Meta-analysis is the application of statistics to combine results from multiple studies and draw appropriate inferences. Its use and importance have exploded over the last 25 years as the need for a robust evidence base has become clear in many scientific areas, including medicine and health, social sciences, education, psychology, ecology, and economics.
Recent years have seen an explosion of methods for handling complexities in meta-analysis, including explained and unexplained heterogeneity between studies, publication bias, and sparse data. At the same time, meta-analysis has been extended beyond simple two-group comparisons of continuous and binary outcomes to comparing and ranking the outcomes from multiple groups, to complex observational studies, to assessing heterogeneity of effects, and to survival and multivariate outcomes. Many of these methods are statistically complex and are tailored to specific types of data.
- Rigorous coverage of the full range of current statistical methodology used in meta-analysis
- Comprehensive, coherent, and unified overview of the statistical foundations behind meta-analysis
- Detailed description of the primary methods for both univariate and multivariate data
- Computer code to reproduce examples in chapters
- Thorough review of the literature with thousands of references
- Applications to specific types of biomedical and social science data
This book is for a broad audience of graduate students, researchers, and practitioners interested in the theory and application of statistical methods for meta-analysis. It is written at the level of graduate courses in statistics, but will be of interest to and readable for quantitative scientists from a range of disciplines. The book can be used as a graduate level textbook, as a general reference for methods, or as an introduction to specialized topics using state-of-the art methods.
Table of Contents
1. Introduction to Systematic Review and Meta-Analysis
Christopher H. Schmid, Ian R. White, Theo Stijnen
2. General Themes in Meta-Analysis
Christopher H. Schmid, Theo Stijnen, Ian R. White
3. Choice of Effect Measure and Issues in Extracting Outcome Data
Ian R. White, Christopher H. Schmid Schmid, Theo Stijnen
4. Analysis of Univariate Study-Level Summary Data Using Normal Models
Theo Stijnen, Ian R. White, Christopher H. Schmid
5. Exact Likelihood Methods for Group-Based Summaries
Theo Stijnen, Christopher H Schmid, Martin Law, Dan Jackson, Ian R. White
6. Bayesian Methods for Meta-Analysis
Christopher H. Schmid, Bradley P. Carlin, Nicky J Welton
Julian PT Higgins, José A López-López, Ariel M Aloe
8. Individual Participant Data Meta-Analysis
Lesley Stewart, Mark Simmonds
9. Multivariate Meta-Analysis
Dan Jackson, Ian R. White, Richard Riley
10. Network Meta-AAnalysis
Adriani Nikolakopoulou, Ian R. White, Georgia Salanti
11. Model Checking in Meta-Analysis
12. Handling Internal and External Biases: Quality and Relevance of Studies
Rebecca M. Turner, Nicky J. Welton, Hayley E. Jones, Jelena Savović
13. Publication and Outcome Reporting Bias
Arielle Marks-Anglin, Rui Duan, Yong Chen, Orestis Panagiotou, Christopher H. Schmid
14. Control Risk Regression
Annamaria Guolo, Christopher H. Schmid, Theo Stijnen
15. Multivariate Meta-Analysis of Survival Proportions
16. Meta-Analysis of Correlations, Correlation Matrices, and Their Functions
Betsy Jane Becker, Ariel M. Aloe, Mike W.-L. Cheung
17. The Meta-Analysis of Genetic Studies
Cosetta Minelli, John Thompson
18. Meta-Analysis of Dose-Response Relationships
Nicola Orsini, Donna Spiegelman
19. Meta-Analysis of Diagnostic Tests
Yulun Liu, Xiaoye Ma, Yong Chen, Theo Stijnen, Haitao Chu
20. Meta-Analytic Approach to Evaluation of Surrogate Endpoints
Tomasz Burzykowski, Marc Buyse, Geert Molenberghs, Ariel Alonso, Wim Van der Elst, Ziv Shkedy
21. Meta-Analysis of Epidemiological Data, with a Focus on Individual Participant Data
Angela Wood, Stephen Kaptoge, Mike Sweeting, Clare Oliver-Williams
22. Meta-Analysis of Prediction Models
Ewout Steyerbeg, Daan Nieboer, Thomas Debray, Hans van Houwelingen
23. Using Meta-Analysis to Plan Further Research
Claire Rothery, Susan Griffin, Hendrik Koffijberg, Karl Claxton
Christopher H. Schmid is Professor of Biostatistics at Brown University. He received his BA in Mathematics from Haverford College in 1983 and his PhD in Statistics from Harvard University in 1991. In 1991, he joined the Institute for Clinical Research and Health Policy Studies at Tufts Medical Center and joined the medical faculty at Tufts University in 1992. He became the director of the Biostatistics Research Center in 2006 and Associate Director of the Tufts Clinical and Translational Research training program in 2009. In 2012, he moved to Brown University to co-found the Center for Evidence Synthesis in Health. In 2016, he became Director of the Clinical Study Design, Epidemiology and Biostatistics Core of the Rhode Island Center to Advance Translational Science and in 2018 became Chair of Biostatistics in the School of Public Health.
