Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: AW: Case-cohort study


From   "Grote, Veit Dr." <[email protected]>
To   <[email protected]>
Subject   st: AW: Case-cohort study
Date   Wed, 23 Apr 2003 09:48:06 +0200

Hi Jonathan,

If you want to make a case-cohort analysis you should use survival analysis. Thats what is is about. If you like you could also use conditional logistic regression, however, that is just a name change.
However, what I did not understand so far: What is the merhodological advantage of a case-cohort over a case-control analysis? Maybe somebody has an easy to understand answer.

First, look at Barlow et al(Barlow WE, Ichicawa L, Rosner D, and Izumi S: Analysis of Case-Cohort Designs; Journal Clinical Epidemiology 1999; 52: 1165-1172) and at the accompanying example at http://lib.stat.cmu.edu/general/robphreg how a case-cohort should be analysed. Are not confused by all the discussion on robust variance. For practical purposes: Stata has the - robust - option handling that issue! 
If you have a dataset that already encompasses all cases and a sub-cohort of your total population than you "translate" the SAS program (relevant extract from the above webpage below) into Stata. To code your weights look at Table2 of above paper of Barlow et al.  
If your dataset includes all cases and the whole population than use :

ssc install stcascoh
ssc install strelpre

The help file of - stcascoh - also gives you the clou how to use stcox in the analysis.
Veit

******************************************************************************
data nickel1; set nickel20;  /* Standard Cox model counting process approach*/
   if (nasal=1) then do; /* Cases */
           survtime=timestop;
           logwt=log(1);
           start=survtime-0.00005;
           output;
           if (subco20=1) then do;   /* Subcohort case treated as control */
                 logwt=log(5);  /* Weight for 20% subcohort=100/20 */
                 start=timestrt;
                 survtime=survtime-0.00005;
                 nasal=2;
                 output;
                 end;
           end;
     else if subco20=1 & nasal=2 then do; /* Subcohort controls */
           survtime=timestop;
           logwt=log(5);
           start=timestrt;
           output;
           end;
     run;
******************************************************************************
-----Urspr�ngliche Nachricht-----
Von: Jonathan Golub [mailto:[email protected]]
Gesendet: Dienstag, 22. April 2003 21:46
An: [email protected]
Betreff: st: Case-cohort study


Hello,
I have analyzed my dataset as a case-control study (via logistic 
regression), and I am now interested in analyzing it as a case-cohort 
study.  The only help I could find online or in the help section 
suggests analyzing my data as survival time analysis which I do not 
want to do.  Please tell me how it is possible to set my data so that 
it can be analyzed as a case-cohort design, via logistic regression.  
The problem I am having is that in order to use the entire population  
(including cases) as my controls I need to define a case-control 
variable that equals 1 for case and 0 for control, but all cases are 
also controls. Thus, defining such a variable seems impossible.

Thank you,
Jonathan Golub

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index