Bioequivalence and Bioavailability Forum

Useless power: over & over again [Regulatives / Guidelines]

posted by Helmut  – Vienna, Austria, 2012-04-25 18:32 (5175 d 06:55 ago) – Posting: # 8462
Views: 5,009

Dear Mumbai!

❝ Dear Vienna,

    ↑↑↑↑↑↑↑↑↑

❝ some time based on less power we are conduction two stage appraoch. but you said power calculations are futile why?

Two Stage studies can help in dealing with uncertain CVs – e.g., if you have only data from literature, old studies, or small pilot studies. The power calculation in the interim analysis is part of the decision tree; no post-hoc power calculation should be performed. What I said about conventional BE studies is applicable here as well: Either the study passed or not.
See also Potvin et al. (2008), Example 2, Method B:

[…] we obtain a two-sided CI for the ratio (T/R) of geometric means of 88.45—116.38%, which meets the 80—125% acceptance criterion. We stop here and conclude BE, irrespective of the fact that we have not yet achieved the desired power of 80% (power = 66.3%).

(my emphasis)

—
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:

• power calculation for partial /full replicate study designs balakotu 2012-04-02 15:06 
  • Useless power Helmut 2012-04-02 16:19
    • Useless power balakotu 2012-04-17 06:11
      • Useless power: over & over againHelmut 2012-04-25 16:32
        • Useless power: over & over again lindseyk 2012-04-27 16:25
          • What for? Helmut 2012-04-28 03:12