The survival of ovarian cancer patients has been reported to be superior at hospitals with a high volume of operations. A population-based study was carried out to assess whether this is true in Japan, where the incidence rate is relatively low as compared with other developed countries. The Osaka Cancer Registry's data were used to investigate associations between hospital procedure volume and survival of ovarian cancer patients. Hospitals were ranked according to the number of operations for ovarian cancer performed per year (high/medium/low/very low). Survival analysis was restricted to the reported 2450 cases who lived in Osaka Prefecture (except for Osaka City) diagnosed in 1975-1995, or those who resided in Osaka City in 1993-1995, since active follow-up data on vital status 5 years after the diagnosis were available. The relative 5-year survival for all ovarian cancer cases was 38.8%, and the survival was higher with greater hospital volume (22.3% / 34.2% / 46.2% / 55.0%). After adjustment for age, histologic type and cancer stage by the Cox regression model, patients receiving care in very-low-volume hospitals were seen to have a 60% higher risk of death than patients receiving care in high-volume hospitals (P < 0.01). Although some limitations existed in this study, the results indicated that further centralization of operative treatment in high-volume hospitals might improve survival of ovarian cancer patients in Japan.