Log in
RegisterReplicate Design/Scaled ABE approach [Design Issues]
Dear joyjac!
Assuming suitable software available (e.g., WinNonlin ≥4.x), it’s possible to obtain the standard average BE evaluation from any replicate design as well.
The point estimate is simply the same, with the additional information of the CI.
I would not 'throw away' half of the data obtained in the study.
IMHO this should be possible.
regards,
Hermann Rotter
❝ Can we revert to the average BE approach if within subject variances obtained from a replicate design/scaled ABE (TRTR) show that the drug is NOT highly variable, or can we apply the usual BE assessment criteria of 90% CI (80-125%) using data from periods I & II?
Assuming suitable software available (e.g., WinNonlin ≥4.x), it’s possible to obtain the standard average BE evaluation from any replicate design as well.
The point estimate is simply the same, with the additional information of the CI.
I would not 'throw away' half of the data obtained in the study.
❝ […] The intention for using replicate design and for turning back to the average BE approach (if the drug is not highly variable), would be stated in the protocol.
IMHO this should be possible.
regards,
Hermann Rotter
Complete thread:
- Replicate Designs for Average Bioequivalence joyjac 2006-01-27 00:58
![[Mix]](https://static.bebac.at/img/mix.png "open in Mix view")
- Replicate Designs for Average Bioequivalence H_Rotter 2006-01-27 12:15
- Replicate Designs joyjac 2006-01-31 06:41
- Replicate Design/Scaled ABE approach joyjac 2006-08-23 09:47
- Replicate Design/Scaled ABE approachH_Rotter 2006-08-23 12:09
- Replicate Designs for Average Bioequivalence H_Rotter 2006-01-27 12:15
Ing. Helmut Schütz