Repeated Measures Design with Generalized Linear Mixed Models for Randomized Controlled Trials

Toshiro Tango

• Publisher: Chapman and Hall/CRC
• Year: 2017
• Print ISBN: 9781498747899
• eBook ISBN: 9781351648141
• Edition: 1
• Topic: Mathematics
• Subtopic: Probability & Statistics
• Index: Mathematics

Repeated Measures Design with Generalized Linear Mixed Models for Randomized Controlled Trials is the first book focused on the application of generalized linear mixed models and its related models in the statistical design and analysis of repeated measures from randomized controlled trials. The author introduces a new repeated measures design called S: T design combined with mixed models as a practical and useful framework of parallel group RCT design because of easy handling of missing data and sample size reduction. The book emphasizes practical, rather than theoretical, aspects of statistical analyses and the interpretation of results. It includes chapters in which the author describes some old-fashioned analysis designs that have been in the literature and compares the results with those obtained from the corresponding mixed models.

The book will be of interest to biostatisticians, researchers, and graduate students in the medical and health sciences who are involved in clinical trials.

Author Website: Data sets and programs used in the book are available at http://www.medstat.jp/downloadrepeatedcrc.html

1

Introduction

In clinical medicine, drugs are usually administered to control some response variable within a specified range reflecting the patient's disease state directly or indirectly. In most randomized controlled trials (RCTs) and randomized animal experiments, to evaluate the efficacy of a new treatment or the toxicity of a new treatment, the primary response variable on the same subject (e.g., a patient, an animal, a laboratory sample) is scheduled to be measured over a period of time, consisting of a baseline period before randomization, treatment period after randomization and, if necessary, a follow-up period after the end of treatment period. These subject-specific repeated measures or subject-specific response profiles are analyzed for assessing difference in changes from baseline among treatment groups, leading to a treatment effect.

1.1 Repeated measures design

Let y_ij denote the primary response variable for the ith subject at the jth time t_ij. An ordinary situation of repeated measures design adopted in RCTs or animal experiments is when the primary response variable is measured once at baseline period (before randomization) and T times during the treatment period (after randomization) where measurements are scheduled to be made at the same times for all subjects t_ij = t_j in the sampling design. We call this design the Basic 1 : T repeated measures design throughout the book, where the response profile vector y_i for the ith subject is expressed as

$y_i={\left(\underset{\text{baseline}\text{data}}{\underbrace{y_{i0}}},\underset{\text{data}\text{after}\text{randomization}}{\underbrace{y_{i1},\dots ,y_{iT}}}\right)}^t.$

(1.1)

In exploratory trials in the early phases of drug development, statistical analyses of interest will be to estimate the time-dependent mean profile for each treatment group and to test whether there is any treatment-by-time interaction. In confirmatory trials in the later phases of drug development, on the other hand, we need a simple and clinically mea...

Table of contents

• Cover
• Title Page
• Copyright Page
• Table of Contents
• Preface
• 1 Introduction
• 2 Naive analysis of animal experiment data
• 3 Analysis of variance models
• 4 From ANOVA models to mixed-effects repeated measures models
• 5 Illustration of the mixed-effects models
• 6 Likelihood-based ignorable analysis for missing data
• 7 Mixed-effects normal linear regression models
• 8 Mixed-effects logistic regression models
• 9 Mixed-effects Poisson regression models
• 10 Bayesian approach to generalized linear mixed models
• 11 Latent profile models: Classification of individual response profiles
• 12 Applications to other trial designs
• Appendix A Sample size
• Appendix B Generalized linear mixed models
• Bibliography
• Index