LECTURE: Estimation in relative survival field

Survival analysis of long term studies is often interested in mortality due to the disease in question but faced with the problem of many deaths occurring due to other causes. Furthermore, the cause of death is often unknown or unreliable. A solution to this problem is to assume that hazard due to other causes can be described by the general population mortality, which enables us to deduct the value of the disease related hazard.
The methodology based on this assumption is referred to as relative survival, its most important field of usage is cancer registry data. Several methods of extracting and summarizing the disease-specific mortality information exist and are used both in standardized reports as well as in in-depth analyzes. Though the methods have different goals, they are often mixed-up or wrongly interpreted.
Several pitfalls exist, some of them in particular relevant when one is aiming at overly standardized reports. In this talk we lay down clearly the different
 concepts and methods of estimation and clarify the interpretation and the limitations of the different concepts, with the net survival being of primary interest. The results are illustrated using Slovene cancer registry and simulated data.