Use of period analysis for providing more up-to-date estimates of long-term survival rates: empirical evaluation among 370 000 cancer patients in Finland

Long-term survival rates, such as 5-year or 10-year survival rates, are essential outcome measures of cancer, and they are now routinely reported by many cancer registries around the world. 1-5 However, long-term cumulative survival estimates obtained by traditional survival analysis pertain to cohorts of patients diagnosed many years ago, 6-8 which hinders early detection of improvements in cancer prognosis. In the case of improved survival, which has been achieved for many common forms of cancer, available traditional long-term survival estimates lag behind the survival experience to be expected by newly diagnosed patients, and they may thereby unduly discourage newly diagnosed cancer patients and their physicians.