C statistics from fitting cox model using PROC PHREG

NicoO_neilly — Tue, 04 Aug 2015 17:22:34 GMT

Hello all,

I have a problem and would really appreciate your input. So here it is:

Context: c-statistics from cox regression analysis

Samples: 16,942 participants below age 75 years old

Events (death): 1,116

Variables: idn, duration, age, sex, censory

Syntax Syntax used is adapted from SAS Global Forum Paper (1): Liu L, Forman S, Barton B. Fitting Cox model using PROC PHREG and beyond in SAS. SAS Glob Forum [Internet]. 2009 [cited 2015 Aug 4]; Available from: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.176.1835&rep=rep1&type=pdf

Syntax:

proc phreg data=Below75allsubjectiveobjective;

id idn;

model duration*censory(0)= age sex;

output out=obs survival=surv;

run;

data evtset; set obs; if censory=1; rename idn=idn_j surv=y_j duration=x_j;

keep idn surv duration; run;

proc sql;

create table allset as

select idn_j, y_j, x_j, idn as idn_i, surv as y_i, duration as x_i

from evtset, obs

where idn_j<>idn;

quit;

data concord;

set allset;

if (x_i<x_j and y_i>y_j) or (x_i>x_j and y_i<y_j) then concord=1;

else concord=0;

run;

data _null_;

set concord end=eof;

retain nch ndh;

if _N_=1 then do;

nch=0;

ndh=0;

end;

if concord=1 then nch+1;

if concord=0 then ndh+1;

if eof=1 then do;

call symput('ch',trim(left(nch)));

call symput('dh',trim(left(ndh)));

call symput('uspairs',trim(left(_n_)));

end;

run;

data _null_;

set Below75allsubjectiveobjective end=eof;

if eof=1 then call symput('totobs',trim(left(_n_)));

run;

%put &ch &dh &uspairs &totobs;

data calculat;

ch=input("&ch",12.0);

dh=input("&dh",12.0);

uspairs=input("&uspairs",12.0);

totobs=input("&totobs",10.0);

pc=ch/(totobs*(totobs-1));

pd=dh/(totobs*(totobs-1));

c_hat=pc/(pc+pd);

w=(2*1.96**2)/(totobs*(pc+pd));

low_ci_w=((w+2*c_hat)/(2*(1+w)))-(sqrt((w**2+4*w*c_hat*(1-

c_hat))/(2*(1+w))));

upper_ci_w=((w+2*c_hat)/(2*(1+w)))+(sqrt((w**2+4*w*c_hat*(1-

c_hat))/(2*(1+w))));

run;

This syntax runs smoothly without any errors.

Output when adding different covariates:

Covariates	Ch	Dh	Uspairs	Totobs	pc	pd	c_hat	w	low_ci_w	upper_ci_w
None	18752674	153482	18906156	16942	0.065	0.001	0.992	0.007	0.977	1.000
Sex	16284756	2621400	18906156	16942	0.057	0.009	0.861	0.007	0.818	0.899
Age, Sex	12821742	6084414	18906156	16942	0.045	0.021	0.678	0.007	0.622	0.732

C-hat from the output is the c-statistics calculated, is that correct?

Problem: Why is C-statistics decreasing when adding more covariates? How do I interpret the c-statistics results?

Thank you in advance

Re: C statistics from fitting cox model using PROC PHREG

NicoO_neilly — Tue, 04 Aug 2015 21:06:05 GMT

Can anyone please explain the output to me? What is pc, ph etc? Or maybe link me to a decent guide?

Also, shouldn't the fourth line of syntax should be 'duration*censory(1)' because censory=1 is death, which is what we're interested in for cox regression?

Re: C statistics from fitting cox model using PROC PHREG

samarda — Wed, 12 Sep 2018 08:02:53 GMT

were you able to figure out what the problem was? i am having the same problem

topic Re: C statistics from fitting cox model using PROC PHREG in Statistical Procedures

C statistics from fitting cox model using PROC PHREG

Re: C statistics from fitting cox model using PROC PHREG

Re: C statistics from fitting cox model using PROC PHREG