0

I'm doing a meta-analysis of proportions using metaprop in R. I'm trying to make my final forest plots look better, but have been having issues removing the heading "Common effect model" or even renaming it to the conventional "Fixed effects model". I've looked through the code for labels that may be correct and also across google, but can't find anyone with a similar issue. Any help would be appreciated.

pes.summary=metaprop(cases,total,authoryear,data=dat,sm="PLO",method.ci="NAsm",method.tau="DL",incr=0.5,allincr=FALSE,addincr=FALSE,title="")
png("forestplotMassDXA.png",width=800,height=1200)
forest(pes.summary,
       xlim = c(0,1), pscale =1, 
       rightcols=c("effect", "ci","w.random"),
       rightlabs=c("Proportion", "95% C.I.","Weights"),
       leftcols = c("studlab", "event", "n"), 
       leftlabs = c("Study", "Successful", "Total"),
       xlab = "Proportion", 
       fs.xlab=12,
       fs.study=12,
       fs.study.lables=12,
       fs.heading=12,
       squaresize = 0.5, col.square="navy", col.square.lines="navy",
       col.diamond="navy", col.diamond.lines="navy",
       comb.fixed=FALSE,
       lty.fixed=0,
       lty.random=2, 
       Modality.study="square",
       type.random="diamond",
       ff.fixed="bold.italic",
       ff.random="bold.italic",
       hetlab = "Heterogeneity:",
       fs.hetstat=8,
       smlab="",
       print.Q=TRUE,
       print.pval.Q=TRUE,
       print.I2=TRUE,
       print.tau2=FALSE,
       col.by="black")

[How my forest plot looks] [1]: https://i.stack.imgur.com/arkDK.png

SirHugo
  • 1
  • 1
  • It's easier to help you if you provide a [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output that can be used to test and verify possible solutions. – MrFlick Jul 12 '22 at 14:57
  • Thank you. I will amend the above now. – SirHugo Jul 12 '22 at 15:06

1 Answers1

0

Not sure if you still need this, but to remove it you could just add common = FALSE in your command so it looks like this:

forest(pes.summary,
       xlim = c(0,1), pscale =1, 
       rightcols=c("effect", "ci","w.random"),
       rightlabs=c("Proportion", "95% C.I.","Weights"),
       leftcols = c("studlab", "event", "n"), 
       leftlabs = c("Study", "Successful", "Total"),
       xlab = "Proportion", 
       fs.xlab=12,
       fs.study=12,
       fs.study.lables=12,
       fs.heading=12,
       squaresize = 0.5, col.square="navy", col.square.lines="navy",
       col.diamond="navy", col.diamond.lines="navy",
       comb.fixed=FALSE,
       lty.fixed=0,
       lty.random=2, 
       Modality.study="square",
       type.random="diamond",
       ff.fixed="bold.italic",
       ff.random="bold.italic",
       hetlab = "Heterogeneity:",
       fs.hetstat=8,
       smlab="",
       print.Q=TRUE,
       print.pval.Q=TRUE,
       print.I2=TRUE,
       print.tau2=FALSE,
       col.by="black",
       common = FALSE)

But to change it to the conventional Fixed-effect model is another story. I am eager to know if someone manages to do it