[Population-based cancer survival: an analysis of 16,922 cases]

In the period of 1982-1991, 16,922 patients with cancers from 15 sites of Qidong population-based cancer registry were analysed for evaluation of cancer survival as well as different cancer control measures.

Observed survival rate (OS) was computed by the Kaplan-Meier method using EGRET statistical software package. Relative survival (RS) which is the ratio of the OS to the expected rate was calculated by using Qidong life table with respect to sex, age and calendar period of observation.

The five-year OS for the 5 leading sites of cancers, liver, stomach, lung, oesophagus, and rectum were 1.8%, 11.6%, 3.0%, 3.3%, and 19.9%, respectively. The five-year RS for the 5 sites were 1.9%, 14.0%, 3.6%, 4.2%, and 23.7%, respectively, in which, 1.7%, 14.8%, 3.4%, 4.2%, and 26.0% for males, and 2.7%, 12.7%, 4.1%, 4.0%, and 22.0% for females, respectively. Female patients with breast cancer and cervix cancer had 5-year RS of 54.6% and 33.0%.

Cancer survival rates for all sites are poor, in which that of the liver is the lowest, while that of the breast, the highest. The survivals of cancers for all sites, especially for breast, cervix, and leukaemia are seen to be lower than those of European countries except for oesophagus, pancreas and lung cancer which do not achieve improved survival both in developing and developed countries. There will be a long way to improve the total cancer survival, as well as the cancer treatment in the developing countries.