I'm working on an extended Cox proportional hazards regression model which is predicting a recurrent event, inpatient hospitalization, during a 2 year period and will include multiple time-varying covariates (at least 10).

My understanding is that the counting process method incorporates multiple records per subject, with each record corresponding to an interval of time during which all variables (including time-varying variables) remain constant and each interval either ends in the event of interest or is right censored.

Is there a limit on the number of time-varying covariates which can be reasonably included in a Cox regression model? I'm wary of sacrificing power by slicing up my time period into too many intervals.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.