The effect of ursodeoxycholic acid administration on liver function tests and on bile acid metabolism was investigated in 18 patients with chronic active hepatitis. Three different doses of ursodeoxycholic acid—250 mg, 500 mg and 750 mg—were administered daily to each patient for consecutive 2-mo periods. The order of doses was randomly assigned according to a replicated Latin-square design. A significant decrease in serum transaminases and γ-glutamyl transpeptidase occurred with the lowest dose of ursodeoxycholic acid, which corresponded to 4 mg/kg body wt/day, and no further significant decrease with the higher doses was seen. Biliary bile acid composition was determined by high-performance liquid chromatography and gas chromatography-mass spectrometry. At entry the relative proportions of major bile acids were similar, to those observed in normal individuals. During treatment the mean percentage of ursodeoxycholic acid in bile (22% with the 250 mg dose, 32% with the 500 mg dose and 34% with the 750 mg dose) was lower than values previously reported for patients with gallstones and normal liver function. The major bile acids were cholic, chenodeoxycholic and deoxycholic acids. A number of unusual bile acids were identified by gas chromatography-mass spectrometry, but these accounted for only 3% to 5% of the total and did not change during ursodeoxycholic acid therapy. No correlation between the improvement in liver function tests and the percentage of ursodeoxycholic acid in bile existed. These data suggest that even a slight enrichment of bile with ursodeoxycholic acid, as is attained with 250 mg/day, is effective in improving biochemical markers of liver function in patients with chronic active hepatitis. (HEPATOLOGY 1991;13:339–344).