Survival statistics can be produced by population-based cancer reg- istries that follow up their cases, either actively or passively. Although survival analysis of data from population-based registries cannot evaluate specific treatments (this can be done only in clinical randomized trials), it provides a useful evaluation of cancer care in the area covered by the registry, since all cancer cases will be included regardless of the type of treatment they may have received. The methods used in survival analyses are those discussed in Chapter 12. The first requirement for the application of these methods is a clear and well defined case definition. This should clearly specify the site of the cancer.
