The ecmeta R package provides tools to implement a meta-analytic framework that adjusts external control studies for additional bias and variability due to their non-randomized design. The meta-analysis is used to compare internal trial controls and external controls and is performed on a set of historical references studies. The parameters of the meta-analytic model are, in turn, used to to adjust estimate of a new study that compares an experimental treatment with an external control arm.
A paper describing the methodology and evauating its performance is available at the GitHub repository here. For an example analysis, see the user guide.
The R package can be installed from GitHub with:
devtools::install_github("phcanalytics/ecmeta")The meta-analytic models can be estimated using Bayesian techniques with JAGS. If you would like to use JAGS, you must first install it. On Linux:
sudo apt update
sudo apt-get install jagsOn OS X or Windows, you can install from the website.
Afterwards, you will also need to install the rjags package, which provides an R based interface:
install.packages("rjags")