Meta-Analysis in Stata: An Updated Collection from the Stata Journal

Meta-analysis allows researchers to combine results of several studies into a unified analysis that provides an overall estimate of the effect of interest and to quantify the uncertainty of that estimate. Stata has some of the best statistical tools available for doing meta-analysis. The unusual thing about these tools is that none of them are part of official Stata. They are all created by and documented by experts in the broader research community who also happen to be proficient Stata developers.

 

Editors Tom Palmer and Jonathan Sterne show how each of the articles in this collection relates to others and how each fits in the overall literature of meta-analysis. For the first edition, Sterne convinced over half the authors to update their software and articles for the collection. In this new edition, Palmer and Sterne have substantially expanded the scope of the collection to cover in more depth many contemporary advances that will help keep the reader up to date.

 

The second edition retains its original topic-specific sections devoted to the fundamentals of meta-analysis: fitting models, meta-regression, and graphical and analytic tools for detecting bias. It also retains a section devoted to advanced methods. Readers of the first edition will find new articles in these sections, in particular ones that take advantage of major changes that occurred in Stata since the first edition, such as the introduction of the gsem command.

 

This edition also adds three new topic-specific sections for multivariate or multiple outcomes meta-analysis, individual participant data (IPD) meta-analysis, and network meta-analysis. The addition of these sections gives readers access to new commands that address recent methodological developments in the field.

 

The new edition adds 11 articles to the original collection of 16 articles. The articles cover topics ranging from standard and cumulative meta-analysis and forest plots to contour-enhanced funnel plots and nonparametric analysis of publication bias. In their articles, the authors present conceptual overviews of the techniques, thorough explanations, and detailed descriptions and syntax of new commands. They also provide examples using real-world data. In short, this collection is a complete introduction and reference for performing meta-analyses in Stata.

Install the software
1. META-ANALYSIS IN STATA: METAN, METAAN, METACUM AND METAP
metan—a command for meta-analysis in Stata
M. J. Bradburn, J. J. Deeks, and D. G. Altman
metan: fixed- and random-effects meta-analysis
R. J. Harris, M. J. Bradburn, J. J. Deeks, R. M. Harbord, D. G. Altman, and J. A. C. Sterne
metaan: Random-effects meta-analysis
E. Kontopantelis and D. Reeves
Cumulative meta-analysis
J. A. C. Sterne
Meta-analysis of p-values
A. Tobias

 

2. META-REGRESSION:METAREG
Meta-regression in Stata
R. M. Harbord and J. P. T. Higgins
Meta-analysis regression
S. Sharp

 

3. INVESTIGATING BIAS IN META-ANALYSIS: METAFUNNEL, CONFUNNEL, METABIAS, METATRIM, AND EXTFUNNEL
Funnel plots in meta-analysis
J. A. C. Sterne and R. M. Harbord
Contour-enhanced funnel plots for meta-analysis
T. M. Palmer, J. L. Peters, A. J. Sutton, and S. G. Moreno
Updated tests for small-study effects in meta-analyses
R. M. Harbord, R. J. Harris, and J. A. C. Sterne
Tests for publication bias in meta-analysis
T. J. Steichen
Tests for publication bias in meta-analysis
T. J. Steichen, M. Egger, and J. A. C. Sterne
Nonparametric trim and fill analysis of publication bias in meta-analysis
T. J. Steichen
Graphical augmentations to the funnel plot to assess the impact of a new study on an existing meta-analysis
M. J. Crowther, D. Langan, and A. J. Sutton

 

4. MULTICVARIATE META-ANALYSIS: METANDI, MVMETA
metandi: Meta-analysis of diagnostic accuracy using hierarchical logistic regression
R. M. Harbord and P. Whiting
Multivariate random-effects meta-analysis
I. R. White
Multivariate random-effects meta-regression: Updates to mvmeta
I. R. White

 

5. INDIVIDUAL PATIENT DATA META-ANALYSIS: IPDFOREST AND IPDMETAN
A short guide and a forest plot command (ipdforest) for one-stage meta-analysis
E. Kontopantelis and D. Reeves
Two-stage individual participant data meta-analysis and generalized forest plots
D. J. Fisher

 

6. NETWORK META-ANALYSIS: INDIRECT NETWORK PACKAGE, NETWORK_GRAPHS PACKAGE
Indirect treatment comparison
B. Miladinovic, I. Hozo, A. Chaimani, and B. Djulbegovic
Network meta-analysis
I. R. White
Visualizing assumptions and results in network meta-analysis: The network graphs package
A. Chaimani and G. Salanti

 

7. ADVANCED METHODS: GLST, METAMISS, SEM, GSEM, METACUMBOUNDS, METASIM, METAPOW AND METAPOWPLOT
Generalized least squares for trend estimation of summarized dose–response data
N. Orsini, R. Bellocco, and S. Greenland
Meta-analysis with missing data
I. R. White and J. P. T. Higgins
Fitting fixed- and random-effects meta-analysis models using structural equation modeling with the sem and gsem commands
T. M. Palmer and J. A. C. Sterne
Trial sequential boundaries for cumulative meta-analyses
B. Miladinovic, I. Hozo, and B. Djulbegovic
Simulation-based sample-size calculation
M. J. Crowther, S. R. Hinchliffe, A. Donald, and A. J. Sutton
Appendix
Author: Tom M. Palmer and Jonathan A. C. Sterne
Edition: Second Edition
ISBN-13: 978-1-59718-147-1
©Copyright: 2016
Versione e-Book disponibile

Meta-analysis allows researchers to combine results of several studies into a unified analysis that provides an overall estimate of the effect of interest and to quantify the uncertainty of that estimate. Stata has some of the best statistical tools available for doing meta-analysis. The unusual thing about these tools is that none of them are part of official Stata. They are all created by and documented by experts in the broader research community who also happen to be proficient Stata developers.