The changes in the quality of Japanese clinical trials were evaluated by comparing the results of Good Clinical Practice (GCP) audits conducted from April 1997 to March 2000 (fiscal year (FY) 1997-1999) with those from April 2001 to March 2002 (FY2001). During both of the periods inspections were undertaken by the Organization for Pharmaceutical Safety and Research (OPSR). The audit findings in the former period were based on the audits that covered 331 hospitals and 775 trials conducted under the old GCP guideline. The audits in the latter period targeted 147 hospitals and 238 trials conducted under the old or new GCP guideline. The total number of deficiencies detected by GCP audits in the former three-year period (FY 1997-1999) was 1529, and the corresponding number in the latter single year (FY 2001) was 912. Two remarkable changes in OPSR's findings were observed between FY 1997-1999 and FY 2001 as follows; the proportion of protocol deviations increased from 14.7% (225/1529) to 53.1% (484/912), while the proportion of errors in case report forms (CRFs) decreased from 43.6% (666/1529) to 15.4% (140/912). The new GCP guideline sets very high standards for a hospital's qualification: to have sufficient equipment and hospital resources, to have capacity for promptly responding to urgent trial-related problems, to have an IRB, and to have appropriate staff including clinical research coordinators (CRCs) assigned to the clinical trial. Our results suggest that the impact of the regulatory changes of applicable standard is large for a hospital's qualification for conducting clinical trials in Japan.