[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: discrete time-varying covariate in cox models

From   "Daniel O. Koralek" <>
Subject   st: discrete time-varying covariate in cox models
Date   Mon, 12 Nov 2007 23:18:15 -0500


I have a scenario where I have a time-varying covariate, the number of screens that an individual has undergone, at up to 6 time points during follow-up. To simplify matters, let's say that I have a single covariate that I am modeling plus my screening variable, so the data set includes (and I am using age as the underlying time metric):
id entryage exitage iscase (my failure variable) x (my exposure) screenage0 - screenage5 (includes the ages at which the potential screening events occurred) and screen0 - screen5 (which simply is the number of screens that have been experienced up to and including screens that may have occurred at the corresponding screenage (i.e. screen2 corresponds to screenage2).

so, the stset looks like:
stset exitage, failure(iscase==1) enter(time entryage) exit(time exitage) scale(1)

Now, what I would like to do is simply control for a single screen variable that equals the number of screens that occurred up to the analysis time. THe examples that I have seen using stsplit seem to only use a single change (in this scenario, up to a certain point screen =0 and after screen = 1), not where multiple changes can occur.

Hopefully this makes sense. Any pointers or references would be greatly appreciated.

Thanks for your time.


Daniel O. Koralek
Department of Epidemiology/Lineberger Comprehensive Cancer Center
The University of North Carolina at Chapel Hill
Chapel Hill, NC 27599-7435

* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index