[R-meta] Meta-analysis including multiple intervention arms

[James Pustejovsky]

The short answer is no. If you are using RVE with a correlated effects working model, the (assumed) correlation between the effect sizes from each intervention arm takes care of the issue. Splitting the control sample size would over-correct, so that the study would end up with less weight than it should receive.

James

> On Nov 14, 2020, at 8:44 AM, Ioana Cristea <ioana.alina.cristea using gmail.com> wrote:
> 
> Dear all
> I am doing a meta-analysis of randomized trials where some trials have
> multiple outcomes of interest (depression measured with different scales)
> and a few have more than one active arms, usually 2 intervention arms and
> control (therapy A compared to control, therapy B compared to control).
> I am using robust variance estimator. My question is: for the cases where
> there are multiple active intervention arms in a trial, compared to the
> same control arm, should I adjust the sample size of the control arm (ie,
> splitting by 2 if there are 2 active interventions compared to control)?
> Thank you
> Ioana
> 
> -- 
> Ioana-Alina Cristea, Ph.D.
> Assistant Professor
> Department of Brain and Behavioral Sciences
> University of Pavia, Piazza Botta 11, 27100 Pavia, Italy
> 
> Research affiliate
> METRICS (Meta-research Innovation Center at Stanford)
> Stanford University, California, USA
> Editorial Advisory Board Lancet Psychiatry
> <https://www.thelancet.com/journals/lanpsy/home>
> International Editorial Board BJPsych Open
> <https://www.cambridge.org/core/journals/bjpsych-open>
> 
> ORCid: http://orcid.org/0000-0002-9854-7076
> Publons: https://publons.com/author/1196647/ioana-alina-cristea#profile
> Researchgate: https://www.researchgate.net/profile/Ioana_Cristea
> Twitter: https://twitter.com/IoanaA_Cristea
> 
>    [[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-meta-analysis mailing list
> R-sig-meta-analysis using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis