Question regarding cox proportional hazard model


New Member

I have a few questions regarding how to handle the situation when performing Cox proportional hazard model (either deleting cases or giving day 1 to this case).

The situation is either the outcome of interest occurs at index date (start of follow-up) or the outcome of interest and time-varying exposure occurs on the same day. Each subject is followed until the outcome of interest or censoring event. Do you think it would be appropriate if this case should be deleted or if the outcome of interest is considered on day 1?

Please let me know.



Less is more. Stay pure. Stay poor.
Well when do other obs have the event? Cursory, I don't know why having the event on day one would be an issue. Can you post your KM or survival plot?