gemtc-package: GeMTC: Network meta-analysis in R

Description Details Author(s) References See Also Examples

Description

An R package for performing network meta-analyses (mixed treatment comparisons).

Details

Network meta-analysis, or mixed treatment comparison (MTC) is a technique to meta-analyze networks of trials comparing two or more treatments at the same time [Dias et al. 2013a]. Using a Bayesian hierarchical model, all direct and indirect comparisons are taken into account to arrive at a single consistent estimate of the effect of all included treatments based on all included studies.

This package allows the automated generation of network meta-analysis models [van Valkenhoef et al. 2012], inclusing both fixed effect and random effects network meta-analysis, node-splitting models to identify inconsistency, and network meta-regression models. Models are estimated using JAGS (through the rjags package).

It is possible to get reproducible results, but as JAGS uses its own pseudo-random number generator, this is somewhat more involved. See mtc.model for details.

The source for GeMTC is available under the GPL-3 on Github.

Author(s)

Gert van Valkenhoef

References

S. Dias, N.J. Welton, D.M. Caldwell, and A.E. Ades (2010), Checking consistency in mixed treatment comparison meta-analysis, Statistics in Medicine 29(7-8, Sp. Iss. SI):932-944.
[doi: 10.1002/sim.3767]

S. Dias, A.J. Sutton, A.E. Ades, and N.J. Welton (2013a), A Generalized Linear Modeling Framework for Pairwise and Network Meta-analysis of Randomized Controlled Trials, Medical Decision Making 33(5):607-617. [doi: 10.1177/0272989X12458724]

S. Dias, A.J. Sutton, N.J. Welton, and A.E. Ades (2013b), Heterogeneity - Subgroups, Meta-Regression, Bias, and Bias-Adjustment, Medical Decision Making 33(5):618-640.
[doi: 10.1177/0272989X13485157]

S. Dias, N.J. Welton, A.J. Sutton, D.M. Caldwell, G. Lu, and A.E. Ades (2013c), Inconsistency in Networks of Evidence Based on Randomized Controlled Trials, Medical Decision Making 33(5):641-656. [doi: 10.1177/0272989X12455847]

A. Gelman, A. Jakulin, M. Grazia Pittau, Y.-S. Su (2008), A weakly informative default prior distribution for logistic and other regression models, The Annals of Applied Statistics 2(4):1360-1383. [doi: 10.1214/08-AOAS191]

R.M. Turner, J. Davey, M.J. Clarke, S.G. Thompson, J.P.T. Higgins (2012), Predicting the extent of heterogeneity in meta-analysis, using empirical data from the Cochrane Database of Systematic Reviews, International Journal of Epidemiology 41(3):818-827. [doi: 10.1093/ije/dys041]

G. van Valkenhoef, G. Lu, B. de Brock, H. Hillege, A.E. Ades, and N.J. Welton (2012), Automating network meta-analysis, Research Synthesis Methods 3(4):285-299. [doi: 10.1002/jrsm.1054]

G. van Valkenhoef, S. Dias, A.E. Ades, and N.J. Welton (2015), Automated generation of node-splitting models for assessment of inconsistency in network meta-analysis, Research Synthesis Methods, accepted manuscript. [doi: 10.1002/jrsm.1167]

G. van Valkenhoef et al. (draft), Modeling inconsistency as heterogeneity in network meta-analysis, draft manuscript.

D.E. Warn, S.G. Thompson, and D.J. Spiegelhalter (2002), Bayesian random effects meta-analysis of trials with binary outcomes: methods for the absolute risk difference and relative risk scales, Statistics in Medicine 21(11):1601-1623. [doi: 10.1002/sim.1189]

See Also

mtc.network, mtc.model, mtc.run

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
# Load the example network and generate a consistency model:
model <- mtc.model(smoking, type="consistency")

# Load pre-generated samples instead of runing the model:
## Not run: results <- mtc.run(model, thin=10)
results <- dget(system.file("extdata/luades-smoking.samples.gz", package="gemtc"))

# Print a basic statistical summary of the results:
summary(results)
## Iterations = 5010:25000
## Thinning interval = 10 
## Number of chains = 4 
## Sample size per chain = 2000 
## 
## 1. Empirical mean and standard deviation for each variable,
##    plus standard error of the mean:
## 
##         Mean     SD Naive SE Time-series SE
## d.A.B 0.4965 0.4081 0.004563       0.004989
## d.A.C 0.8359 0.2433 0.002720       0.003147
## d.A.D 1.1088 0.4355 0.004869       0.005280
## sd.d  0.8465 0.1913 0.002139       0.002965
## 
## 2. Quantiles for each variable:
## 
##          2.5%    25%    50%    75% 97.5%
## d.A.B -0.2985 0.2312 0.4910 0.7530 1.341
## d.A.C  0.3878 0.6720 0.8273 0.9867 1.353
## d.A.D  0.2692 0.8197 1.0983 1.3824 2.006
## sd.d   0.5509 0.7119 0.8180 0.9542 1.283

Example output

Loading required package: coda

Results on the Log Odds Ratio scale

Iterations = 5010:25000
Thinning interval = 10 
Number of chains = 4 
Sample size per chain = 2000 

1. Empirical mean and standard deviation for each variable,
   plus standard error of the mean:

        Mean     SD Naive SE Time-series SE
d.A.B 0.4965 0.4081 0.004563       0.004746
d.A.C 0.8359 0.2433 0.002720       0.003022
d.A.D 1.1088 0.4355 0.004869       0.005256
sd.d  0.8465 0.1913 0.002139       0.003034

2. Quantiles for each variable:

         2.5%    25%    50%    75% 97.5%
d.A.B -0.2985 0.2312 0.4910 0.7530 1.341
d.A.C  0.3878 0.6720 0.8273 0.9867 1.353
d.A.D  0.2692 0.8197 1.0983 1.3824 2.006
sd.d   0.5509 0.7119 0.8180 0.9542 1.283

gemtc documentation built on May 15, 2021, 1:07 a.m.