topic Re: Proc PHREG and competing risks in Statistical Procedures

Proc PHREG and competing risks

n6 — Tue, 12 May 2009 19:35:54 GMT

I am doing a competing risks survival model, which I'm new at but I don't think it is so new that it shouldn't be in a SAS Proc. And yet I can't seem to find anything about it in my SAS Help. The idea is that it's a survival model but instead of just death being an outcome, there is also another event that can take people out of the study. That normally would just be a censoring issue, but in this case this other event is related to death and so when people get close to death they are less likely to experience death while in the study because they're more likely to be removed from the study.

Make sense? Hopefully. But anyway, the main question is, can SAS do this? Surely, it must be able to. Okay, I'll stop calling you Shirley. But in the meantime, any info is appreciated.

Re: Proc PHREG and competing risks

plf515 — Wed, 13 May 2009 02:00:06 GMT

See

Extending the use of PROC PHREG in survival analysis

by Christopher Ake and Art Carpenter,

presented at various user groups.

Re: Proc PHREG and competing risks

OsoGris — Wed, 13 May 2009 02:52:32 GMT

Allison(1995) has an entire chapter (Chapter 6) on competing risks. Anyone doing survival analysis in SAS should have this book.

reference:
Allison, Paul D.(1995) Survival Analysis Using the SAS System - A Practical Guide, SAS Institute Inc., Cary, NC

The basic idea is to model each risk separately, censoring on the other exits.

Re: Proc PHREG and competing risks

n6 — Wed, 13 May 2009 19:35:18 GMT

Thanks a lot, folks, that was very useful info. I appreciate it.

Re: Proc PHREG and competing risks

r123 — Fri, 24 Sep 2010 00:13:04 GMT

I also want to do competing risks survival analysis. Is the "Extending the use of PROC PHREG in survival analysis" talk available online?

Re: Proc PHREG and competing risks

Dale — Fri, 24 Sep 2010 17:04:42 GMT

Well, I just Googled "ake carpenter phreg" (without the quotes - the quotes here serve only to indicate the terms that were in the Google search). Low and behold, right on top was a a paper with the title "Extending the use of PROC PHREG in survival analysis" which was presented at a SUG.

Re: Proc PHREG and competing risks

mycotropic — Wed, 31 Aug 2011 20:48:56 GMT

OsoGris and N6; The Allison method of handling competing risks must done with care. If you violate the assumption of non-informative exits and cannot FULLY control for it with covariates then his approach is not appropriate. For more recent work on this problem see this talk by Pintilie; www.stat.ucl.ac.be/edt/20082009/Pintilie/CR.pdf

After spending weeks running and then writting up the Allison approach only to have it tanked by my committee believe me I know it's an issue and an important one.