Evidence of benefit from centralised treatment of ovarian cancer: a nationwide population-based survival analysis in Finland.

To assess the effect of different hospital types or surgical volume on the survival of ovarian cancer patients, a nationwide and population-based analysis was carried out in Finland. The study included all 3,851 ovarian cancer patients operated from 1983-94. The patients were classified according to the hospital of the first surgery. The hospitals were categorized by type (university, central or other hospital) and, separately, into quartiles by the number of operated patients (surgical volume). The patients operated at university hospitals had better survival than those operated in central hospitals, the 5-year relative survival rates (RSR) being 45% (95% CI = 42-48%) and 37% (34-40%), respectively. RSR in the 'other hospital' category was 45% (42-48%). The RSR for the patients operated in the highest volume hospitals was 47% (43-50%), and by decreasing volume (quartile) the RSR was 40% (36-43%), 40% (36-43%) and 42% (38-45%), respectively. After controlling for potential confounding by stage and age using regression models, the results remained practically the same. The results indicate that further centralizing of operative treatment of ovarian cancer may still improve survival rates on a population level in Finland. Copyright 2002 Wiley-Liss, Inc.