How to drew survival curves by SAS based on Survival Probability

[Mingming1992]

Hello all, I was hoping to find some help regarding this dataset. I want to drew survival curves by SAS based on it but I do not know how to do it. I am sure this would be a simple resolution for someone with more experience utilizing SAS. Any help would be very much appreciated. Thank you!

Here is the dataset:(Number in the column blow each surv means the Survival Probability on these time (days).

time	surv1	surv2	surv3	surv4
0	1	1	1	1
0	1	0.993631	1	0.965475
1	1	0.980955	1	0.954309
2	1	0.968365	1	0.941546
3	1	0.962017	1	0.931951
4	0.990686	0.962017	0.989159	0.927144
5	0.990686	0.962017	0.989159	0.922343
6	0.981322	0.955658	0.957124	0.908054
7	0.981322	0.955658	0.946394	0.900084
…………
345	0.612125	0.774457	0.574568	0.592467
350	0.612125	0.774457	0.574568	0.590918
355	0.612125	0.774457	0.574568	0.589368
356	0.612125	0.767991	0.574568	0.589368
360	0.612125	0.761524	0.574568	0.589368

[FreelanceReinh · Accepted solution]

Hello @Mingming1992,

I would start with a basic step plot:

proc sgplot data=have;
step x=time y=surv1;
step x=time y=surv2;
step x=time y=surv3;
step x=time y=surv4;
run;

Next, there are a lot of options to modify the plot (like line types and colors, axis labels, tick marks, the legend, etc.).

[Mingming1992]

Thank you FreelanceReinhard. It helps a lot.

[Norman21]

Following on from the idea of FreelanceReinhard and using the code from one of Sanjay Matange's blogs (https://blogs.sas.com/content/graphicallyspeaking/2018/02/19/survival-plot-twist-using-sgplot-proced...) you can produce something like this:

title 'Product-Limit Survival Estimates';
footnote j=l h=6pt italic 'This visual is for discussion of graph features only.'
' The actual details should be customized by user to suit their application.';

data my;
input time	surv1	surv2	surv3	surv4  ;
label 
      time='Survival Time (days)'
      surv1='Group 1'
      surv2='Group 2'
      surv3='Group 3'
      surv4='Group 4'
	  ;
cards;
0 1 1 1 1
0 1 0.993631 1 0.965475
1 1 0.980955 1 0.954309
2 1 0.968365 1 0.941546
3 1 0.962017 1 0.931951
4 0.990686 0.962017 0.989159 0.927144
5 0.990686 0.962017 0.989159 0.922343
6 0.981322 0.955658 0.957124 0.908054
7 0.981322 0.955658 0.946394 0.900084
345 0.612125 0.774457 0.574568 0.592467
350 0.612125 0.774457 0.574568 0.590918
355 0.612125 0.774457 0.574568 0.589368
356 0.612125 0.767991 0.574568 0.589368
360 0.612125 0.761524 0.574568 0.589368
;

proc sgplot data=my noborder;
    step x=time y=surv1 / ;
    step x=time y=surv2 / ;
    step x=time y=surv3 / ;
    step x=time y=surv4 / ;
run;

[Mingming1992]

Thank you Norman21. It helps a lot.

[ballardw]

If you use one of the several SAS procedures such as Lifereg, Lifetest or Phreg to create the survival probabilities from "raw" data, they have a number of Plot options available.