Three-year survival from diagnosis in surgically treated patients in designated and non-designated cancer care hospitals in Japan.

The Japanese national and prefectural governments accredited high-capacity, high-experience cancer care hospitals as "designated cancer care hospitals" to standardize cancer care, centralize patients, and improve clinical outcomes, but the performance of these designated hospitals has not been evaluated. We retrospectively compared 3-year patient survival in national, prefectural, and non-designated cancer care hospitals in 2010-2012 in Osaka using registry-based data of 86,456 surgically treated cancer patients aged 15 years or older. Hazard ratios and 3-year survival probabilities were compared among national, prefectural, and non-designated hospitals using a Cox-proportional hazard regression model. Subgroup analyses for six cancers (stomach, colorectum, lung, breast, uterus, and prostate) and other cancers were performed. In 2010-2012, 36,634 (42.4%), 38,048 (44.0%), and 11,774 (13.6%) patients were treated at national, prefectural, and non-designated hospitals, respectively. The mortality hazard for all-site cancer was significantly lower in national and prefectural designated hospitals (adjusted hazard ratio 0.60, 95% confidence interval 0.53-0.68 and 0.72, 0.66-0.80) than in non-designated hospitals. The adjusted 3-year survival probabilities for all-site cancer were 86.6%, 84.2%, and 78.8% in national, prefectural, and non-designated hospitals, respectively. Site-specific subgroup analyses revealed significantly lower hazard ratios in national and prefectural hospitals than in non-designated hospitals for stomach, colorectal, lung, breast, and other cancers. To conclude, the majority of cancer patients underwent surgeries at designated hospitals and had higher 3-year survival probabilities than those treated at non-designated hospitals. Further centralization of patients from non-designated to designated hospitals may improve population-level survival.