Quantcast

Comparing survival curves with "survdiff" "strata" help

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Comparing survival curves with "survdiff" "strata" help

Richardson, Patrick
ExpeRts,

I'm trying to compare three survival curves using the function "survdiff" in the survival package.  Following is my code and corresponding error message.
> survdiff(Surv(st_months, status) ~ strata(BOR), data=mydata)
Error in survdiff(Surv(st_months, status) ~ strata(BOR), data = mydata) :
  No groups to test

When I check the "strata" of the variable.  I get . . .
> strata(BOR)
Levels: BOR=PD BOR=PR BOR=SD

So I know there are levels associated with the BOR variable, but I can't seem to get the strata statement to work within "survdiff".  I checked ?strata and (unless I am not understanding) the help, it seems to be working as I get the following output
> levels(strata(BOR))
[1] "BOR=PD" "BOR=PR" "BOR=SD"

I can get the survival tables and KM plot with no problem, but I can't seem to be able to compare the survival curves.  I can get an overall comparison if I remove the "strata" statement
> survdiff(Surv(st_months, status) ~ BOR, data=mydata)
Call:
survdiff(formula = Surv(st_months, status) ~ BOR, data = mydata)

n=35, 5 observations deleted due to missingness.

        N Observed Expected (O-E)^2/E (O-E)^2/V
BOR=PD 13       13     6.51     6.464      8.87
BOR=PR 10       10    18.48     3.888     10.59
BOR=SD 12       12    10.01     0.395      0.61

 Chisq= 13.5  on 2 degrees of freedom, p= 0.00117

But when I use survdiff(Surv(st_months, status) ~ strata(BOR), data=mydata)  it doesn't want to work.  Any suggestions?

Many thanks for your assistance!


_____________________________________________
Patrick Richardson
Biostatistician - Program of Translational Medicine
Van Andel Research Institute - Webb Lab
333 Bostwick Avenue NE
Grand Rapids, MI  49503

This email message, including any attachments, is for th...{{dropped:9}}

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Comparing survival curves with "survdiff" "strata" help

Richardson, Patrick
ExpeRts,

I'm trying to compare three survival curves using the function "survdiff" in the survival package.  Following is my code and corresponding error message.
> survdiff(Surv(st_months, status) ~ strata(BOR), data=mydata)
Error in survdiff(Surv(st_months, status) ~ strata(BOR), data = mydata) :
  No groups to test

When I check the "strata" of the variable.  I get . . .
> strata(BOR)
Levels: BOR=PD BOR=PR BOR=SD

So I know there are levels associated with the BOR variable, but I can't seem to get the strata statement to work within "survdiff".  I checked ?strata and (unless I am not understanding) the help, it seems to be working as I get the following output
> levels(strata(BOR))
[1] "BOR=PD" "BOR=PR" "BOR=SD"

I can get the survival tables and KM plot with no problem, but I can't seem to be able to compare the survival curves.  I can get an overall comparison if I remove the "strata" statement
> survdiff(Surv(st_months, status) ~ BOR, data=mydata)
Call:
survdiff(formula = Surv(st_months, status) ~ BOR, data = mydata)

n=35, 5 observations deleted due to missingness.

        N Observed Expected (O-E)^2/E (O-E)^2/V
BOR=PD 13       13     6.51     6.464      8.87
BOR=PR 10       10    18.48     3.888     10.59
BOR=SD 12       12    10.01     0.395      0.61

 Chisq= 13.5  on 2 degrees of freedom, p= 0.00117

But when I use survdiff(Surv(st_months, status) ~ strata(BOR), data=mydata)  it doesn't want to work.  Any suggestions?

Many thanks for your assistance!


_____________________________________________
Patrick Richardson
Biostatistician - Program of Translational Medicine
Van Andel Research Institute - Webb Lab
333 Bostwick Avenue NE
Grand Rapids, MI  49503

This email message, including any attachments, is for th...{{dropped:9}}

______________________________________________
[hidden email] 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.
Loading...