Skip to content

[R-meta] using netmeta::netsplit()

4 messages · Dale Steele, Dr. Gerta Rücker, Guido Schwarzer

#
I'm doing a network meta-analysis, using netmeta (v2.9-0). There are a
large number
of potential comparisons in my network.

My current plot displays all comparisons (too many) where
least one study contributes direct evidence.

ns1 <- netsplit(net1)
forest(ns1, show = "both")

My goal is to produce a forest plot showing only comparisons for which there
are at least 3 studies with direct evidence.

I think I want the 7 comparisons where k > 2
[1] 120 127 130 132 147 237 253

Thanks!

--Dale
4 days later
#
Hi Dale,

I don't know the answer to your question (perhaps Guido has an idea), but there is a different way to (hopefully) shorten your output. By
you see the results only for comparisons with the reference treatment.

Best,
Gerta


UNIVERSIT?TSKLINIKUM FREIBURG
Institute for Medical Biometry and Statistics

Dr. Gerta R?cker
Guest Scientist

Stefan-Meier-Stra?e 26 ? 79104 Freiburg
gerta.ruecker at uniklinik-freiburg.de

https://www.uniklinik-freiburg.de/imbi-en/employees.html?imbiuser=ruecker


-----Urspr?ngliche Nachricht-----
Von: Dale Steele via R-sig-meta-analysis <r-sig-meta-analysis at r-project.org> 
Gesendet: Samstag, 16. Dezember 2023 00:29
An: r-sig-meta-analysis at r-project.org
Cc: Dale Steele <dale.w.steele at gmail.com>
Betreff: [R-meta] using netmeta::netsplit()

I'm doing a network meta-analysis, using netmeta (v2.9-0). There are a
large number
of potential comparisons in my network.

My current plot displays all comparisons (too many) where
least one study contributes direct evidence.

ns1 <- netsplit(net1)
forest(ns1, show = "both")

My goal is to produce a forest plot showing only comparisons for which there
are at least 3 studies with direct evidence.

I think I want the 7 comparisons where k > 2
[1] 120 127 130 132 147 237 253

Thanks!

--Dale


_______________________________________________
R-sig-meta-analysis mailing list @ R-sig-meta-analysis at r-project.org
To manage your subscription to this mailing list, go to:
https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis
#
Dale,

On GitHub, I added a new argument 'subset' to forest.netsplit() and print.netsplit() which can be used to define a subset of comparisons to print. E.g.,
forest(ns1, subset = k >= 3)

Best,
Guido
#
Guido,

Works like a charm!

Best.

--Dale

On Thu, Dec 21, 2023 at 12:30?PM Dr. Guido Schwarzer <
guido.schwarzer at uniklinik-freiburg.de> wrote: