SaraCHenriques
☆    

Portugal,
2020-06-01 20:24
(233 d 16:32 ago)

Posting: # 21496
Views: 2,635
 

 TDDS non-inferiority analysis [Software]

Hi everyone! :-)

During a Phase I bioequivalence trial of a TDDS, we recorded the adhesion of the patches (T and R), at the end of dosing interval, as percentage of area that remains adhered, according to EMA's Guideline.

I ask if anyone could advise me on how to perform a non-inferiority analysis for this data, on Phoenix WinNonlin.

Many thanks in advance!

Sara


Edit: Category changed; see also this post #1[Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-06-14 12:05
(221 d 00:51 ago)

@ SaraCHenriques
Posting: # 21538
Views: 1,665
 

 Phoenix: One-sided tests

Hi Sara,

sorry for not replying earlier.

» […] how to perform a non-inferiority analysis for this data, on Phoenix WinNonlin.

Not directly supported in Phoenix. However, there is a simple workaround. Non-inferiority tests (and non-superiority as well) are one-sided. Perform the BE-analysis as usual in Phoenix (i.e., two-sided for equivalence at α 0.05) and assess whether the lower confidence limit is ≥80.00% of the reference. Ignore the upper limit.

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

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

Portugal,
2020-06-26 23:10
(208 d 13:46 ago)

@ Helmut
Posting: # 21581
Views: 1,553
 

 Phoenix: One-sided tests

Hi Helmut,

Thank you so much for your answer!

Regarding the fixed and random effects, do you have any recommendation? Should we use all effects fixed as for PK BE analysis; or only Subject and Treatment fixed as when comparing irritation scores?

Thank you once again!

Sara

Sara
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-06-29 14:54
(205 d 22:01 ago)

@ SaraCHenriques
Posting: # 21601
Views: 1,544
 

 Phoenix: One-sided tests

Hi Sara,

» Regarding the fixed and random effects, do you have any recommendation? Should we use all effects fixed as for PK BE analysis;

Since the PKWP seemingly is obsessed with “all effects fixed”, likely yes. ;-)

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

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

Portugal,
2020-06-29 20:12
(205 d 16:44 ago)

@ Helmut
Posting: # 21606
Views: 1,523
 

 Phoenix: One-sided tests

Hi Helmut

Thank you very much! :-)

Sara
Activity
 Admin contact
21,308 posts in 4,444 threads, 1,489 registered users;
online 9 (1 registered, 8 guests [including 2 identified bots]).
Forum time: Thursday 11:56 CET (Europe/Vienna)

Politicians use statistics like drunkards use lampposts:
not for illumination, but for support.    attributed to Hans Kuhn

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