Dr. Schmid has a long record of collaborative research and training activities in many different clinical and public health research areas. His research focuses on Bayesian methods for meta-analysis, including networks of treatments and N-of-1 designs, as well as open-source software tools, as well as methods for developing and assessing predictive models using data from multiple databases, e.g., the current standard biomarker prediction tool for GFR, glomerular filtration rate. He is the author of nearly 300 publications, including coauthored consensus CONSORT reporting guidelines for N-of-1 trials and single-case designs, and PRISMA guidelines extensions for meta-analysis of individual participant studies and for network meta-analyses as well as the Institute of Medicine report that established US standards for systematic reviews.
Dr. Schmid is an elected member of the Society for Research Synthesis Methodology and co-founding editor of its journal, Research Synthesis Methods. He is a Fellow of the American Statistical Association and long-time statistical editor of the American Journal of Kidney Diseases.
Ian White is Professor of Statistical Methods for Medicine at the Medical Research Council Clinical Trials Unit at University College London, UK. He originally studied mathematics at Cambridge University, and his first career was as a teacher of mathematics in The Gambia, Cambridge and London. He obtained his MSc in statistics from University College London, where he subsequently worked in the Department of Epidemiology and Public Health. He was then Senior Lecturer in the Medical Statistics Unit at the London School of Hygiene and Tropical Medicine and for 16 years programme leader at the Medical Research Council Biostatistics Unit in Cambridge. He received his PhD by publication in 2011.
His research interests are in statistical methods for the design and analysis of clinical trials, observational studies and meta-analyses. He is particularly interested in developing methods for handling missing data, correcting for departures from randomised treatment, novel trial designs, simulation studies, and network meta-analysis. He runs courses on various topics and has written a range of Stata software.
Theo Stijnen is Emeritus Professor of Medical Statistics at the Leiden University Medical Center, The Netherlands. He obtained his MSc in mathematics at Leiden University in 1973 and received his PhD in mathematical statistics at the University of Utrecht in 1980. Then he decided to leave mathematical statistics and to specialise in applied medical statistics, a choice he has never regretted. In 1981 he was appointed assistant professor of medical statistics at the Leiden University Medical Faculty. In 1987 he became associate professor of biostatistics at the Erasmus University Medical Center in Rotterdam, where he was appointed full professor in 1998. In 2007 he returned to Leiden again to become the head of the Department of Medical Statistics and Bioinformatics, which was recently renamed the Department of Biomedical Data Sciences. He has a broad experience in teaching statistics to various audiences and his teaching specialties include mixed modelling, survival analysis, epidemiological modelling and meta-analysis. In 2009 he was a co-founder of the MSc program Statistical Science for the Life and Behavioral Sciences, the first MSc program in this field in The Netherlands. He has extensive experience in statistical consultancy for medical researchers, resulting in more than 400 co-authorships in the medical scientific literature, of which about 25 on medical meta-analyses. His biostatistical research interests include clinical trials methodology, epidemiological methods, mixed modelling and meta-analysis. He is (co-)author of over 70 methodological articles, of which about 25 on meta-analysis. He retired on December 14, 2016. He now works part-time as an independent biostatistical consultant and continues doing research.
