Bioequivalence and Bioavailability Forum

Dear All

We completed a reference replicate design (3 WC) for FDA submission


while calculating the Geo mean ratio in SABE we had a confusion in using the SAS codes as below

we calculated the GMR as per the below code

pointest=exp(estimate);

poinREF=exp(estimate);

data dglmno;
set dglmn;
DIFF= (pointest-poinREF)**2;
theta=((log(1.25))/0.25)**2;
y=-theta*s2wr;
boundy=y*dfd/cinv(0.95,dfd);
ucl= DIFF- boundy;
lcl= DIFF+ boundy;
sWR=sqrt(s2wr);
critbound=(x+y)+sqrt(((boundx-x)**2)+((boundy-y)**2));
Geomean_ratio= pointest/ poinREF;


But other statistician used diffrent code as below

pointest=exp(estimate);

data dglmno;
set dglmn;
theta=((log(1.25))/0.25)**2;
y=-theta*s2wr;
boundy=y*dfd/cinv(0.95,dfd);
sWR=sqrt(s2wr);
critbound=(x+y)+sqrt(((boundx-x)**2)+((boundy-y)**2));
Geomean_ratio= pointest;


So kinldy help me out in differntiating in which one is the correct way to calculate the GMR

Thanks

Dinesh

---

Edit: Category changed. [Helmut]