Log in
Register|
BE-proff ● 2015-12-26 17:44 (3826 d 15:32 ago) Posting: # 15767 Views: 6,544 |
|
|
Hi All, If to conduct a BE-study with alfa error 5% and power 80% it means that a customer has 5% risk to buy non-BE medication and a developer has 20% risk to reject bioequivalent drug. So, if we failed BE-study because PK-parameters were out of acceptance range may be we were "unlucky" and fall into 20% risk?  I wonder if there are any ways to decrease risk except sample sizing? |
|
Helmut ★★★   Vienna, Austria, 2015-12-26 18:44 (3826 d 14:32 ago) @ BE-proff Posting: # 15768 Views: 5,119 |
|
|
Hi BE-proff, you got it! The answers to your questions are: 2 × yes – yes – no. Of course I assume that: You cannot manufacture a formulation with less deviation from the reference (or find a batch of the reference which is closer to yours). You have done the best to reduce within-subjects variability (in this order: conduct of the clinical part in general > sample handling/transport/storage > bioanalytics). β = type II error, power = 1 – β Post #1,000 in the category Power / Sample Size ★ — Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
Ing. Helmut Schütz