In longitudinal studies it is often of interest to investigate how a marker that is repeatedly measured in time is associated with a time to an event of interest, e.g., prostate cancer studies where longitudinal PSA level measurements are collected in conjunction with the time-to-recurrence. Joint Models for Longitudinal and Time-to-Event Data: With Applications in R provides a full treatment of random effects joint models for longitudinal and time-to-event outcomes that can be utilized to analyze such data. The content is primarily explanatory, focusing on applications of joint modeling, but sufficient mathematical details are provided to facilitate understanding of the key features of these models.All illustrations put forward can be implemented in the R programming language via the freely available package JM written by the author. All the R code used in the book is available at:
Table of Contents
Inferential Objectives in Longitudinal Studies
Organization of the Book
Analysis of Longitudinal Data
Features of Repeated Measures Data
Linear Mixed Effects Models
Dropout in Longitudinal Studies
Analysis of Time-to-Event Data
Features of Event Time Data
Relative Risk Models
Joint Models for Longitudinal and Time-to-Event Data
The Standard Joint Model
Connection with the Dropout Framework
Extensions of the Standard Joint Model
Multiple Failure Times
Latent Class Joint Models
Residuals for the Longitudinal Submodel
Residuals for the Survival Submodel
Random Effects Distribution
Prediction and Accuracy in Joint Models
Dynamic Predictions for the Survival and Longitudinal Outcomes
Effect of the Parameterization on Predictions
Prospective Accuracy Measures for Longitudinal Markers
Dimitris Rizopoulos is an Assistant Professor at the Department of Biostatistics of the Erasmus University Medical Center in the Netherlands. Dr. Rizopoulos received his M.Sc. in Statistics in 2003 from the Athens University of Economics and Business, and a Ph.D. in Biostatistics in 2008 from the Katholieke Universiteit Leuven.
Dr. Rizopoulos wrote his dissertation, as well as a number of methodological articles on various aspects of joint models for longitudinal and time-to-event data. He currently serves as an Associate Editor for Biometrics and Biostatistics, and has been a guest editor for a special issue in joint modeling techniques in Statistical Methods in Medical Research.
"Overall, the book provides a nice introduction to joint models and the R package "JM". It is well written, readable, and comprehensive. With the availability of the R package for joint models, it is expected that joint models will become increasingly popular in practice, especially in medical research. In summary, the book makes an important contribution to the research and application of joint models."
—Lang Wu, Department of Statistics, The University of British Columbia, Vancouver, Canada, in the Journal of Biopharmaceutical Statistics
"The book is well written in a matter-of-fact style that makes even unfamiliar readers understand the concept of joint models and furthermore provides them with a guide for getting started with their own analysis. The more joint model-savvy reader will, on the other hand, find inspiration for further foraging into the subject of model extensions, diagnostics, prediction, and accuracy. … a handy guide for anyone with a need to analyze survival data in the presence of a time-dependent covariate that is measured several times. As the author incorporates a longitudinal model for such a covariate into the relative risk regression modeling framework, we observe the advantage of being able to account for measurement errors within our covariate; a fortification of our research outcomes. All in all a satisfying book on joint models with a solid payout for fellow researchers."
—Maral Saadati, Biometrical Journal, 55, 2013
"This new addition to the genre is based on the JM package written by the author and has been done well. … I particularly liked the sections on numerical methods, which manage to give a useful overview of what the package is actually doing but without scaring off the mathematically reluctant. The dreaded problem of non-convergence is met head-on, with an illustration and discussion of how a little knowledge of the fitting algorithms can help to overcome such problems. This alone is worth the price of the book! … To summarize, this is a very well-crafted introduction to an active research area that I would recommend to anyone interested in getting into this field or in learning to analyze such data."
—Geoff Jones, Australian & New Zealand Journal of Statistics, 2013