[R-meta] Question RE 2-stage MA of TTE outcomes

[Viechtbauer, Wolfgang (SP)]

Dear Theodosia,

If all that you really have is the number of events and non-events per study, you do not have the necessary information for a time-to-event outcomes analysis, because you have no time information. The best you can do then is to analyze the proportions (or transformations thereof) in the usual manner. Packages meta (function metaprop()) and metafor (functions rma() and rma.glmm()) provide the necessary functionality for this.

If you actually have time information, in particular, the total follow-up time for each study, then you can compute incidence rates (or transformations thereof) and analyze those. Again, meta (function metarate()) and metafor (functions rma() and rma.glmm()) could be used for that.

Best,
Wolfgang

-----Original Message-----
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Salika, Theodosia
Sent: Thursday, 23 May, 2019 8:46
To: r-sig-meta-analysis using r-project.org
Subject: [R-meta] Question RE 2-stage MA of TTE outcomes

Dear all,

I hope this email finds you well.

I was wondering whether there is an option in R MA packages to conduct a 2-stage meta-analysis of time-to-event outcomes having only the number of events and non-events per study without having lnHR and SE or O-E and logrank V.

I know there is a command in STATA which can provide you with the necessary results having only no. of events and non-events. But I am analysing my results in R and I was wondering if this exist in R packages as well or should I calculate it step by step by myself. I will be using the complementary loglog link in order to derive the results.

I apologise in advance if I am asking something very simple but I am new both in the meta-analysis field and in R.

Thank you in advance for your help.

Best wishes,
Theodosia