I still think that T/R-ratios are useful [GxP / QC / QA]

posted by Helmut Homepage – Vienna, Austria, 2022-05-25 14:04 (932 d 02:21 ago) – Posting: # 23019
Views: 7,366

Dear Ohlbe,

❝ Well, remember GVK ? According to the ANSM inspection report (document A3, page 3), …


Didn’t know that the documents are in the public domain in the meantime. ;-)

❝ … the whistleblower said that "the switch", as ElMaestro calls it, was done only for some sponsors, who paid the head of the bioanalytical lab to do it. The report does not mention any evidence being found to support this allegation, but that's indeed a possibility.


However, as I wrote above it’s a pretty trivial exercise for anybody to critically inspect the data. No need to buy a pig in a poke.

An example from my files: 2×2×2 crossover of a HVD back in the day when reference-scaling was not acceptable, assumed T/R-ratio 95%, CV 40%, targeted at ≥80% power, anticipated dropout-rate 10%. Hence, the large sample size (74 dosed). In the study (71 eligible subjects) the PE was with 96.77% slightly better than expected (90% CI: 89.82–104.26%) and the CV with 27.1% as well.

[image]
Possibly the statistician was fired by the man in the Armani suit shouting
»You’re a bloody looser‼ Why didn’t we perform the study in 16 subjects‽
I calculated post hoc power in FARTSSIE and it’s already 81.6%! What’s the point of having 99.4% at the end?«

(With a sigh, the statistician types cmd rmdir "C:\Users\Genius" /s /q
and tries to become a regulator or an academic…)


❝ – if the sponsor selected the CRO precisely because it is doing "the switch" (be the decision to select that CRO or to make "the switch" made by the Man in the Penguin Armani Suit or by somebody not willing to bring bad news to this man), well...


Likely he receives annual gratifications cause all studies pass. If sumfink is uncovered, he moves to the next company.
Nowadays being employed by numerous companies is understood as sign of competence and ‘solution based flexibility’.  Boomers  Dinosaurs who stayed for decades in the same company are on their way to retire.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

UA Flag
Activity
 Admin contact
23,336 posts in 4,902 threads, 1,669 registered users;
20 visitors (0 registered, 20 guests [including 5 identified bots]).
Forum time: 15:26 CET (Europe/Vienna)

Biostatistician. One who has neither the intellect for mathematics
nor the commitment for medicine but likes to dabble in both.    Stephen Senn

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