Animal and in vitro studies suggest that ursodeoxycholic acid (UDCA) can induce cytochrome P450 3A (CYP3A) expression and enhance its activities. On the other hand, Becquemont et al. demonstrated that UDCA had no influence on intestinal CYP3A activities. The aim of this study was to investigate the effects of UDCA on the intestinal and hepatic CYP3A activities by administration of midazolam (MDZ), as a specific probe for CYP3A activity, in humans. This was a randomized, open-label, crossover study with two phases in 14 healthy volunteers. The volunteers received UDCA (300 mg/day) or placebo orally for 9 days. The pharmacokinetics and pharmacodynamics of intravenous MDZ (5 microg/kg) and oral MDZ (15 microg/kg) were assessed on days 8 and 9, respectively. The pharmacodynamics of MDZ was estimated by measuring peak saccadic velocity, postural away length, critical fusion flicker frequency, and visual analogue scale. UDCA did not affect the pharmacokinetic and pharmacodynamic parameters of intravenous and oral MDZ administrations. Our study suggests that the clinical dosage of UDCA could not affect both hepatic and intestinal CYP3A activities and that the drug interaction between UDCA and substrates for CYP3A is unlikely in humans.