Missing Data in Clinical Studies

by Molenberghs, Geert; Kenward, Michael

• ISBN13: 9780470849811
• ISBN10: 0470849819
• Edition: 1st
• Format: Hardcover
• Copyright: 2007-04-16
• Publisher: WILEY

Summary

Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on to discuss more advanced approaches. The authors focus on practical and modeling concepts, providing an extensive set of case studies to illustrate the problems described. Provides a practical guide to the analysis of clinical trials and related studies with missing data. Examines the problems caused by missing data, enabling a complete understanding of how to overcome them. Presents conventional, simple methods to tackle these problems, before addressing more advanced approaches, including sensitivity analysis, and the MAR missingness mechanism. Illustrated throughout with real-life case studies and worked examples from clinical trials. Details the use and implementation of the necessary statistical software, primarily SAS. Missing Data in Clinical Studies has been developed through a series of courses and lectures. Its practical approach will appeal to applied statisticians and biomedical researchers, in particular those in the biopharmaceutical industry, medical and public health organisations. Graduate students of biostatistics will also find much of benefit.

Author Biography

Geert Molenberghs and Michael Kenward are the authors of Missing Data in Clinical Studies, published by Wiley.

Table of Contents

Preface

Acknowledgements

Preliminaries

Introduction

From Imbalance to the Field of Missing Data Research

Incomplete Data in Clinical Studies

MAR, MNAR, and Sensitivity Analysis

Outline of the Book

Key Examples

Introduction

The Vorozole Study

The Orthodontic Growth Data

Mastitis in Dairy Cattle

The Depression Trials

The Fluvoxamine Trial

The Toenail Data

Age-Related Macular Degeneration Trial

The Analgesic Trial

The Slovenian Public Opinion Survey

Terminology and Framework

Modelling Incompleteness

Terminology

Missing Data Frameworks

Missing Data Mechanisms

Ignorability

Pattern-Mixture Models

Classical Techniques and the Need for Modelling

A Perspective on Simple Methods

Introduction

Simple Methods

Problems with Complete Case Analysis and Last Observation Carried Forward

Using the Available Cases: a Frequentist versus a Likelihood Perspective

Intention to Treat

Concluding Remarks

Analysis of the Orthodontic Growth Data

Introduction and Models

The Original, Complete Data

Direct Likelihood

Comparison of Analyses

Example SAS Code for Multivariate Linear Models

Comparative Power under Different Covariance Structures

Concluding Remarks

Analysis of the Depression Trials

View 1: Longitudinal Analysis

Views 2a and 2b and All versus Two Treatment Arms

Missing at Random and Ignorability

The Direct Likelihood Method

Introduction

Ignorable Analyses in Practice

The Linear Mixed Model

Analysis of the Toenail Data

The Generalized Linear Mixed Model

The Depression Trials

The Analgesic Trial

The Expectation-Maximization Algorithm

Introduction

The Algorithm

Missing Information

Rate of Convergence

EM Acceleration

Calculation of Precision Estimates

A Simple Illustration

Concluding Remarks

Multiple Imputation

Introduction

The Basic Procedure

Theoretical Justification

Inference under Multiple Imputation

Efficiency

Making Proper Imputations

Some Roles for Multiple Imputation

Concluding Remarks

Weighted Estimating Equations

Introduction

Inverse Probability Weighting

Generalized Estimating Equations for Marginal Models

Weighted Generalized Estimating Equations

The Depression Trials

The Analgesic Trial

Double Robustness

Concluding Remarks

Combining GEE and MI

Introduction

Data Generation and Fitting

MI-GEE and MI-Transition

An Asymptotic Simulation Study

Concluding Remarks

Likelihood-Based Frequentist Inference

Introduction

Information and Sampling Distributions

Bivariate Normal Data

Bivariate Binary Data

Implications for Standard Software

Analysis of the Fluvoxamine Trial

The Muscatine Coronary Risk Factor Study

The Crepeau Data

Concluding Remarks

Analysis of the Age-Related Macular Degeneration Trial

Introduction

Direct Likelihood Analysis of the Continuous Outcome

Weighted Generalized Estimating Equations

Direct Likelihood Analysis of the Binary Outcome

Multiple Imputation

Concluding Remarks

Incomplete Data and SAS

Introduction

Complete Case Analysis

Last Observation Carried Forward

Direct Likelihood

Weighted Estimating Equations

Multiple Imputation

Missing Not at Random

Selection Models

Introduction

The Diggle-Kenward

Table of Contents provided by Publisher. All Rights Reserved.