Modelling Recurrent Event Data with Application to Cancer Research

The aim of this book is to show how to analyze survival data with the presence of recurrent events applied to cancer settings. Throughout, the emphasis is on presenting analysis of real data. Many of the models discussed are those widely used in this area. In addition, a new model specially designed for analyzing cancer data is presented. Modern techniques such as penalized likelihood approach, nonparametric smoothig and bootstrapping are developed and used when appropriate. The author, jointly with other colleagues, has written three R packages, freely available at CRAN (http:://www.r-project.org) designed to analyze recurrent event data: gcmrec, survrec and frailtypack. These packages also contain the real data sets analyzed in this book. Each chapter of this book ends with an illustration of how to use these packages to fit models. These analyses should help biostatisticians, clinicians or medical doctors to analyze their own data arising form studies where the main aim is to describe those clinical factors that are associated with the time until a new event occurs taking into account the repeated nature of the data.