We feel privileged to present this special issue to honor David Oakes, Professor of Biostatistics and Computational Biology at University of Rochester.

Since David received his PhD degree with his thesis titled “Semi-Markov representations of some stochastic point processes” under the direction of Professor Sir David Cox at Imperial College, London, David’s contribution to the fields of Statistics and Biostatistics has been enormous and influential. His seminal work spans many areas, including Stochastic Processes and Probability, Univariate and Multivariate Censored Survival Data, Frailty and Copula Models, Mean Residual Life, and Win Ratio for Semi-Competing Risks Data. His influential paper “Bivariate Survival Models Induced by Frailties” (JASA 1989), which has been cited over 1100 times to date, introduced a new class of bivariate survival distributions that are induced by a common dependence structure (frailty) under the proportional hazards model. The monograph “Analysis of Survival Data”, co-authored with Sir David Cox (Cox and Oakes 1984), has received over 11,000 citations. At the turning point of the century, his recognition in the field led to invitations to write a Vignette on Survival Analysis for Journal of the American Statistical Association (JASA 2000) and Biometrika Centenary in Survival Analysis (Biometrika 2001). David’s contribution to medical research has been also influential and well recognized in the areas of Environmental Health, Neurology, Toxicology, and Cardiology.

David started his academic career as an Assistant Professor in the Department of Statistics at Harvard University and then as a Senior Lecturer in Occupational Health Statistics at the London School of Hygiene and Tropical Medicine. In 1983 he joined the Departments of Statistics and Biostatistics at the University of Rochester. He chaired the Department of Statistics from 1989 to 1995 and the Department of Biostatistics from 1995 to 2002. He is an elected fellow of the American Statistical Association, an elected member of the International Statistical Institute, and an elected fellow of the Institute of Mathematical Statistics. David served as an Associate Editor for Biometrika for almost three decades (from 1979 to 2007) and is currently one of the three editors for Lifetime Data Analysis.

This special issue includes 10 invited articles. Prentice reviews various existing methods for multivariate failure time regression data and applies the recently developed multivariate marginal hazard methods to the Women’s Health Initiative hormone therapy trial data. Cook, Lawless and Xie propose a novel expectation–maximization algorithm for the marker-failure-visit process to reduce bias under the joint multistate model framework. Andersen, Wandall and Perme review methods for estimating transition probabilities in multi-state models and propose regression analysis based on pseudo-observations by relaxing the Markov assumption. Rice, Johnson and Strawderman study the importance of adhering to recommended screening policies for chronic diseases such as cancer, and develop new methodology to better optimize screening policies when adherence is imperfect. Lee, Lawrence, Chen and Whitmore investigate the issue of delayed entry into observational studies and clinical trials by modeling time to event for an individual as a first hitting time of an event threshold by a latent disease process. Wang and Zhu point out that the commonly adopted age-specific risk probability for cross-sectionally sampled data with binary disease outcome yields biased estimation of the population risk and propose an approach which reassigns a portion of the observed binary outcome, 0 or 1, to the other disease category. Cui and Peng propose a flexible testing framework to properly assess either constant or dynamic covariate effects under the quantile regression setting for time-to-event data. Hougaard discusses the issue of transparency and interpretability of the treatment from multi-state modeling on two time scales, i.e., time since start of the study (running time) or time since most recent event (gap time). Sinha, Basak and Lipsitz propose three novel models for clustered time-to-event data via random effects or copula models to account for within-cluster association, from both frequentist and Bayesian perspectives. Wang, Zeng and Lin propose a semiparametric proportional hazards model by incorporating an interaction between treatment and a single index of covariates through an unknown monotone link function in the context of discovering optimal treatment regimens for survival outcomes.

When preparing for this special issue we were pleased to extend an invitation to Professor Sir David Cox to share his reflections. Not long before his recent passing, this icon of the statistical field wrote: “It is a privilege to have the chance of congratulating David Oakes for his dedication and achievements over the years. It was a great pleasure to work with him that long time ago in London, culminating in the 1984 publication of our book, Analysis of Survival Data, reprinted several times. I greatly appreciate the opportunity to send very warm good wishes to David and to congratulate him on his achievements”.

Finally, we express our utmost gratitude to contributors to this special issue and reviewers for their time and enthusiasm. We also would like to thank the Editor-in Chief of Lifetime Data Analysis, Mei-Ling Ting Lee, for giving us this opportunity to honor David and for her helpful suggestions and guidance to complete this issue.