Metaanalysis with zero event studies

General
1

Hello everyone,

I am working on a metaanalysis of proportions in RStudio looking at the proportion of complications between "treatment A (cases)" & "Treatment B (controls)". The studies from the control group reported 0 events and/or 0 complications and/or 100% success, for which I would have, Cases= 0 , given any total sample size per group.

This is my code for the "control group"

pes.logit.control=rma(yi,vi,data=ies.logit,subset=case_control=="Control",method="DL")

When I am pooling the effect size, I get this error message:

Error: Stopped because k = 0 after subsetting.

Could anybody help me figuring out this problem? I cannot exclude the control group because having 0 events is also important and informative.

Appreciate any help,
Claudia

2

I challenge you to justify that this is important and informative...

3

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.