A Network Meta-Analysis Toolkit

Available Online Material & Software for Network Meta-Analysis

mvmeta command – performing NMA in STATA

-          Source: http://www.mrc-bsu.cam.ac.uk/software/stata-software

The mvmeta command in STATA employs a recent approach to network meta-analysis that handles the different treatment comparisons appeared in studies as different outcomes. The command can perform fixed and random effects network meta-analysis assuming either a common or different between-study variances across comparisons. Both consistency and inconsistency models (the ‘design-by-treatment model’ or ‘Lu & Ades model’) have been implemented as well as network meta-regression models that can incorporate covariates. The command contains also an option that enables the estimation of ranking probabilities.

See also:

White IR. Multivariate random-effects meta-regression: Updates to mvmeta.The STATA Journal2011, 11: 255-270 

White IR, Barrett JK, Jackson D, Higgins JPT. Consistency and inconsistency in network meta-analysis: model estimation using multivariate meta-regression.Res Synth Meth2012, 3: 111-125

Higgins JPT, Jackson D, Barrett JK, Lu G, Ades AE, White IR. Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies.Res Synth Meth2012, 3: 98-110

Lu G, Ades AE. Assessing evidence inconsistency in mixed treatment comparisons.J Amer Statist Assoc2006, 101: 447-459

Netmeta package – performing NMA in R

-          Source: http://cran.r-project.org/web/packages/netmeta/index.html

The netmeta package in R is based on a novel approach for network meta-analysis that follows the graph-theoretical methodology. This method exploits the analogy between treatment networks and electrical networks to construct the network meta-analysis model accounting for the correlated treatment effects in multi-arm trials. Fixed and random effects models have been implemented in netmeta; the latter is constructed under the assumption of a common heterogeneity across all comparisons. Additional outputs of the package are Q-statistics for heterogeneity and inconsistency, forest plots of the pooled treatments effects versus a common reference treatment and network diagrams. Also, the ‘netheat’ plot (developed by Krahn et al. 2013) has been implemented in the package; this is a graph that helps to identify pairwise comparisons that might be potential sources of important inconsistency in the network.

See also:

Rucker G, Schwarzer G. Package ‘netmeta’: Network meta-analysis with R. The R Project website: http://cran.r-project.org/web/packages/netmeta/netmeta.pdf

Rucker G. Network meta-analysis, electrical networks and graph theory. Res Synth Meth2012, 3: 112-124

Krahn U, Binder H, Konig J. A graphical tool for locating inconsistency in network meta-analyses. BMC Med Res Methodol2013, 13: 35

Lu G, Welton NJ, Higgins JPT, White IR, Ades AE. Linear inference for mixed treatment comparison meta-analysis: A two-stage approach.Res Synth Meth 2011, 2: 43-60

Konig J, Krahn U, Binder H. Visualizing the flow of evidence in network meta-analysis and characterizing mixed treatment comparisons. Stat Med 2013, 32(30): 5414-5429

Indirect command – performing indirect comparisons in STATA

-          Source: http://www.stata-journal.com/article.html?article=st0325

The indirect command in STATA uses the ‘Bucher method’ to perform indirect comparisons using either fixed or random effects within each pairwise comparison. The command gives also in the output the estimated ‘indirect Q-statistic’ (see Bucher et al. 1996) for assessing heterogeneity in the indirect comparison.

See also: 

Miladinovic B, Hozo I, Chaimani A, Djulbegovic B. Indirect treatment comparison. Stata Journal, 2014, 14, 1, 76-86

Bucher HC, Guyatt GH, Griffith LE, Walter SD. The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials.J Clin Epidemiol1997, 50: 683-691

GeMTC software – performing NMA in Bayesian framework

-          Source: http://drugis.org/gemtc

The GeMTC software can be used either via a GUI (graphical user interface) application or via R. The software automatically generates models for network meta-analysis suitable for MCMC software, like WinBUGS, OpenBUGS and JAGS. The models can be then run directly or exported. It can generate random effects consistency and inconsistency models (the ‘Lu & Ades model’) as well as ‘node-splitting models’ for checking inconsistency. The output of the consistency model includes also the estimation of ranking probabilities.

See also:

van Valkenhoef G, Lu G, de Brock B, Hillege H, Ades AE, Welton NJ. Automating network meta-analysis.Res Synth Meth2012, 3: 285-299

van Valkenhoef G, Tervonen T, de Brock B, Hillege H. Algorithmic Parameterization of Mixed Treatment Comparisons. Statistics & Computing2012, 22: 1099-1111

Lu G, Ades AE. Assessing evidence inconsistency in mixed treatment comparisons.J Amer Statist Assoc2006, 101: 447-459

Dias S, Welton NJ, Caldwell DM, Ades AE. Checking consistency in mixed treatment comparison meta-analysis.Stat Med2010, 29: 932-944

BUGS codes – performing NMA in Bayesian environment

-          Sources: http://www.bris.ac.uk/social-community-medicine/projects/mpes & http://www.mtm.uoi.gr

The websites of MPES program (University of Bristol) and IMMA project (University of Ioannina) provide codes that can be used to perform network meta-analysis in WinBUGS, OpenBUGS or JAGS. Consistency fixed and random effects models are available for different types of data. The two websites provide also material (such as software codes and example datasets) from published papers in the field of networks meta-analysis.

Network graphs package – graphical tools for NMA in STATA

-          Source: http://www.mtm.uoi.gr

The network graphs package in STATA contains 8 commands that produce graphs for network meta-analysis. These graphs can be used to present the evidence base, the assumptions and the results of a network meta-analysis and aim to make the methodology accessible also to non-statisticians. Some of the commands are used combined with the mvmeta command. 

See also:

Chaimani A, Higgins JPT, Mavridis D, Spyridonos P, Salanti G. Graphical tools for network meta-analysis in STATA.Plos One2013, 8(10): e76654

ITC software – application for performing indirect comparisons

-          Source: https://www.cadth.ca/resources/itc-user-guide/

The ITC software is an application suitable for Windows environment that uses the ‘Bucher method’ to perform indirect comparisons. It requires as input the mean relative treatment effect and the respective 95% confidence interval for each direct comparison involved in a specific indirect root. The application can incorporate estimates on different effect measures (odds ratio, risk ratio, risk difference, mean difference, hazard ratio).

See also:

Wells GA, Sultan SA, Chen L, Khan M, Coyle D. Indirect evidence: Indirect treatment comparisons in meta-analysis. Ottawa,Canadian Agency for Drugs and Technologies in Health2009

Bucher HC, Guyatt GH, Griffith LE, Walter SD. The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials.J Clin Epidemiol1997, 50: 683-691

MetaXL - a free add-in for Microsoft Excel for Windows.

MetaXL supports conducting meta-analyses in Excel. Starting with version 4.0 (currently downloadable from http://www.epigear.com) it has added support for network meta-analysis.