Data handling for Cochrane-Armitage trend test

[Eve]

Hi,

I currently work on a new project where I was requested to perform a Cochrane Armitage trend test across adverse events severity grades between two arms.

My data consists of variables:

Category, MedDRA_Term, arm, sev1, sev2, sev3, sev4 ;

where sev1-sev4 correspond to ae_severity and include the number of subjects (please refer to attached document)

Does anybody know how I can edit my dataset in order to perform the test? I would need to combine sev1-sev4 into one variable to use the code below?

proc freq
data=mydata;

tables arm*severity / trend measures cl

exact trend / maxtime=60;

run;

I would really appreciate any input from you!

Many thanks,

Eve

[Haikuo]

For your data on hand, you can choose to put everything into one dataset and subset it on the fly or you can subset upfront by disease. So for example, the subset or individual table should look like (Anemia):

arm	severity	counts
1	1	2
1	2	1
1	3	0
1	4	0
2	1	0
2	2	2
2	3	1
2	4	0

Then the following code (minor modification comparing to yours) can then be run:

PROC FREQ DATA = MYDATA

     ORDER=INTERNAL

;

     WEIGHT counts;

     TABLES arm * severity /

           MEASURES

           TREND

           SCORES=TABLE

      CI

           ALPHA=0.05;

     EXACT MEASURES TREND  /    MAXTIME=60

      ;

  RUN;

Good luck,

Haikuo

[Eve]

Thank you very much Haiko!

Eve