Skip to content

Relative Survival in R - Getting 5 year and Median Survival

2 messages · Josh Rosen, Joonas Miettinen

#
Hello,

I am trying to compute the relative survival curves for a cohort of cancer
patients that I have data on. I have been trying to use some of the
elements of the relsurv package and have gotten to the point where I have
built my US Life Table, and have my data set in the proper form and am able
to run rs.surv rs.add and rs.surv.add. I also think I have been able to
plot the appropriate survival curve using some code found online. However,
what I am wondering is how to best extract the 5-year (and 3-year) and
median relative survivals for my cohort.

I am also trying to figure out the best way to then go about validating my
choices of method for relative survival computation as well.

I am fairly new to R (usually use SAS) so any help that you can provide
would be greatly appreciated.

Thank you!
-JR
#
Hey Josh,

use |summary| to get estimates for specific times. E.g.:

|data(slopop) data(rdata) #calculate the relative survival curve #note 
that the variable year is given in days since 01.01.1960 and that #age 
must be multiplied by 365 in order to be expressed in days. rs <- 
rs.surv(Surv(time,cens)~sex+ratetable(age=age*365,sex=sex, 
year=year),ratetable=slopop,data=rdata) summary(rs, times = c(3, 
5)*365.25) |

?
Not sure what you mean about the validation part, but you might try 
estimating both EdererII and Pohar-Perme curves and see if they're similar.

BR: Joonas

2015-11-06, 15:15, Josh Rosen kirjoitti: