Bioequivalence and Bioavailability Forum

statistical significance ≠ clinical relevance [Study Assessment]

posted by Helmut – Vienna, Austria, 2025-12-23 14:52 (183 d 09:27 ago) – Posting: # 24538
Views: 1,699

Hi Darborn,

❝ Has anyone ever encountered a BE study result where the CI did not include "1"?

Very rarely and only for drugs with low variability combined with unreasonably large sample sizes. For details see this article.

❝ If the regulatory agency asks for an assessment of whether bioequivalence can still be demonstrated, …

It demonstrates their lack of understanding the underlying concept of BE (statistical significance ≠ clinical relevance).

❝ … how should I reply?

See the last two sentences of the Summary of the linked article.

—
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

Robustness of bioequivalence when 90%CI doesn’t include 1 Darborn 2025-12-23 08:55
- statistical significance ≠ clinical relevanceHelmut 2025-12-23 13:52
  - statistical significance ≠ clinical relevance Darborn 2025-12-24 01:07
    - Δ (20% in general) Helmut 2025-12-24 09:59