Is there any possibility to add a time-dependent covariate in Fine-Gray model (a proportional hazards model for the subdistribution of a competing risk; crr, cmprsk R package) as it is possible in Cox regression?
If you have access to Stata there is a nice user-written program that prepares the data for you perform the Fine-Gray model. The program is called stcrprep and can be downloaded by typing: ssc install stcrprep
That said, this command is based on the R function, crprep written by Ronald Geskus, which is part of the mstate package. So if you want to stay within R, this may be something for you to investigate.
I hope this helps
Ariel
Hi. I have the same question. I investigated the crprep function, but that seems to be a function to convert datasets from short to long format. In itself this does not facilitate time-dependent variables in the Fine-Gray model.
Time-dependent variables is not possible using cmprsk/crr. Is there other packages that can do this?
It seems that the inclusion of time-dependent covariates in the Fine and Gray model leads to biased results (Latouche A., Porcger R. & Chevret S. (2005) and Putter H., Fiocco R. and Geskus R. (2007)).
- Badges
- Science topic
- Similar topics
- Force