topic Re: Competing risk analysis when proportional hazards assumption is not met in Statistical Procedures

Competing risk analysis when proportional hazards assumption is not met

kc — Mon, 18 Aug 2025 03:48:59 GMT

I am using Fine-Gray subdistribution hazard model to analyze time to event in the presence of competing risks. But, it assumes proportional hazards for the subdistribution hazard and one of the important covariates doesn't meet this assumption. I am wondering how do I go about addressing this issue. Alternatively, to address the non-proportionality, we can use the accelerated failure time models (AFT). But, how do you adjust for competing risk in these models?

Any help (with code) would be greatly appreciated!

Thanks in advance.

Re: Competing risk analysis when proportional hazards assumption is not met

Ksharp — Mon, 18 Aug 2025 05:27:07 GMT

https://communities.sas.com/t5/Statistical-Procedures/Dealing-with-nonproportionality-in-a-Cox-PH-Model/m-p/968428#M48690

Re: Competing risk analysis when proportional hazards assumption is not met

kc — Mon, 18 Aug 2025 12:41:38 GMT

Sorry if I am missing anything but this post/embedded links do not talk about the scenario where competing risks and non-proportionality are involved.