Create a Kaplan-Meier cumulative incidence curve by strata

[Reeza]

Use ODS TRACE to see the table/graph names to ensure your ODS statement is correct.

Also what does 'it doesn't work' mean? That could be anything from a syntax error to SAS crashing.

[ANKH1]

There was no output.

[ANKH1]

I used the following:

proc lifetest data=sample plots=survival(f)
outsurv=survival_data timelist=(0 to 200 by 12) reduceout;
strata group;
time time2event*censor(1);
run;

I get the dataset to be use in proc sgplot but I am not sure what to put in the Y axis. The examples I've seen are with "Survival" not like the graph for cumulative incidence. I tried to use Failure but it says that the variable does not exist.

[Reeza]

That outputs the raw data, you'll have to re-do the calculations. You were on the right track trying to get the plot data instead, but you have the wrong graph name - it's a failure plot not survival plot. Use ODS TRACE, check the log to get the correct ODS name.

https://blogs.sas.com/content/iml/2017/01/09/ods-output-any-statistic.html

[Reeza]

Why? What did the log say?

[ANKH1]

with this code:

proc sgplot data=survival_data;
step x=time2event y=survival/group=stratum name="Percentage of patients reporting the first event";
xaxis values=(0 to 24 by 2) valueshint label="Time to event (Months)";
yaxis offsetmin=0.02 min=0 offsetmax=0.1 max=1.0 label="Percentage of patients reporting the First Event";
run;

I get this output:

[Reeza]

Correct, that is the survival plot, not the failure plot. The plots=survival(f) -> the f option requests failure which is what you're looking for instead (IMO).

[ANKH1]

Yes, you are right, the failure plot is what we need. I used ODS TRACE ON as you advised and got the right name:

ods trace on;
ods output failureplot=sp;
proc lifetest data=sample plots=survival(f)
outsurv=survival_data timelist=(0 to 24 by 2) reduceout;
strata group;
time time2event*censor(1);
run;

proc sgplot data=sp;
step x=time y=_1_survival_/group=stratum name="Percentage of patients reporting the first event";
xaxis values=(0 to 24 by 2) valueshint label="Time to event (Months)";
yaxis offsetmin=0.02 min=0 offsetmax=0.1 max=1.0 label="Percentage of patients reporting the First Event";
run;

and got this:

Looks ok right? I used variable _1_survival_ instead of survival. I have a question if we want to show the censored, how will you add it? 

[Reeza]

https://blogs.sas.com/content/graphicallyspeaking/2018/02/19/survival-plot-twist-using-sgplot-proced...

[ANKH1]

Thanks for sharing. What is the difference between these two statements?:

scatter x=time y=censored / markerattrs=(symbol=plus) name='c';
scatter x=time y=censored / markerattrs=(symbol=plus) GROUP=stratum;

And since this is a failure plot, should I use _1_censored_ instead of censored?

proc sgplot data=sp;
step x=time y=_1_survival_/group=stratum name="Percentage of patients reporting the first event";
xaxis values=(0 to 24 by 2) valueshint label="Time to event (Months)";
yaxis offsetmin=0.02 min=0 offsetmax=0.1 max=1.0 label="Percentage of patients reporting the First Event";
scatter x=time y=_1_censored_ / markerattrs=(symbol=plus) name='censored';
scatter x=time y=_1_censored_ / markerattrs=(symbol=plus) GROUP=stratum;
run;

This is the figure.