Confidence interval between two study for same test and reference lot [Study As­sess­ment]

posted by ElMaestro  – Denmark, 2021-08-04 23:41 (485 d 02:10 ago) – Posting: # 22508
Views: 1,663

Hi RG628,

❝ We conducted two BE study for one of the ANDA, in one study we got 90% CI (77-108) and another study we got (106-125.8) for same test and RLD lot. Will it possible to combine both study two justify product is bio-equivalent based on observe variability for regulatory submissions.


Very unfortunate.
I would check first if T and R were by coincidence switched somehow in the evaluation. I.e. if you had a 90% CI for T/R of 77-108 in one study, and then got a CI of 106-126 for R/T (rather than T/R) in the next. This would be consistent.
Such an issue could for example be related to treatment coding and alphabetical order etc. If T and R were called T and R in one study, but A and B in the other, and if both datasets were evaluated with the same script then strange things could happen depending on how the scripts were written.

I know this is a long shot from my side and I hope it does not offend you that I am mentioning it as a potential cause for a surprising observation. I can only confess that this is the type of mistake I could so easily make myself.

Pass or fail!
ElMaestro

Complete thread:

UA Flag
Activity
 Admin contact
22,426 posts in 4,694 threads, 1,600 registered users;
14 visitors (0 registered, 14 guests [including 9 identified bots]).
Forum time: 00:52 CET (Europe/Vienna)

The rise of biometry in this 20th century,
like that of geometry in the 3rd century before Christ,
seems to mark out one of the great ages or critical periods
in the advance of the human understanding.    R.A. Fisher

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5