topic Re: Adjusted cumulative incidence curve with proc phreg in Statistical Procedures

Adjusted cumulative incidence curve with proc phreg

kayeee_ — Tue, 19 Mar 2024 07:00:42 GMT

Dear all,

I am struggling with drawing adjusted cumulative incidence curve in proc phreg.

While searching, I found a paper that mentioned using the following SAS macro called %CIFCOX. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3377442/)

But I couldn't find it on the web site.

If anyone knows a code or macro that can draw such graphs, could you please share it?

Many thanks in advance.

Re: Adjusted cumulative incidence curve with proc phreg

ballardw — Tue, 19 Mar 2024 14:26:17 GMT

The paper in the link references a web page that is no longer active. Due to the age of the article, 2011, I would suggest that you provide us some example data in the form of a data step, dummy data is fine as long as it behaves like your actual data, and the code you are using for the analysis.

13 years is long enough to have gone through several updates to SAS and many of the updates have involved graphic options. So the macro may very well be duplicating what may be done easier these days.

Instructions here: https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-data-AKA-generate/ta-p/258712 will show how to turn an existing SAS data set into data step code that can be pasted into a forum code box using the </> icon or attached as text to show exactly what you have and that we can test code against.

Re: Adjusted cumulative incidence curve with proc phreg

kayeee_ — Wed, 20 Mar 2024 02:19:51 GMT

Thanks for your advice.

I also, found the macro of adjusted survival curve. and It has CIF and SREVERSE options to y-axis: 1-Survival.

But It is not possible to use in adjusted model.

https://communities.sas.com/t5/SAS-Communities-Library/Kaplan-Meier-Survival-Plotting-Macro-NEWSURV/ta-p/479747/page/2/show-comments/true

Here is the sample data below:

proc format;
   value grpLabel 1='ALL' 2='AML low risk' 3='AML high risk';
run;
 
data BMT;
        input DIAGNOSIS Ftime Status Gender education@@;
        label Ftime="Days";
        format Diagnosis grpLabel.;
datalines;
1	2081	0	1	2	3	677	2	1	1
1	1496	0	1	0	3	168	2	0	2
1	1433	0	1	2	3	16	2	0	1
1	1330	0	1	2	3	625	1	0	0
1	226	0	0	1	3	273	1	1	0
1	1111	0	1	1	3	76	1	1	0
1	1182	0	0	2	3	363	2	1	2
1	418	2	1	2	1	1602	0	1	0
1	276	2	0	1	1	1462	0	0	2
1	609	1	1	1	1	1377	0	1	0
1	487	2	1	2	1	996	0	1	2
1	194	2	0	1	1	1199	0	1	2
1	526	2	1	0	1	530	0	1	2
1	129	1	0	2	1	1167	0	0	1
1	122	1	0	1	1	383	1	1	2
1	466	2	1	0	1	104	1	1	2
1	109	1	1	0	1	172	2	0	1
1	1	2	1	1	1	662	1	1	0
1	110	1	0	0	1	230	1	0	0
2	2569	0	1	2	1	122	2	1	0
2	2409	0	1	1	1	74	1	1	1
2	1857	0	0	1	1	86	2	1	1
2	1562	0	1	1	1	192	1	1	2
2	1363	0	1	0	1	55	1	0	0
2	860	0	0	2	1	107	2	1	2
2	2246	0	0	1	1	332	2	1	0
2	1799	0	1	0	2	2506	0	1	2
2	1674	0	1	0	2	2218	0	1	1
2	1527	0	0	0	2	1829	0	1	0
2	957	0	1	1	2	1470	0	1	1
2	847	0	1	2	2	1030	0	0	1
2	1850	0	0	2	2	1258	0	0	1
2	1535	0	0	0	2	1870	0	0	2
2	1384	0	0	0	2	1709	0	0	0
2	2204	2	0	0	2	1568	0	1	0
2	481	2	1	2	2	1324	0	1	1
2	641	2	1	1	2	932	0	0	0
2	288	2	1	0	2	848	0	1	2
2	79	2	0	1	2	1843	0	0	1
2	486	1	0	1	2	1447	0	0	0
2	272	1	0	1	2	414	2	1	0
2	381	1	0	2	2	1063	2	1	2
2	53	2	0	2	2	105	2	1	2
2	35	2	0	2	2	390	2	1	0
2	704	2	0	0	2	421	1	1	1
2	219	1	1	0	2	748	1	1	0
3	2640	0	1	2	2	48	2	0	2
3	2252	0	1	2	2	1074	2	1	1
3	2133	0	0	2	2	10	2	1	1
3	1631	0	1	1	2	80	2	0	0
3	1345	0	1	2	2	248	1	1	0
3	845	0	0	2	2	211	1	1	2
3	162	2	1	2	2	606	1	1	1
3	100	1	1	2	3	2430	0	1	1
3	47	1	1	0	3	2140	0	1	0
3	456	1	1	0	3	1238	0	1	2
3	318	2	0	2	3	2024	0	0	2
3	467	1	0	2	3	1136	0	1	2
3	390	1	1	1	3	422	1	0	2
3	105	2	1	2	3	84	1	0	1
3	164	2	0	0	3	2	2	1	1
3	120	1	0	1	3	242	1	1	1
3	64	1	0	1	3	268	1	0	2
3	74	2	0	0	3	32	1	1	2
3	157	1	0	1	3	47	1	1	1
3	48	1	0	0	3	183	2	0	0
3	63	2	1	2	3	115	1	0	1
3	113	1	0	1	3	93	1	0	2
3	80	2	1	0					

;
run;


proc phreg data=bmt;
class DIAGNOSIS(ref='1') Gender education;
Model Ftime*Status(0)=DIAGNOSIS Gender education; /*gender and education are covariates*/
run;

Re: Adjusted cumulative incidence curve with proc phreg

Ksharp — Wed, 20 Mar 2024 05:25:23 GMT

Check the example in Doc.

SAS Help Center: Nonparametric Analysis of Competing-Risks Data

proc lifetest data=sashelp.bmt   plots=cif ;
time T*Status(0)/eventcode=1; 
strata group;
run;

Re: Adjusted cumulative incidence curve with proc phreg

Ksharp — Wed, 20 Mar 2024 05:36:32 GMT

And if you are using PROC PHREG,
Check this out:

https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/statug/statug_phreg_examples15.htm

Re: Adjusted cumulative incidence curve with proc phreg

acozzilepri — Mon, 25 Mar 2024 20:04:35 GMT

This ?

https://github.com/rsparapa/rasmacro/blob/master/_cifstrata.sas