# Define the data as a string
data <- "
Study_ID\tsample\tAdv_Post_op\tAdv_12_mon\tAdv_long\tsubgroup
Christmas 2010\t20\t19\t15\t105\tsurgery
Subramanina 2016\t45\t26\t4\t0\tsurgery
Josue 2019\t10\t10\t8\tNA\tsurgery
Bergfeld 2016\t25\t14\tNA\tNA\tstimulation
Malone 2008\t15\t3\tNA\tNA\tstimulation
Dougherty 2014\t16\t7\tNA\tNA\tstimulation
"
# Read the data into a dataframe
adv <- read.table(text = data, header = TRUE, sep = "\t", na.strings = "NA")
# View the dataframe
print(adv)
## Study_ID sample Adv_Post_op Adv_12_mon Adv_long subgroup
## 1 Christmas 2010 20 19 15 105 surgery
## 2 Subramanina 2016 45 26 4 0 surgery
## 3 Josue 2019 10 10 8 NA surgery
## 4 Bergfeld 2016 25 14 NA NA stimulation
## 5 Malone 2008 15 3 NA NA stimulation
## 6 Dougherty 2014 16 7 NA NA stimulation
library(meta)
## Loading required package: metadat
## Loading 'meta' package (version 7.0-0).
## Type 'help(meta)' for a brief overview.
## Readers of 'Meta-Analysis with R (Use R!)' should install
## older version of 'meta' package: https://tinyurl.com/dt4y5drs
meta_analysis1 <- metaprop(event = Adv_Post_op, n = sample, data = adv,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID,
byvar = subgroup
)
summary(meta_analysis1)
## proportion 95%-CI %W(random) subgroup
## Christmas 2010 0.9500 [0.7513; 0.9987] 10.9 surgery
## Subramanina 2016 0.5778 [0.4215; 0.7234] 23.6 surgery
## Josue 2019 1.0000 [0.6915; 1.0000] 6.9 surgery
## Bergfeld 2016 0.5600 [0.3493; 0.7560] 21.8 stimulation
## Malone 2008 0.2000 [0.0433; 0.4809] 17.0 stimulation
## Dougherty 2014 0.4375 [0.1975; 0.7012] 19.8 stimulation
##
## Number of studies: k = 6
## Number of observations: o = 131
## Number of events: e = 79
##
## proportion 95%-CI
## Random effects model 0.5925 [0.3792; 0.7759]
##
## Quantifying heterogeneity:
## tau^2 = 0.7375 [0.3196; 12.2732]; tau = 0.8588 [0.5653; 3.5033]
## I^2 = 72.2% [35.9%; 88.0%]; H = 1.90 [1.25; 2.88]
##
## Test of heterogeneity:
## Q d.f. p-value
## 18.01 5 0.0029
##
## Results for subgroups (random effects model):
## k proportion 95%-CI tau^2 tau Q I^2
## subgroup = surgery 3 0.8626 [0.4245; 0.9816] 2.6811 1.6374 8.97 77.7%
## subgroup = stimulation 3 0.4124 [0.2271; 0.6265] 0.3322 0.5764 4.58 56.3%
##
## Test for subgroup differences (random effects model):
## Q d.f. p-value
## Between groups 3.45 1 0.0632
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies
## - Continuity correction of 0.5 in studies with zero cell frequencies
# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis1, layout = "RevMan")
Adverse effects