Hi! I am running a test for bioequivalence for a three-sequence, three-period study (Ref, Test 1, Test 2). In this study, there was a lot of drop-out, and many subjects only received the drug during the first period. I would like to test bioequivalence for test 1 vs. Reference, and for that I would like to exclude these individuals that didn’t receive both Test 1 and Ref. Is there a code I can add in Phoenix to run the bioequivalence only the individuals that received both Test 1 and Reference (ie exclude individuals with missing data from the calculation of geometric LS mean and ratios)? Right now, I am running the NCA on the whole study, then manually delete the results from individuals that didn’t received both Test 1 and Reference from the pivoted PK parameters table. Thank you! Alex
have you tried filtering (held within the DataWizard tool from 6.3 onwards?) You may need/want to use/ set a flag first to make it easier to filter out the non-completers Simon.
Hi Alex, Simon is right that you can’t do this directly in Bioeq, and would need to build a workflow. We’ve been doing something similar while working on workflows that can do reference-scaled average bioeq, so you could do something like: Filter to keep only Ref and Test1 data. Send that result to Split Worksheet, using Subject as Sort, and check the box to “Create unique value worksheet” (this will give you a worksheet with a count of administrations per subject, either 1 or 2, since Test2 was already filtered out). Filter to keep subjects with 2 administrations. Merge this last result with the original data where the merge is done on Subject. Hope that helps, Linda
I think there is an example of what Linda is describing in the project that Helmut posted here; Certara | Drug Development Solutions Simon
Thank you Simon and Linda! 
This is very helpful! I’ll try filtering the data as proposed! Thank you so much! Alex