Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: the use of stcompet


From   "Philip Ryan" <philip.ryan@adelaide.edu.au>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: the use of stcompet
Date   Sun, 5 Apr 2009 13:37:31 +0930

Natalie

Enzo Coviello would no doubt be the best person to assist, but in the
meantime, recall  that the help file for -stcompet- says:

"  Note that each created variable contains the function for all competing
    events : i.e. the event of interest specified in stset statement and the
    events in compet# options. "

So, if your failure variable "path" has categories OTHER than 1, 2 or 3
(these are the only three that -stcompet- knows about in your data),
-stcompet- will not estimate the cumulative incidence function for those
categories, and a missing value is the result. Typically, a failure event
variable in a competing risk situation might still have an administrative
censoring category, coded as 0. You might choose to exclude admin censoring
as a competing risk, and so event=0 will end up with a missing cumulative
incidence.

So, in your case, with "path" as the event variable, see what  -tabulate
path-  reveals.

[My advice is based on the following information after issuing -which
stcompet-:
*! stcompet 1.0.4 EC - MB 2 FEB 2004
*! Cumulative Incidence in the Presence of Competing Events.

It is helpful if you state the version of the command you are dealing with,
especially if it is a user-written one.]

I hope this helps.

Phil


Philip Ryan
Professor and Director
Data Management & Analysis Centre
Discipline of Public Health
University of Adelaide
Australia 

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of natalie chan
Sent: Sunday, 5 April 2009 10:49 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: the use of stcompet


Dear Statalister,

I am trying to use Stcompet to determine the cause specific mortality rate
on a dataset with about 3500 subjects, 30000 observations and 3 possible
causes.

I use stset first:

stset time1, id(id) time0(time0) failure(path==1)

the results are as follows:

 id:  id
     failure event:  path == 1
obs. time interval:  (time0, time1]
 exit on or before:  failure

----------------------------------------------------------------------------
--
    30367  total obs.
     2671  obs. begin on or after (first) failure
----------------------------------------------------------------------------
--
    27696  obs. remaining, representing
     3521  subjects
      563  failures in single failure-per-subject data
    27696  total analysis time at risk, at risk from t =         0
                             earliest observed entry t =1
                                  last observed exit t =15


then I use stcompet:

stcompet test1=ci test2=se, compet1(2) compet2(3)

In the new variables generated, there are numbers when the failure variable
(=1, 2 or 3), but in other rows, the value is missing.

I have searched Statalist Archive and found that someone had the same
problem but I did not find the answer. I would be grateful if anyone could
help me on this.

Many thanks.

Natalie
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

No virus found in this incoming message.
Checked by AVG - www.avg.com 
Version: 8.0.238 / Virus Database: 270.11.40/2039 - Release Date: 04/03/09
17:54:00


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index