Objective: Chinese cancer registration data provide information on the national cancer burden but howrepresentative they are of the real situation is uncertain. Mortality data from cancer registration and the thirdnational death survey were therefore compared to determine the accuracy of estimates in China.
Methods: Thedata were from the Cancer Registration Annual Report, 2004, China and the third National Death Survey,2004-205. Negative binomial regression was used to estimate site-specific cancer mortality rate ratios betweenthe two.
Results: The estimated cancer mortality for all sites from national cancer registration was representativefor China, especially in urban areas, but mortality was over-estimated for rural areas, with large differences insome cancer site-specific mortalities.
Conclusion: Although cancer registration data are representative at thecountry level and for urban areas, they may not reflect real cancer burden in rural areas, particularly with somecancer types. Setting up new cancer registries in non-high risk areas in rural areas should be enhanced infurther cancer surveillance plans.