Cancer registries will submit their data to CONCORD according to a well-defined submission protocol before the end of 2012. The data will then be cleaned, validated, and analysed in a standard way, under guidance from the steering group.
The 3 main analytical tasks are:
- The estimation of population-based life-tables that match each cancer registration population to adjust for differing background mortality. Various smoothing-techniques will be employed to standardize life-tables particularly as different regions may lack population estimates in every calendar year, or in all single year age groups up to age 99.
- The estimation of net survival using a modeling technique that minimizes biases known to occur in direct estimation techniques, such as relative survival.
- The use of data imputation techniques to handle significant proportions of missing stage data when analyzing stage-specific net survival.