Skip to content

[R-meta] Calculation of z scores in NMA analysis

2 messages · Roopa Subbaiah, Gerta Ruecker

Roopa Subbaiah
# 
Hi All,

Need help regarding calculation of z scores.

We are working on nma analysis with three treatments compared to placebo.
We are using netmeta package for stats. We have curated mean and sd values
for each endpoint w. r. t trt and placebo. Need help to calculate weighted
mean z scores with standard error. Please suggest ways of doing it. Thank
you
Gerta Ruecker
# 
Hi Roopa,

First of all, I want to remind you of Wolfgang's e-mail referring to the 
transparency rules of this mailing list. He wrote on 24 of March: "When 
posting to the mailing list, please use your real name and include 
information about any institutional/organizational affiliation if possible."

Now to your question: It is not fully clear to me what you want to do 
and why. If this is a network meta-analysis (NMA) of differences of 
means, you need mean, sd, and sample size (n) for each comparison. 
netmeta will then provide the NMA estimates for all comparisons in the 
network, again as differences of means (MD). The resulting z-scores are 
just tabulated with the printout, as you can see with the built-in 
example data Senn2013:

library(netmeta)
data(Senn2013)
net1 <- netmeta(TE, seTE, treat1.long, treat2.long, studlab,
 ? data = Senn2013, sm = "MD", random = FALSE, nchar.trts = 4)
net1

Isn't this what you want? If not, what else do you expect?

Best,

Gerta

Am 16.04.2022 um 12:23 schrieb Roopa Subbaiah: