Bioequivalence and Bioavailability Forum • S × F interaction

S × F interaction [Design Issues]

posted by Helmut – Vienna, Austria, 2023-06-07 12:34 (738 d 17:03 ago) – Posting: # 23580
Views: 2,986

Hi Brus,

❝ I don't understand the subject-by-formulation effect, …

You are not alone.

❝ … can it only be studied in full replicated designs?

Yes.

❝ If the subject-by-formulation effect is the change between test and reference difference means from one subject to another, in a 2x2 design it should be possible to study it. Right?

No. In a 2×2×2 crossover we have already aliased effects (are victims of confounding). See this article.
Adding S × F will not work. Didn’t try it yet. Either you get zero degrees of freedom in a fixed effects model or a mixed-effects model will collapse.

❝ And in that case it could also be measured in partial replicate. Right?

Nope.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

Complete thread:

Switchability with replicate design studies Achievwin 2022-05-20 04:44 [Design Issues]
- Switchability = IBE Helmut 2022-05-20 09:17
  - Switchability = IBE Achievwin 2022-05-20 14:19
    - S × F interaction Helmut 2022-05-20 15:33
      - S × F interaction Achievwin 2022-05-20 23:32
      - S × F interaction Brus 2023-06-06 11:02
        
        S × F interactionHelmut 2023-06-07 10:34
        
        S × F interaction Achievwin 2023-06-08 10:21