"Handbook of Meta-Analysis is a most laudable and detailed treatise on meta-analysis. It successfully covers – with gusto and substance – the full range of statistical methodology used in meta-analysis in a statistically rigorous and up-to-date manner, exuding a good balance of theory and applications (with real data and software syntax provided). It provides a comprehensive, coherent, and unified overview of the statistical foundations behind meta-analysis. Crafted by experts on the topic, each chapter is written with lucidity and surgical precision. It is elegantly organized, encyclopedic in breadth and depth, and fluent in exposition on the multidimensional role of meta-analysis: core material (background, systematic review process, data extraction, study-level results, frequent and Bayesian approaches); key extensions (meta-regression, individual data, multivariate meta-analysis, network meta-analysis, model checking, bias); and advances in particular fields of biomedical and social research (control risk regression, survival data, correlation matrices, genetic data, dose-response relationships, diagnostic tests, surrogate endpoints, complex observational data, prognostic models). It is a tour de force, a premier, and an indispensable reference that is highly recommended – and a must for serious researchers and practitioners engaged in meta-analysis. This state-of-the-science handbook is destined to be a classic."
- Joseph C. Cappelleri, PhD, MPH, MS, Executive Director of Biostatistics, Pfizer Inc
"For many researchers in social, medical, life and environmental sciences, it has become an essential part of their activities to synthesize evidence from the body of relevant research. The Handbook of Meta-analysis provides the most comprehensive and up-to-date coverage of the quantitative part of evidence synthesis, i.e., meta-analysis. Therefore, this handbook is a must-have for all researchers who wish to unlock and understand the power and potential of meta-analysis, but also for those who have already found and benefited from it. The authors of this edited volume are an interdisciplinary all-star team of statisticians and methodologists; probably, each of them could have written a textbook on meta-analysis. Here, they introduce both basics and advanced techniques that they have been leading to develop over their career. For many statisticians, a meta-analysis may be just one type of linear models (Chapters 1-11), yet, as this book demonstrates, meta-analyses can come in diverse forms and serve different purposes (see Chapters 14-22). Further, there are specific statistical issues meta-analysis needs to grapple with, such as publication bias (Chapters 12-13). The book ends with a chapter on how to use meta-analysis to plan our future work (Chapter 23) – what all scientists should be doing to reduce research waste and to accelerate scientific progress."
- Shinichi Nakagawa, Professor of Evolutionary Biology and Synthesis, University of New South Wales, Sydney, Australia
"This is an important book on an important subject, covering both theory and application, and it should be valuable to a wide range of readers in statistics and applied fields.""...The Handbook of Meta-Analyses is a “must have” resource for: 1) statisticians, other professionals, and students conducting statistical research in meta-analysis; 2) practitioners conducting meta-analyses as part of systematic reviews or otherwise; and 3) educators and students who want to either start, or continue, to learn more about meta-analysis. The breadth and depth of up-to-date coverage of meta-analysis methods, wide range of areas of application, and examples, including online software code and data, is impressive. The contents are weighted towards frequentist strategies, but Bayesian strategies are highlighted in the core materials and revisited elsewhere. The Handbook is a pleasure to read. The editors and other co-authors guide the reader in a cohesive, unified fashion, from the foundational core material through increasingly sophisticated and wider ranging methods and applications. Their tone is conversational, with forwards-and-backwards sign-posting which integrates the contents in a tutorial-like fashion. Statistical notation is used with purpose, without excess, while maintaining statistical rigor in content. An abundance of graphs, figures, and tables reinforce the statistical concepts and methods, and visualize the examples. Both novice and more experienced readers will benefit...The Handbook of Meta-Analysis is a significant contribution which provides a palpable opportunity to improve future decision-making and policy setting."
- Andrew Gelman, Columbia University
- Thomas Bradstreet, Appeared in the Journal of Biopharmaceutical Statistics