[R] Surv analysis with multiple internal time-dep covariates measured over different time intervals

From: <z.dalton_at_lancaster.ac.uk>
Date: Tue 18 Jul 2006 - 22:37:06 EST


I am analysing survival data (diagnosis time until death/cens) with time-dependent

 covariates. I would like to fit a cox model using the (start, stop] variable.    

  In summary, I have the multiple internal time dependent covariates as follows;

1). LAS score (measured weekly on low risk patients, monthly on high risk)

2). EORTC score (measured monthly on low risk patients and every 3 months on

 high risk)
3). BMI (measured monthly on low risk patients and every 3 months on high risk)

I have referred to the John Fox 'Cox Proportional-Hazards Regression for Survival


and the corresponding script file at

and also to Therneau and Grambsch.

My problem is creating the dataset, possibly using the fold function (as described

 in Fox, p9) with more than one time-dependent covariate (which I successfully

 did with LAS). I have longitudinal measurements for each subject (with each

 date of assessment) as above with some missing data in a period of time before

 death (which I have entered as NA).

Since the measurements in time depend on whether the patient is high or low

 risk and are made at different time intervals for each covariate, I wasn't

 sure how to code this in R.

I would be really grateful if anyone could start me off on this.

Thank you to those who respond.


R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Wed Jul 19 02:03:25 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Wed 19 Jul 2006 - 04:16:54 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.