Bioequivalence and Bioavailability Forum

Hi J,

❝ 2) or I should remove treatment A and B data from the dataset so that it only contains C and D data before running proc mix?

This is the correct one according to the questions and answers of the EMA's BE guideline (Helmut kept a copy online) .

❝ I also got an advice from a stat fellow who told me to convert the study into a 2-way 2-sequence study before running proc mixed(????). What he meant was remove treatment A and B data and also convert the 4 sequences - ABDC, BCAD, CDBA, DACB, by removing the A + B and leave only C + D so to end up with 2 sequences CD and DC.

Yes and no. For C vs. D, yes to remove A and B from the whole data set before the analysis, but the original values of treatment, sequence, period etc should be maintained according to the same Q&A mentioned above, i.e., you'll still have treatment of C and D only, 4 sequences of ABCD/DCBA (etc), 4 periods (1/2/3/4) in your data. In EU, you can use PROC GLM; in US, I'd say that you can use it too but I'm not sure since sometimes they prefer PROC MIXED.