Network meta-analysis in diabetes
Network meta-analysis in diabetes comparing effects of a number of drugs on the HbA1c value.
These data are used as an example in Senn et al. (2013) and have been preprocessed for use in R package netmeta.
A data frame with the following columns:
| TE | treatment effect |
| seTE | standard error of treatment effect |
| treat1 | treatment 1 |
| treat2 | treatment 2 |
| treat1.long | treatment 1 (full treatment names) |
| treat2.long | treatment 2 (full treatment names) |
| studlab | Study label |
Treatment labels provided by columns treat1 and
treat2 have been abbreviated:
acar = Acarbose
benf = Benfluorex
metf = Metformin
migl = Miglitol
piog = Pioglitazone
plac = Placebo
rosi = Rosiglitazone
sita = Sitagliptin
sulf = Sulfonylurea
vild = Vildagliptin
Full treatment names are available in columns treat1.long
and treat2.long.
Senn S, Gavini F, Magrez D, Scheen A (2013): Issues in performing a network meta-analysis. Statistical Methods in Medical Research, 22, 169–89
data(Senn2013)
# Fixed effects model
#
net1 <- netmeta(TE, seTE, treat1, treat2,
studlab, data = Senn2013,
comb.random = FALSE)
net1
net1$Q.decomp
# Forest plot
#
forest(net1, ref = "plac")
## Not run:
# Comparison with reference group
#
netmeta(TE, seTE, treat1.long, treat2.long,
studlab, data = Senn2013,
reference = "plac")
# Random effects model
#
net2 <- netmeta(TE, seTE, treat1, treat2, studlab, data = Senn2013,
comb.fixed = FALSE)
net2
forest(net2, ref = "plac")
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.