topic Re: How to show the 95% CI of hazard ratios of PROC PHREG? in Statistical Procedures

How to show the 95% CI of hazard ratios of PROC PHREG?

TomHsiung — Thu, 25 Jan 2018 08:43:38 GMT

By default, the PROC PHREG procedure results in a fixed value of hazard ratio, like in the screenshot below.

PS: The confidence intervals of "Parameter Estimate" and "Hazard Ratio" were both missing. Based on the theory behind Cox proportional hazard model, I need the 95% CI.

Tom

Re: How to show the 95% CI of hazard ratios of PROC PHREG?

TomHsiung — Thu, 25 Jan 2018 14:27:04 GMT

I searched the Internet and found the code of

/ rl

The complete code I used to make the Cox proportional hazard regression analysis is

proc phreg data=work.d200;
model Time_to_first_therapeutic_INR__d*First_therapeutic_INR_reached_(0)=age__y_ BSA gender_code af
hypertension chf VAR94 AKI_for_T_test Potential_amiodarone_DDI
T_test___indication AF_and_Warfarin_History
var33 var34 / rl selection=stepwise SLE=0.05 SLS=0.20;
assess ph / resample;
run;

Please pay attention to the last syntax of

assess ph / resample;

This code resulted in following output, what is that? Someone knows? This graph is interesting but it did not seem to like a goodness-of-fit test.

Re: How to show the 95% CI of hazard ratios of PROC PHREG?

Reeza — Thu, 25 Jan 2018 15:50:50 GMT

ASSESS Statement

ASSESS <VAR=(list)> <PH> </ options>;

The ASSESS statement performs the graphical and numerical methods of Lin, Wei, and Ying (1993) for checking the adequacy of the Cox regression model. The methods are derived from cumulative sums of martingale residuals over follow-up times or covariate values. You can assess the functional form of a covariate or you can check the proportional hazards assumption for each covariate in the Cox model. PROC PHREG uses ODS Graphics for the graphical displays. You must specify at least one of the following options to create an analysis.

http://documentation.sas.com/?docsetId=statug&docsetTarget=statug_phreg_syntax02.htm&docsetVersion=14.3&locale=en

Re: How to show the 95% CI of hazard ratios of PROC PHREG?

ballardw — Thu, 25 Jan 2018 20:23:16 GMT

Since you are looking for hazard ratios and confidence intervals is there any reason you aren't using a HAZARDRATIO statement? The option CL= on the Hazardratio statement allows requesting Wald or profile-likelihood limits or both.