Re: [R] SAS and R code hazard ratios

From: Thomas Lumley <>
Date: Wed 10 Jan 2007 - 20:50:30 GMT

On Wed, 10 Jan 2007, wrote:
> I am new to R and have been comparing CPH survival analysis hazard ratios
> between R and SAS PhReg. The binary covariates' HRs are the same, however
> the continuous variables, for example age, have quite different HRs
> although in the same direction. SAS PhReg produces HRs which are the
> change in risk for every one increment change in the independent variable.
> How do I interpret the HRs produced by R? Thanks much, C

What function did you use to fit the model in R? If you used coxph(), in the survival package then you should get the same answers as SAS. If you used cph() in the design package then the output says what the increment is that correponds to the quoted hazard ratio, and the default is the interquartile range.

         -thomas mailing list PLEASE do read the posting guide and provide commented, minimal, self-contained, reproducible code. Received on Thu Jan 11 07:54:58 2007

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Wed 10 Jan 2007 - 22:30:27 GMT.

Mailing list information is available at Please read the posting guide before posting to the list.