在使用R进行的生存分析中，考克斯比例危害模型中surv函数的目的是什么？

Question

I am currently looking at a document that states to use the Cox Proportional Hazards model, your response variable for the formula portion of

coxph(formula, data=, weights, subset, 
      na.action, init, control, 
      ties=c("efron","breslow","exact"), 
      singular.ok=TRUE, robust=FALSE, 
      model=FALSE, x=FALSE, y=TRUE, tt, method, ...)

must surv() in the formula portion.

Can someone tell me what the surv() function does? I understand it states it is a survival object, but I am not sure if it is something that is necessarily required. Thanks!

Answer 1

This is a case where you need to just read the documentation and run the examples therein. The first example in ? coxph ? coxph shows the following:

# Create the simplest test data set 
test1 <- list(time=c(4,3,1,1,2,2,3), 
              status=c(1,1,1,0,1,1,0), 
              x=c(0,2,1,1,1,0,0), 
              sex=c(0,0,0,0,1,1,1)) 
# Fit a stratified model 
coxph(Surv(time, status) ~ x + strata(sex), test1) 

Clearly, you need to have the lefthandside/response part of the formula be the output from Surv (which also has clear documentation that you can read; see ?Surv ). If you take a look at that object:

> str(Surv(test1$time,test1$status))
 Surv [1:7, 1:2] 4  3  1  1+ 2  2  3+
 - attr(*, "dimnames")=List of 2
  ..$ : NULL
  ..$ : chr [1:2] "time" "status"
 - attr(*, "type")= chr "right"

And see how it reflects the information contained in the time and status columns:

> with(test1, cbind.data.frame(time, status, Surv(time,status)))
  time status Surv(time, status)
1    4      1                 4 
2    3      1                 3 
3    1      1                 1 
4    1      0                 1+
5    2      1                 2 
6    2      1                 2 
7    3      0                 3+

Then, to answer your question about whether it is necessary, you can just try running coxph without it and see what happens:

> coxph(time ~ x + strata(sex), test1) 
Error in coxph(time ~ x + strata(sex), test1) : 
  Response must be a survival object

Answer 2

Surv() is a function to create the survival object. For survival analysis you need the follow-up time (or time intervals in case of time-dependent variables) and the status of the individual. Obviously, it is necessarily required.

You should read the survival package documentation first. I also suggest that you read this very well explained book about survival analysis: Survival Analysis: A Self-Learning Text