[R] Multiple events Cox's model and proportional hazards

[natalie.vanzuydam]

Hi,

I am using the survival package to perform a Cox's regression analysis on
multiple events of myocardial infarctions.  I have been using the Andersen
and Gill model: 
coxph(Surv(time1,time2,status)~factor(treatment)+age+sex+cluster(id).

I was just wondering if this model should satisfy proportional hazards
assumptions.  I have run  the cox.zph function and the age parameter
violates the proportional hazards?  What would be the best way to construct
this model.  Should I include time dependent covariates?

Thanks,
Natalie

-----
Natalie Van Zuydam

PhD Student
University of Dundee
nvanzuydam at dundee.ac.uk
--
View this message in context: http://r.789695.n4.nabble.com/Multiple-events-Cox-s-model-and-proportional-hazards-tp3783031p3783031.html
Sent from the R help mailing list archive at Nabble.com.