1st Edition
Handbook of Statistical Methods for Randomized Controlled Trials
Statistical concepts provide scientific framework in experimental studies, including randomized controlled trials. In order to design, monitor, analyze and draw conclusions scientifically from such clinical trials, clinical investigators and statisticians should have a firm grasp of the requisite statistical concepts. The Handbook of Statistical Methods for Randomized Controlled Trials presents these statistical concepts in a logical sequence from beginning to end and can be used as a textbook in a course or as a reference on statistical methods for randomized controlled trials.
Part I provides a brief historical background on modern randomized controlled trials and introduces statistical concepts central to planning, monitoring and analysis of randomized controlled trials. Part II describes statistical methods for analysis of different types of outcomes and the associated statistical distributions used in testing the statistical hypotheses regarding the clinical questions. Part III describes some of the most used experimental designs for randomized controlled trials including the sample size estimation necessary in planning. Part IV describe statistical methods used in interim analysis for monitoring of efficacy and safety data. Part V describe important issues in statistical analyses such as multiple testing, subgroup analysis, competing risks and joint models for longitudinal markers and clinical outcomes. Part VI addresses selected miscellaneous topics in design and analysis including multiple assignment randomization trials, analysis of safety outcomes, non-inferiority trials, incorporating historical data, and validation of surrogate outcomes.
Part I. Introduction to Randomized, Controlled Trials
1. Introduction
KyungMann Kim
Part II. Analytic Methods for Randomized, Controlled Trials
2. Dichotomous and ordinal: chi-square and Fisher's exact tests and binary regression models
Garrett Fitzmaurice, Stuart Lipsitz
3. Continuous: t-test, Wilcoxon-test, and linear or non-linear regression models
Fang-Shu Ou
4. Time to event subject to censoring: logrank test, Kaplan-Meier estimation and Cox proportional hazards regression models
Daniel Scharfstein, Yuxin Zhu, Anastasios Tsiatis
5. Count: Poisson and negative binomial regression models
Jianguo "Tony" Sun, Xin He
6. Longitudinal: Linear and generalized linear mixed models, GEE
Myunghee Cho Paik, Soeun Kim
7. Recurrent events
Richard Cook, Yujie Zhong
8. Cross-over design
Stephen Senn
9. Factorial design
Bibhas Chakraborty, Palash Ghosh
10. Cluster randomized design
Martin Bland, Mona Kanaan
11. Randomization, stratification, and outcome-adaptive allocation
Oleksandr Sverdlov, Yevgen Ryeznik
12. Sample size estimation and power analysis: Dichotomous, ordinal, continuous and count
Keaven Anderson, Oliver Bautista
13. Sample size estimation and power analysis: Time-to-event data subject to censoring
Keaven Anderson, Oliver Bautista
14. Sample size estimation and power analysis: Longitudinal data
Sin-Ho Jung
15. Group sequential methods, triangular methods and stochastic curtailments
Michael Proschan
16. Sample size re-estimation
Tim Friede, Tobias Mütze
17. Adaptive designs
Gernot Wassmer, Franz Koenig, Martin Posch
18. Multiple testing
Jason Hsu, Yi Liu, Szu-Yu Tang
19. Subgroup analysis
Rui Wang
20. Competing risks
Haesook Kim
21. Joint models for longitudinal markers and clinical outcomes
Helene Jacqmin-Gadda, Cécile Proust-Lima, Loïc Ferrer
22. Sequential multiple assignment randomization trial (SMART) for dynamic treatment allocation
Michael Kosorok, Emily Butler
23. Safety data analysis
Amy Xia, Brenda Crowe, Jesse Berlin
24. Non-inferiority trials
Brian Wiens
25. Incorporating historical data into RCTs
Heinz Schmidli, Sandro Gsteiger, Beat Neuenschwander
26. Validation of surrogate outcomes
Geert Molenberghs
Biography
KyungMann Kim is Professor of Biostatistics and Statistics and Director of Clinical Trials Program, Department of Biostatistics and Medical Informatics, University of Wisconsin-Madison. He is a former associate editor of Biometrics and an elected Fellow of the American Statistical Association, the Society for Clinical Trials, and the American Association for Advancement of Science.
Frank Bretz is a Distinguished Quantitative Research Scientist at Novartis. He is also an Adjunct Professor at the Hannover Medical School (Germany) and the Medical University Vienna (Austria). He is a former editor-in-chief of Statistics in Biopharmaceutical Research. a Fellow of the American Statistical Association, and a recipient of the Susanne-Dahms-Medal from the German Region of the International Biometric Society.
Ying Kuen (Ken) Cheung is Professor of Biostatistics and Associate Dean for Faculty in the Mailman School of Public Health at Columbia University. He is a recipient of the IBM Faculty Award on Big Data and Analytics. He is a Fellow of the American Statistical Association and a Fellow of the New York Academy of Medicine.
Lisa Hampson is a Director in Statistical Methodology at Novartis.
"This book is the product of a large and outstanding group of editors and collaborative authors who undertook a huge effort of summarizing, in one volume, a subject spanning a wide crosssection of topics related to clinical trials. ... Throughout, many topics are illustrated with examples of recently reported trials adding to the applicability of the corresponding theory. The emphasis on sample size estimation is a very nice touch and a strong feature of the book. In some cases, authors have included code in R and SAS to assist users."
-Daniel Zelterman, in Technometrics, July 2022
