ABSTRACT
Multiple event data arises when individuals experience repeated occurrences of the same type of event, or events of different types. The sequence of events is termed an event history. Multiple event and event history data can be analysed using an extension of Cox regression modelling to the situation where more than one event type may occur. This extension involves a probability model known as a counting process and so this chapter begins with an introduction to the relevant theory. It is shown how survival data can be expressed in counting process format, used in modelling time-dependent variables. Models for the recurrence of the same type of event are then presented, and it is shown how they can be fitted once the recurrent event data have been expressed in counting process format. The occurrence of non-fatal events during the follow-up of an individual in a survival study may help in the understanding of the underlying mortality process and can be investigated using multistate models. These are described and illustrated in the final section of this chapter.
