Statalist


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

Re: st: the use of stcompet


From   Enzo Coviello <enzo.coviello@tin.it>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: the use of stcompet
Date   Sun, 05 Apr 2009 08:29:30 +0200

Hi Natalie,

the Phil's answer is complete about your problem.

I would add that you can download the paper where the use of -stcompet- is illustrated from
http://www.statajournal.com/sjpdf.html?articlenum=st0059

Here some practical example can help you in using the output of -stcompet- for making graphs.

Note that -stcompet- does not estimate the cause specific mortality rates but the cumulative incidence function in the presence of the competing risks, a quite different function.

I wrote on this topic a further macro -stcompadj- that computes the covariate-adjusted cumulative incidence according to a method proposed by Rosthoj and Andersen. The macro is not yet available on the SSC Archive because I would add a delta-method estimate of the variance of this function. At present -stcompadj- computes the bootstrap confidence intervals of the adjusted cumulative incidence that seem quite good but I am not completely convinced about their validity. I would pleased to send privately to other Stata users -stcompadj- for further checks and, eventually, for adding the delta-method variance estimate of the covariate-adjusted cumulative incidence.

Best wishes.

Enzo



natalie chan ha scritto:
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/



--
Enzo Coviello
Unità di Epidemiologia e Statistica ASL Bari
Piazza V. Emanuele, 14 - 70054 GIOVINAZZO (BA)
Italy
tel +39 080 3357883 - fax +39 080 3357867
mobile 347 5016016 ^^^^
Home
via Imbriani, 186 - 70055 MINERVINO MURGE (BA)
tel +39 0883 695055

*
*   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