Hel­mert trans­for­ma­tion? [Software]

posted by Shatha – 2022-04-05 14:00 (1171 d 16:23 ago) – Posting: # 22910
Views: 5,695

❝ You can’t verify anything in science.

❝ In statistics you can only try reject a properly stated null hypothesis – which is in your case what? Any test at level \(\small{\alpha}\) has an unavoidable false positive rate. Do you want to exclude \(\small{(100\times\alpha)\%}\) of subjects although they are in (pseu­do) steady state?


I will not exclude any subjects from analysis, I just need to check achievement of steady state in compliance with EMA guideline. The guideline didn't specify the verification method.
I have reviewed the comments raised for this guidance, the proposed changes in page 53 and page 54 include visual inspection of the last three pre-dose concentrations to verify steady state achievement and the feasibility to use the last 2 pre-dose concentrations and the Ctau instead of the last 3 pre-dose concentrations.
EMA has evaluated the comments.
They mentioned in page 4 that clarifications on steady state evaluation was included in the final guidance following stakeholder comments. I searched for this final guidance but I didn't find it, the published effective version is still the one came into effect in 2015.
Is it feasible to verify steady state visually?

❝ ❝ I need your help to perform this verification using Helmert transformation method


❝ Do you have a reference demonstrating its application in PK? Never came across this approach in 40+ years. See there why statistical tests are nonsense – essentially because we don’t have a proper null.


No, I don't have references. It was suggested as the best method for steady state verification by a reviewer during protocol preparation two years ago and accordingly documented in the protocol.

Thanks for your cooperation.

Complete thread:

UA Flag
Activity
 Admin contact
23,424 posts in 4,927 threads, 1,676 registered users;
49 visitors (0 registered, 49 guests [including 8 identified bots]).
Forum time: 06:24 CEST (Europe/Vienna)

Complex, statistically improbable things are by their nature
more difficult to explain than
simple, statistically probable things.    Richard Dawkins

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