AE Summary Tables by SOC

Adverse Events Summary is one of the important tables generated in clincal trials industry.

Let us see how we could construct an AE table..

Important Points to remember:
1. AE Verbatim terms are to be coded with standard medical dictionaries, MeDRa.
2.SOC Name and Pt name are important variables that would be used for developing table.
3.Each Patient or Unique Patient Id is Counted only once for each of the SOC term and once for the Pt Name in each SOC term.

Code:

proc sort data=ae out=ae1;
by rand site ptid socname;
run;

proc sort data=ae out=ae2;
by rand site ptid socname ptname;
run;


proc freq data=ae1;
tables socname/out=ae1;
by rand;
run;

proc freq data=ae2;
tables socname*ptname/out=ae2;
by rand;
run;

Now set the datasets ae1 (with counts for soc term) and ae2 (with the counts for soc term *pt term) by soc term.

you would have the counts in each of the trt group(rand) for the soc term and pt term. Transpose the datasets or set accordingly to result in the counts for each of the trt group. Percents are calculated by using denominator N from the number of Patients enrolled(Like Demographics Dataset for ITT) and N from the number of Patients intended to treat from Randomization dataset for the safety analysis.