Bayesian Adaptive Methods for Clinical Trials

Scott M. Berry is the President and Senior Statistical Scientist at Berry Consultants, a statistical consulting group specializing in adaptive clinical trial design in pharmaceutical and medical device research and development.Bradley P. Carlin is Mayo Professor in Public Health and Head of the Division of Biostatistics at the University of Minnesota.J. Jack Lee is Professor of Biostatistics at the University of Texas M.D. Anderson Cancer Center.Peter Müller is a Robert R. Herring Distinguished Professor in Clinical Research in the Department of Biostatistics at the University of Texas M.D. Anderson Cancer Center.

• Statistical Approaches for Clinical Trials. Basics of Bayesian Inference. Phase I Studies. Phase II Studies. Phase III Studies. Special Topics. References. Indices.

… researchers/statisticians who work in oncology clinical trials would especially benefit from this book. Having said that, much of the book should be accessible to nonstatistician readers with interest/involvement in clinical trials across different disease areas, while even Bayesian medical statistician readers should find the book to be a helpful resource. The authors, while clearly advocating the use of Bayesian approaches, nevertheless take a very pragmatic approach to the issue. They argue for Bayesian methods that demonstrate good frequentist properties and that are practical to use. This is a refreshing change from some other books and papers that also advocate Bayesian methods, but which while theoretically interesting, are difficult to implement in practice. In summary, I found this book to be well written and interesting. It is very timely given the interest in adaptive designs, and should be a useful resource for statisticians and nonstatisticians alike interested in adaptive clinical trials from a Bayesian perspective.—Biometrics, 67, September 2011This is the first serious text on adaptive designs using the Bayesian approach. … This is the right book to get if you are interested in Bayesian methods for adaptive designs.—Michael R. Chernick, Technometrics, August 2011… a rich introduction to Bayesian clinical trial design in general. It is written in an extremely readable style and is furnished with numerous examples and a great deal of helpful supplementary code. … This work provides a good overall look at the Bayesian approach to clinical trials. It covers the theoretical framework, provides software for the many excellent examples, and even delves into the practical regulatory issues that arise with the use of the designs. The book would be a worthy addition to the practicing statistician’s library.—Journal of Statistical Software, November 2010, Volume 37This fine book represents the most recent and exciting developments in this area, and gives ample justification for the power and elegance of Bayesian trial design and analysis. … This book, based on the many years of cumulative experience of the authors, manages to deal with [ideological, bureaucratic, practical and pragmatic] difficulties. Adaptive studies are a perfect application for a Bayesian approach, and I am confident that this book will be a major contribution to the science and practice of clinical trials.—From the Foreword by David J. Spiegelhalter, MRC Biostatistics Unit, University of Cambridge, UK"This excellent book can be recommended to any biostatistician with a professional interest in clinical trials, who has mainly applied frequentist methods up to now and who wants to gain a better understanding of the Bayesian methodology. The book can also be recommended to clinical investigators and regulators who have a minimum level of formal training. The practice-oriented and real-problem-solving approach of the book shows the strengths, potentials, difficulties, and yes, also weaknesses of the Bayesian adaptive approach for drug and medical device development. The book is well written, contains a lot of real life examples, and provides web links to software code (WinBUGS and R). In short, it is a fine practical handbook."—Harald Heinzl, Zentralblatt MATH 1306