I'm learning about time-dependent coefficients in R using the material below. https://cran.r-project.org/web/packages/survival/vignettes/timedep.pdf

In 3.4 PBC data, the code below correctly assumes the hazard associated with changing bili and protime values.

fit2 <- coxph(Surv(tstart, tstop, death==2) ~ log(bili) + log(protime), pbc2)

My question is, how can you expand this analysis to predict survival after liver transplant using pre-transplant bilirubin and protime values? (Not comparing transplant vs no transplant (medical treatment) as explained in the YouTube video: https://www.youtube.com/watch?v=Lbqdvn7YHh4)

This is not compatible with pbc/pbcseq database, but assume all patients who died in pbc/pbcseq database receive liver transplant at death date (status = 0 or 1 only), and we also have post-liver transplant survival data, as in jasa dataset depicted in 3.3.

Is it possible with R?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.