[R] Survival Curves

[David Winsemius]

On Apr 16, 2012, at 1:02 PM, robgriffin247 wrote:

> actually, maybe I haven't quite solved this - the plot doesn't  
> appear to
> include the 23 that are still alive at the end.
>
> I have tried using a status variable (1 or 0  dependant upon  
> survival as in
> page 800, roughly, in the R book)
> ...
> Call: survfit(formula = Surv(ssl$long, ssl$status) ~ ssl$sex)
> 23 observations deleted due to missingness
>
> (these 23 will be the 23 long - longevity - scores of NA)

If a case is censored you need to code the longevity value as the  
duration at last observation, .... not NA. (Crawley's "R Book" is not  
particularly good at explaining this or many other aspects of R, IMO).

   The code will take care of the censoring because you code its  
status as "alive" == 0 as of last observation. When I do this, I  
create my survival-years as:

   surv.yr <- ifelse(status,  death.dt-entry.dt, study.close.dt - 
entry.dt )/365.25

-- 
David.
>
> and with the etype
> (http://stat.ethz.ch/R-manual/R-devel/library/survival/html/survfit.formula.html 
> )
> and censor=T
> (http://stat.ethz.ch/R-manual/R-devel/library/survival/html/survfit.coxph.html 
> )
> arguments but I can't seem to get them to work, maybe it's because  
> its a
> monday night, but I seem to have hit a dead end... bad times.
>
>
> robgriffin247 wrote
>>
>> Hi David thanks for the help, i was just looking back to say I have  
>> found
>> my solution (i think) and as you suggested survfit was the way to go.
>>
>> my solution was this:
>> model<-survfit(Surv(a$long)~a$sex)
>> plot(model,ylab="proportion alive",xlab="time  
>> (days)",col=c("red","blue"))
>>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/Survival-Curves-tp4561685p4562155.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at 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.

David Winsemius, MD
West Hartford, CT