I am at the beginning of setting up a survival analysis in R. I took a look in this book here: https://www.powells.com/book/modeling-survival-data-9780387987842/ but struggle to properly set the data up in the first place. So this is a very basic question to survival analysis as I can not find a good example online.
I'd to understand how to incorporate the consorized data into my surv() function. I understand the inputs in surv are:
- 0 = right censored
- 1 = event
- 2 = left censored
- 3 = interval censored
Right Censored: The time of study ends before an event takes place (ob1)
Left Censored: The event has already happend before the study starts
Event: Typically, death or some other form of expected outcome (marked by x)
Intervall Censored: The observation starts at some point in the study and has an event / drops out before end of study (ob5)
Left truncated: Ob 3,4,5 are left truncated
To better understand what I am talking about I sketched the described types of censored data below:
"o" marks beginning of data / first occurance in data set
"x" marks event
Start of study End of observation
ob1 o-|-----------------------------------------------------------|--------
| |
ob2 o-|-------------------------------xo |
| |
ob3 | o-----------------------------------xo |
| |
ob4 | o------------------x-|----------o
| |
ob5 | o----------------------------o |
|--------------------------------------------------------------
1999 2010
Finally, what would i like to know:
- Did I classify ob1- ob5 correctly?
- How about the other types of observations?
- How do I represent these as input for the surv function? If for example right censored is true, i.e. the study ends how does a "0" indicate so? What is the input for the time series when neither event(1) nor end of observation occur (0)? what happens at a time when "nothing" happens?
- When and how is the interval censored data marked? 3 for beginning and end?
I can provide some sample code if needed.
Again, thank you for your help on this and valuable questions!
