Dear all,
I would like to draw three forest plots to represent results at years 1, 2
and 3. I have the data as point estimates and 95% confidence intervals.
Using the following code I can get three basic forest plots - the first
which has the table of results. I have to plot each separately as the usual
par(mfrow=c(3,1)) does not work with the function forestplot within rmeta.
I can easily put them next to one another within powerpoint or similar
though so that's not a problem.
#Read data in
dattabrem1 <- read.table("risk factors rem1.txt",header=TRUE) # year 1
results
dattabrem2 <- read.table("risk factors rem2.txt",header=TRUE) # year 2
results
dattabrem3 <- read.table("risk factors rem3.txt",header=TRUE) # year 3
results
# Set up table of results for the three plots
plotextr <-
rbind(c("Age","Gender","Seizures","Treatment"),c("10","M","2","CBZ"),c("10","F","2","CBZ"),
c("10","M","2","LTG"),c("10","F","2","LTG"),c("10","M","10","CBZ"),c("10","F","10","CBZ"),
c("10","M","10","LTG"),c("10","F","10","LTG"),c("40","M","2","CBZ"),c("40","F","2","CBZ"),
c("40","M","2","LTG"),c("40","F","2","LTG"),c("40","M","10","CBZ"),c("40","F","10","CBZ"),
c("40","M","10","LTG"),c("40","F","10","LTG"),c("75","M","2","CBZ"),c("75","F","2","CBZ"),
c("75","M","2","LTG"),c("75","F","2","LTG"),c("75","M","10","CBZ"),c("75","F","10","CBZ"),
c("75","M","10","LTG"),c("75","F","10","LTG"))
# 1 year results
estimate1y <- c(NA,dattabrem1$HR)
lowerd1y <- c(NA,dattabrem1$CIlower)
upperd1y <- c(NA,dattabrem1$CIupper)
# 2 year results
estimate2y <- c(NA,dattabrem2$HR)
lowerd2y <- c(NA,dattabrem2$CIlower)
upperd2y <- c(NA,dattabrem2$CIupper)
# 3 year results
estimate3y <- c(NA,dattabrem3$HR)
lowerd3y <- c(NA,dattabrem3$CIlower)
upperd3y <- c(NA,dattabrem3$CIupper)
# Draw forest plots
forestplot(plotextr,estimate1y,lowerd1y,upperd1y,is.summary=c(TRUE,rep(FALSE,24)),zero="",align="c",xlab="Remission
@ 1 year")
forestplot(plotext2r,estimate2y,lowerd2y,upperd2y,zero="",xlab="Remission @
2 years")
forestplot(plotext2r,estimate3y,lowerd3y,upperd3y,zero="",xlab="Remission @
3 years")
Having managed to obtain these basic plots I need the x-axes to be the
same. Usually xlim would be sufficient to do this but this function is not
avaialble within forestplot so does anyone know how I can make the x-axes
the same over all the plots?
Additionally, within each plot, two treatments are compared (top two blocks
are treatment 1, 2nd 2 blocks are treatment 2, next 2 are treatment 1 etc.)
Is there any way I can colour code the boxes to show this?
Many thanks,
Laura
P.S. I'm using Windows, R 2.9.2
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
