I am doing survival analysis in R and I want to perform a log-rank test to detect differences between survival curves with a time-dependent covariate.
I use:
fit_seasons<- survfit(Surv(time1, time2, status) ~ season,data=dados7)
summary(fit_seasons)
to compute the Kaplan-Meier estimated survival.
The survdiff() function in the survival package does not accept models with time-dependent covariates. And since I am using the counting process, ictest() will not work.
What is an alternative way to do this analysis without using the coxph() function?
