Closed m7pr closed 1 year ago
Thanks for the post @m7pr !
Does survMisc add risk tables to survival figures?
@ddsjoberg anf @m7pr - survMisc is one of the packages we considered for the comparison table on the package website. But I think we didn't include as it doesn't look like it's an active project anymore. For example, I wasn't able to run the autoplot
example without error i.e.
library(survMisc)
data("bmt", package="KMsurv")
s2 <- survfit(Surv(time=t2, event=d3) ~ group, data=bmt)
autoplot(s2)
For transparency, the related packages provides a good overview of alternatives to ggsurvfit
.
@bailliem I get such message when I run your code (extended with library(survival)
)
! Objects of class
are not supported by autoplot.
So I guess there are other functions to support plotting survfit in survMisc.
Based on examples in ?survMisc::autoplot.survfit
you can see you need to apply ten
function over survfit
object
t1 <- ten(survfit(Surv(time, delta) ~ type, data=kidney))
autoplot(t1)
autoplot(t1, type="fill", survLineSize=2, jitter="all")
autoplot(t1, timeTicks="months",
type="CI", jitter="all",
legLabs=c("surgical", "percutaneous"),
title="Time to infection following catheter placement \n
by type of catheter, for dialysis patients",
titleSize=10, censSize=2)$plot
first two calls of autoplot()
with minimal parameter setup will fail (weeird)
> autoplot(t1)
Error in if (is.nan(times1[1])) times1 <- times :
argument is of length zero
> autoplot(t1, type="fill", survLineSize=2, jitter="all")
Error in if (is.nan(times1[1])) times1 <- times :
argument is of length zero
but this last, broader specification, gives this
autoplot(t1, timeTicks="months",
type="CI", jitter="all",
legLabs=c("surgical", "percutaneous"),
title="Time to infection following catheter placement \n
by type of catheter, for dialysis patients",
titleSize=10, censSize=2)$plot
@ddsjoberg
Does survMisc add risk tables to survival figures?
I think they do not
Thanks for following up with those code examples @m7pr !
I took a look at the available vignette on CRAN, and it is full of errors https://cran.r-project.org/web/packages/survMisc/vignettes/plots.pdf
given that some pretty basic code is erring, I don't think there is a benefit to adding it to the list. moreover, we could perhaps remove all the packages that don't add risktable (since creating a survival curve without the risktable is somewhat trivial)...
Thanks again for the post!
Yes, you are right
There are a couple more popular R packages for drawing survival curves. Would you consider adding survMisc https://cranlogs.r-pkg.org/badges/grand-total/survMisc ? It has 985K downloads so would be a good comparison CC @rossfarrugia