Abstract

OBJECTIVES
A method of estimation using 8 population-based cancer registries databases in Korea (KRCR DB) has been introduced as another strategy for validly estimating the national cancer incidence (NCI) in Korea. The purpose of this study was to evaluate the validity of this method with using the 7 KRCR DBs, excluding Seoul covering 21% of the total population of Korea. METHODS: We designed the study method (NCSE_7) as same as the estimating method with using 8 KRCR DBs (NCSE_8) in order to ensure maximal comparability. We defined the expected number of cancer cases in each registry as the number of observed cases and then we added the weighted observed cases according to gender, age and the proportion of the population covered by each registry for the population of the seven regions and the population of all areas, with excluding these seven regions. From the expected number of total cancer incidents, the estimated NCI was calculated by dividing the expected number of cancer cases by the number of the total population. The standard error (SE) of the estimated incidence was also taken from the expected number of total cancer incidents. RESULTS: Compared with the results of the NCSE_8, the overall age-standardized rates (ASR) in men and women became over-estimated and under-estimated, respectively. Primary sites that showed statistically significant differences were the colo-rectum, prostate, breast and thyroid. The index of death certificate only (DCO)and microscopically verified (MV)% indicating levels of data quality were decreased, especially for the brain in DCO% and kidney in the MV%. CONCLUSIONS: The database of Seoul regional cancer registry has a key role for the method to estimate the valid nationwide cancer statistics in Korea with using the population-based cancer registries' databases.

• Keywords: NeoplasmsIncidenceRegistriesEstimation technique