Applied Meta-Analysis with R and Stata

Ding-Geng (Din) Chen, Karl E. Peace

• Publisher: Chapman and Hall/CRC
• Year: 2021
• Print ISBN: 9780367183837
• eBook ISBN: 9780429590238
• Edition: 2
• Topic: Computer Science
• Subtopic: Data Mining
• Index: Computer Science

Review of the First Edition:

The authors strive to reduce theory to a minimum, which makes it a self-learning text that is comprehensible for biologists, physicians, etc. who lack an advanced mathematics background. Unlike in many other textbooks, R is not introduced with meaningless toy examples; instead the reader is taken by the hand and shown around some analyses, graphics, and simulations directly relating to meta-analysis… A useful hands-on guide for practitioners who want to familiarize themselves with the fundamentals of meta-analysis and get started without having to plough through theorems and proofs.

— Journal of Applied Statistics

Statistical Meta-Analysis with R and Stata, Second Edition provides a thorough presentation of statistical meta-analyses (MA) with step-by-step implementations using R/Stata. The authors develop analysis step by step using appropriate R/Stata functions, which enables readers to gain an understanding of meta-analysis methods and R/Stata implementation so that they can use these two popular software packages to analyze their own meta-data. Each chapter gives examples of real studies compiled from the literature. After presenting the data and necessary background for understanding the applications, various methods for analyzing meta-data are introduced. The authors then develop analysis code using the appropriate R/Stata packages and functions.

What's New in the Second Edition:

• 
  
  

• Adds Stata programs along with the R programs for meta-analysis

• 
  
  

• Updates all the statistical meta-analyses with R/Stata programs

• 
  
  

• Covers fixed-effects and random-effects MA, meta-regression, MA with rare-event, and MA-IPD vs MA-SS

• 
  
  

• Adds five new chapters on multivariate MA, publication bias, missing data in MA, MA in evaluating diagnostic accuracy, and network MA

Suitable as a graduate-level text for a meta-data analysis course, the book is also a valuable reference for practitioners and biostatisticians (even those with little or no experience in using R or Stata) in public health, medical research, governmental agencies, and the pharmaceutical industry.

1

Introduction to R and Stata for Meta-Analysis

In this chapter, we begin with a basic introduction to the R and Stata for meta-analysis. We focus this introduction on R since R will be the main software in this book and the Stata program will be included at the end of each chapter for Stata users to practise the meta-analysis in each chapter.

We assume that readers have no experience in R. So this introduction to R will be from the very basic R system on where to get R and how to install R and upgrade R packages. We then proceed to show how easy it is to use R for data management as well as to simulate and analyze data from multi-center studies with a brief introduction to meta-analysis. We conclude the chapter with a brief summary and some recommendations for further reading and references.

The main goal for this chapter is to introduce R to readers. For readers who already know and have familiarity with R, you can skip this chapter and go directly to any of the remaining chapters.

1.1 Introduction to R for Meta-Analysis

1.1.1 What is R?

R was initially created by Ihaka and Gentleman (1996) from University of Auckland, New Zealand. Since its creation in the middle of 1990s, R has quickly become a popular programming language and an environment for statistical computing. The continuing development of R is carried out by a core team from different institutions around the world.

To obtain an introduction to R, go to the official home page of the R project at http://www.R-project.org and click “What is R?”:

R is a language and environment for statistical computing and graphics. It is a GNU project which is similar to the S language and environment which was developed at Bell Laboratories (formerly AT&T, now Lucent Technologies) by John Chambers and colleagues. R can be considered as a different implementation of S. There are some important differences, but much code written for S runs unaltered under R.

R provides a wide variety of statistical (linear and nonlinear modeling, classical statistical tests, time-series analysis, classification, clustering, ...) and graphical techniques, and is highly extensible. The S language is often the vehicle of choice for research in statistical methodology and R provides an Open Source route to participation in that activity.

One of R’s strengths is the ease with which well-designed publication-quality plots can be produced, including mathematical symbols and formulae wherever needed. Great care has been taken over the defaults for the minor design choices in graphics, but the user retains full control.

R is available as Free Software under the terms of the Free Software Foundation’s GNU General Public License in source code form. It compiles and runs on a wide variety of UNIX platforms and similar systems (including FreeBSD and Linux), Windows, and MacOS.

To some users, “free” software may be a “negative” word for software that is difficult to use, has lower quality, or utilizes procedures that have not been validated or verified, etc. However to other users, “free” software means software from an open source that not only allows use of the software but also permits modifications to handle a variety of applications. This latter description is the fundamental principle for R system.

We now proceed to the steps for installing and using R.

1.1.2 Steps on Installing R and Updating R Packages

In general, the R system consists of two parts. One is the so-called R base system for the core R language and associated fundamental libraries. The other consists of user contributed packages that are more specialized applications. Both the base system and the packages may be obtained from the Comprehensive R Archive Network (CRAN) from the weblink:

http://CRAN.r-project.org

Installation of R system is described in the following sections.

1.1.2.1 First Step: Install R Base System

The base system can be downloaded from

http://CRAN.r-project.org

for different platforms of “Linux,” “MacOS X,” and “Windows.” In this book, we illustrate the use of R for “Windows.” “Windows” users can download the latest version of R using the link:

http://CRAN.r-project.org/bin/windows/base/release.htm

(At the writing of this book, version R 4.0.0 is available.). To download and install R to your computer simply follow the instructions from the installer to install R to the “Program Files” subdirectory in your C. You are ready to use R for statistical computing and data analysis.

Note to L^AT_EX and R/Sweave users (in fact, this book is written using R/Sweave): L^AT_EX will complain about the extra space in the path as in “Program Files.” Therefore if you want to use R along with L^AT_EX, you need to make a subdirectory without space in the path to install ...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Preface for the Second Edition
• Preface for the First Edition
• Authors
• List of Figures
• List of Tables
• 1 Introduction to R and Stata for Meta-Analysis
• 2 Research Protocol for Meta-Analyses
• 3 Fixed Effects and Random Effects in Meta-Analysis
• 4 Meta-Analysis with Binary Data
• 5 Meta-Analysis for Continuous Data
• 6 Heterogeneity in Meta-Analysis
• 7 Meta-Regression
• 8 Multivariate Meta-Analysis
• 9 Publication Bias in Meta-Analysis
• 10 Strategies to Handle Missing Data in Meta-Analysis
• 11 Meta-Analysis for Evaluating Diagnostic Accuracy
• 12 Network Meta-Analysis
• 13 Meta-Analysis for Rare Events
• 14 Meta-Analyses with Individual Patient-Level Data versus Summary Statistics
• 15 Other R/Stata Packages for Meta-Analysis
• Bibliography
• Index