ABSTRACT
Analysing time until death is the archetypical example of a survival analysis, hence the name. In practice, however, time-to-event endpoints are often composites. Common examples in oncology include progression-free survival, disease-free survival and relapse-free survival (Mathoulin-Pelissier et al., 2008). Relapse-free survival is the time until relapse or death, whatever occurs first. Death can also be considered to be a composite, if we distinguish between disease-specific death, say, and death from other causes (Cuzick, 2008). In other words, the “survival time” is typically the time until the first of a number of possible events. It is often tacitly understood that sooner or later every individual experiences at least one of these events. Then the distribution of the survival time will approach one as time progresses. For instance, the cumulative event probabilities of disease-specific death and of death from other causes, respectively, add up to the cumulative all-cause mortality distribution. The latter will ultimately approach one.
