Systematic Reviews in Health Research

Meta-Analysis in Context

Matthias Egger, Julian P. T. Higgins, George Davey Smith, Matthias Egger, George Davey Smith, Julian P. T. Higgins, Matthias Egger, George Davey Smith, Julian P. T. Higgins

• Publisher: BMJ Books
• Year: 2022
• Print ISBN:
  9780727914880
• eBook ISBN:
  9781119099383
• Edition: 3
• Topic: Medicine
• Subtopic: Medical Theory, Practice & Reference

Systematic Reviews in Health Research

Explore the cutting-edge of systematic reviews in healthcare

In this Third Edition of the classic Systematic Reviews textbook, now titled Systematic Reviews in Health Research, a team of distinguished researchers deliver a comprehensive and authoritative guide to the rapidly evolving area of systematic reviews and meta-analysis. The book demonstrates why systematic reviews—when conducted properly—provide the highest quality evidence on clinical and public health interventions and shows how they contribute to inference in many other contexts. The new edition reflects the broad role of systematic reviews, including:

• Twelve new chapters, covering additional study designs, methods and software, for example, on genetic association studies, prediction models, prevalence studies, network and dose-response meta-analysis
• Thorough update of 15 chapters focusing on systematic reviews of interventions
• Access to a companion website offering supplementary materials and practical exercises (www.systematic-reviews3.org)

A key text for health researchers, Systematic Reviews in Health Research is also an indispensable resource for practitioners, students, and instructors in the health sciences needing to understand research synthesis.

CHAPTER 1
Systematic Reviews in Health Research: An Introduction

Matthias Egger, Julian P.T. Higgins, and George Davey Smith

The volume of data that needs to be considered by practitioners and researchers is constantly expanding. In most areas, it has become impossible for the individual to read, critically evaluate, and synthesize the state of current knowledge, let alone keep updating this on a regular basis. Reviews have become essential tools for anybody who wants to keep up with the new evidence that is accumulating in their field of interest. Reviews are also required to identify areas where the available evidence is insufficient and further studies are needed. In 1987, Cynthia Mulrow drew attention to the poor quality of traditional reviews, pointing out that “current medical reviews do not routinely use scientific methods to identify, assess, and synthesize information” [1]. In response to this situation, methods and guidance on systematically reviewing studies were developed to produce explicitly formulated, reproducible, and up‐to‐date summaries of the effects of health care interventions. The focus was initially on randomized controlled trials (RCTs), but soon expanded to other study designs. This is illustrated by the sharp increase, since around 2005, in the number of reviews that used formal methods to synthesize evidence (Figure 1.1).

This chapter aims to clarify terminology and scope, provide some historical background, introduce the potentials, promise, and limitations of systematic reviews and meta‐analysis, and give an overview of the topics covered in this book.

1.1 SYSTEMATIC REVIEW, META‐ANALYSIS, OR EVIDENCE SYNTHESIS?

A number of terms are used to describe the process of systematically reviewing and integrating research evidence, including “systematic review,” “meta‐analysis,” “overview,” and “pooling.” In the foreword to the first edition of this book, Chalmers and Altman [2] defined systematic review as a review that was prepared using a “systematic approach to minimizing biases and random errors documented in a materials and methods section.” A systematic review may, or may not, include a meta‐analysis: “a statistical analysis of the results from independent studies, which generally aims to produce a single, typical estimate of a treatment effect” [3].

FIGURE 1.1 Number of publications concerning meta‐analysis, 1987–2020. Results from MEDLINE search using text word and medical subject heading (MESH) “meta‐analysis” and text word “systematic review.”

The distinction between systematic review and meta‐analysis is essential because it is always appropriate and desirable to review a body of data systematically. In contrast, it may sometimes be inappropriate, or even misleading, to pool results from separate studies statistically [4]. Indeed, reviewers often find it hard to resist the temptation of combining studies even when such meta‐analysis is questionable or inappropriate. According to the Royal Society [5], “evidence synthesis” describes the process of synthesizing information from a range of sources and disciplines to inform policymaking. Such an evidence synthesis typically includes several systematic reviews and meta‐analyses, or may involve the statistical synthesis of a range of different types of evidence (see also discussion of triangulation of evidence in Chapter 19).

Other types of reviews that have emerged in recent years include “scoping reviews,” “rapid reviews,” and “umbrella reviews” [6]. Scoping reviews describe or map key concepts, types of evidence, and gaps in a defined research area [7]. In rapid reviews, time is gained by using less rigorous processes to identify studies [8]. Umbrella reviews, which are meta‐reviews (or “overviews”) of systematic reviews and meta‐analyses, reflect the explosion in the number of such publications and address the fact that systematic reviews and meta‐analyses addressing the same question may reach opposite conclusions [9] (see also Chapter 7).

1.2 THE SCOPE OF META‐ANALYSIS

An important distinction can be made between meta‐analysis of RCTs and meta‐analysis of observational studies of interventions (Chapter 15) or etiology (Chapter 19). Consider a set of high‐quality trials that examined the same intervention in comparable patient populations: each trial should provide an unbiased estimate of the same underlying treatment effect. The variability observed between the trials can be attributed to random variation, and meta‐analysis should provide an equally unbiased estimate of the treatment effect with increased precision. A fundamentally different situation typically arises in observational epidemiological studies, for example case–control studies, cross‐sectional studies, or cohort studies. Due to confounding and selection bias, these studies may produce estimates of causal associations that deviate from the underlying causal associations beyond what can be attributed to chance. Combining a set of observational studies will thus often provide spuriously precise, biased estimates of causal associations. The thorough consideration of heterogeneity between observational study results, particularly of possible confounding and bias, will generally provide more insights than the mechanical calculation of an overall measure of effect.

The fundamental difference between observational studies and RCTs does not mean that the latt...

Table of contents

• Cover
• Table of Contents
• Title Page
• Copyright Page
• Preface
• Tribute
• List of Contributors
• About the Companion Website
• CHAPTER 1: Systematic Reviews in Health Research
• PART I: PRINCIPLES AND PROCEDURES
• PART II: META‐ANALYSIS
• PART III: SPECIFIC STUDY DESIGNS
• PART IV: COCHRANE AND GUIDELINE DEVELOPMENT
• PART V: OUTLOOK
• PART VI: SOFTWARE
• Index
• End User License Agreement