I am analyzing a dataset of recurring events for a sample of patients. Each interval is categorized as ending in a hospitalization or not a hospitalization. I am looking to define the final interval of each patient. I understand that if the patient died, the final row would end with the date of their death. If the patient survives, would the final row end with their last recorded event or proceed to the last date their chart was checked?
$\begingroup$ $\endgroup$
1 
- $\begingroup$ Could you please edit the question to explain the nature of your data a bit more? I usually think of an Andersen-Gill model in terms of modeling event times in continuous time, but from the question it seems that you are working with time intervals like in a discrete-time survival model. Also, it's not clear how much of a problem having death as a competing event type might pose. Showing some samples of your data could help clarify. $\endgroup$EdM– EdM2025-11-19 15:50:26 +00:00Commented Nov 19 at 15:50
Add a comment |