Sub-group plot from meta-analysis using ggplot2-CodePudding

can someone help me with the construction of the meta-analysis sub-plot with the ggplot2 library.

Here is the reproducible sample:

library(ggplot2)
library(meta)

study <- c("Study 1","Study 2","Study 3","Study 4","Study 5",
           "Study 6","Study 7","Study 8","Study 9","Study 10",
           "Study 11","Study 12","Study 13","Study 14","Study 15")
nT <- c(155,31,75,18,8,57,34,110,60,20,11,32,36,97,80)
meanT <- c(55,27,64,66,14,19,52,21,30,45,32,80,40,25,70)
sdT <- c(47,7,17,20,8,7,45,16,27,11,6,22,31,4,32)
nC <- c(156,32,71,18,13,52,33,183,52,22,14,32,44,93,81)
meanC <- c(75,29,119,137,18,18,41,31,23,16,44,65,22,11,90)
sdC <- c(64,4,29,48,11,4,34,27,20,5,21,37,8,2,55)
pH <- c("high","low","high","low","low","high","low","high",
        "high","low","low","low","low","high","high")
dt <- data.frame(study,nT,meanT,sdT,nC,meanC,sdC,pH)
head(dt, 5)

# meta-analysis model
dt$pH <- factor(dt$pH, levels = c("low", "high"))
m1 <- metacont(nT, meanT, sdT,
               nC, meanC, sdC,
               fixed = FALSE,
               random = TRUE,
               subgroup = pH,
               data = dt)
summary(m1)

# forest plot with two pH levels (and individual studies)
forest(m1, subgroup = TRUE, print.byvar = FALSE)

I want to see the plot with the effect sizes (and CIs) only for two pH levels without individual studies.

Many thanks in advance!

CodePudding user response：

What you could do is create a dataframe which is a group_by of your pH where you calculate the mean per columns, so you have two rows which describe the low and high pH like this:

library(ggplot2)
library(dplyr)
library(meta)

study <- c("Study 1","Study 2","Study 3","Study 4","Study 5",
           "Study 6","Study 7","Study 8","Study 9","Study 10",
           "Study 11","Study 12","Study 13","Study 14","Study 15")
nT <- c(155,31,75,18,8,57,34,110,60,20,11,32,36,97,80)
meanT <- c(55,27,64,66,14,19,52,21,30,45,32,80,40,25,70)
sdT <- c(47,7,17,20,8,7,45,16,27,11,6,22,31,4,32)
nC <- c(156,32,71,18,13,52,33,183,52,22,14,32,44,93,81)
meanC <- c(75,29,119,137,18,18,41,31,23,16,44,65,22,11,90)
sdC <- c(64,4,29,48,11,4,34,27,20,5,21,37,8,2,55)
pH <- c("high","low","high","low","low","high","low","high",
        "high","low","low","low","low","high","high")
dt <- data.frame(study,nT,meanT,sdT,nC,meanC,sdC,pH)
head(dt, 5)
#>     study  nT meanT sdT  nC meanC sdC   pH
#> 1 Study 1 155    55  47 156    75  64 high
#> 2 Study 2  31    27   7  32    29   4  low
#> 3 Study 3  75    64  17  71   119  29 high
#> 4 Study 4  18    66  20  18   137  48  low
#> 5 Study 5   8    14   8  13    18  11  low

# This part to aggregated all colums
dt <- dt %>%
  group_by(pH) %>%
  summarise(across(everything(), mean))

# meta-analysis model
dt$pH <- factor(dt$pH, levels = c("low", "high"))
m1 <- metacont(nT, meanT, sdT,
               nC, meanC, sdC,
               fixed = FALSE,
               random = TRUE,
               subgroup = pH,
               data = dt)

# forest plot with two pH levels (and individual studies)
forest(m1, subgroup = TRUE, print.byvar = FALSE)

^{Created on 2022-07-23 by the}

I guess this function simply has too many parameters to make it easy to find the right options...