Applied Survival Analysis

This textbook introduces survival analysis from standard methods for a single event to approaches for complex time-to-event outcomes. It connects statistical concepts with practical questions arising in biomedical research, using real data examples and accompanying R code to illustrate how methods are applied and results interpreted. The book is organized into three parts. Part I develops the foundations of univariate survival analysis, including nonparametric estimation and testing, Cox regression, alternative regression models, study design, and methods for truncated and interval-censored data. Part II covers multivariate and recurrent events, competing and semi-competing risks, joint analysis of longitudinal and survival data, multistate models, and composite endpoints. Part III introduces causal inference and machine learning for survival outcomes, connecting these topics to the concepts developed earlier. Throughout, the emphasis is on understanding what a method estimates, the assumptions it requires, and how to use it in practice. Exercises range from methodological derivations to data analyses in R. Designed primarily for advanced undergraduate and graduate students in statistics and biostatistics, the book also serves researchers and quantitatively trained health science professionals who work with time-to-event data.