R help - Apr 2013 - Convert results from print(survfit(formula, ...)) into a matrix or data frame

Hello All,

Below is some sample survival analysis code. I'd like to able to get the
results from print(gehan.surv) into a matrix or data frame, so I can manipulate
them and then create a table using odfWeave. Trouble is, I'm not quite sure
how make such a conversion using the results from a print method.

Is there some simple way of doing this?

Thanks,

Paul

require(survival)
require(MASS)

attach(gehan)
gehan.surv <- survfit(Surv(time, cens) ~ treat, data= gehan, conf.type =
"log-log")
print(gehan.surv)

On Apr 15, 2013, at 11:31 AM, Paul Miller <pjmiller_57 at yahoo.com>
wrote:
> Hello All,
> 
> Below is some sample survival analysis code. I'd like to able to get
the results from print(gehan.surv) into a matrix or data frame, so I can
manipulate them and then create a table using odfWeave. Trouble is, I'm not
quite sure how make such a conversion using the results from a print method.
> 
> Is there some simple way of doing this?
> 
> Thanks,
> 
> Paul
> 
> require(survival)
> require(MASS)
> 
> attach(gehan)
> gehan.surv <- survfit(Surv(time, cens) ~ treat, data= gehan, conf.type =
"log-log")
> print(gehan.surv)
> 


Hi Paul,

If I am correctly understanding the output that you want to save, which is the
summary results table output:
> gehan.surv
Call: survfit(formula = Surv(time, cens) ~ treat, data = gehan, conf.type =
"log-log")

              records n.max n.start events median 0.95LCL 0.95UCL
treat=6-MP         21    21      21      9     23      13      NA
treat=control      21    21      21     21      8       4      11


You can use:
> summary(gehan.surv)$table
              records n.max n.start events median 0.95LCL 0.95UCL
treat=6-MP         21    21      21      9     23      13      NA
treat=control      21    21      21     21      8       4      11

> str(summary(gehan.surv)$table)
 num [1:2, 1:7] 21 21 21 21 21 21 9 21 23 8 ...
 - attr(*, "dimnames")=List of 2
  ..$ : chr [1:2] "treat=6-MP" "treat=control"
  ..$ : chr [1:7] "records" "n.max" "n.start"
"events" ...


See ?summary.survfit and note the 'table' part of the Value section.

Regards,

Marc Schwartz

Hi Marc,

Oh, yes. That is quite simple. Silly me for not recognizing this. Thanks very
much for your help.

Paul