[R-pkgs] New package MBNMAdose: dose-response meta-analysis
I would like to introduce you to a new package available on CRAN for
performing meta-analysis of dose-response relationships, 'MBNMAdose'.
For anyone who has performed a meta-analysis or network meta-analysis
before, you might be aware of the difficulties of including different doses
of a drug. Typically one either assumes that different doses have a broadly
similar effect, or that they are totally independent.
MBNMAdose allows you to explicitly model the dose-response relationship in
a statistically robust framework. By modelling dose-response relationships
this can connect networks of evidence that might otherwise be disconnected,
and can improve precision on treatment estimates.
There's a detailed vignette included with the package that explains
everything in a lot more detail for those that are interested.
Any feedback or suggestions for improvements would be hugely welcome.
Senior Research Associate
Population Health Sciences
University of Bristol