile acid metabolism was investigated in a multi-center randomized controlled dose study for chronic hepatitis C. Twenty, 18 and 19 patients were administered 150, 600 and 900 mg/day, respectively of ursodeoxycholic acid every day for 16 wk. Serum liver parameters and bile acid composition in the treatment groups were compared with 17 control patients. A similarly significant decrease of serum alanine aminotransferase and serum γ-glutamyltransferase was observed in patients administered 600 and 900 mg of ursodeoxycholic acid. Serum bile acid composition was determined by high-performance liquid chromatography. At entry, the relative proportions of major bile acids were similar to those observed in normal individuals. Maximal concentrations of total ursodeoxycholic acid were 0.30 μmol/L, 5.59 μmol/L, 21.42 μmol/L and 14.73 μmol/L in the control, 150, 600 and 900 mg/day groups, respectively. The fraction of the total ursodeoxycholic acid increased in a dose-dependent manner, and it was significantly higher than in controls (p < 0.001). The hydrophobicity index of bile acids was calculated by the method of Heuman, and its correlation with serum parameter levels was analyzed. In the 600 and 900 mg/day dose groups, serum alanine aminotransferase decreased in the cases in which hydrophobicity index significantly decreased during treatment. The same correlation was observed between the hydrophobicity index and serum γ = glutamyltransferase in these two groups. There was no correlation between these parameters in the control and 150-mg groups. There was no correlation between reduction rate of serum alanine aminotransferase and initial liver histology. These data suggest that ursodeoxycholic acid at a dose of 600 mg/day may be effective in improving liver function in patients with chronic hepatitis C. (HEPATOLOGY 1994;20:558–564).