Silva ☆ Portugal, 2014-04-17 10:57 (3634 d 04:30 ago) Posting: # 12846 Views: 2,541 |
|
Dear all Is it advisable to use two stage design approach in partial replicate design? If yes, which alpha level should be used for the second stage (combined stage I + stage II data)? Best regards Silva |
ElMaestro ★★★ Denmark, 2014-04-17 11:17 (3634 d 04:10 ago) @ Silva Posting: # 12847 Views: 1,636 |
|
Hi Silva, ❝ Is it advisable to use two stage design approach in partial replicate design? ❝ If yes, which alpha level should be used for the second stage (combined stage I + stage II data)? There are no publications about two-stage studies and replicated designs of any type. Therefore I would only attempt to do it after a scientific advice. They may ask you to prove by your own simulations if/how the method performs in various scenarios. — Pass or fail! ElMaestro |
Silva ☆ Portugal, 2014-04-17 11:36 (3634 d 03:51 ago) @ ElMaestro Posting: # 12848 Views: 1,690 |
|
Hi Helmut Once I met someone with the same doubts and he told me that he have asked for a CHMP SA on the subject. As I remember from the talk, his opinion was that the agency was also not able to deal with this complex statistical analysis and therefore the advice was to consider the second stage group as a group effect and therefore only 90% CI's were calculated and not the 94.12%! But this is not mandatory, as no public information is available! Rgs Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Helmut] |
Helmut ★★★ Vienna, Austria, 2014-04-17 17:33 (3633 d 21:54 ago) @ Silva Posting: # 12850 Views: 3,237 |
|
Hi Silva, ❝ […] a CHMP SA on the subject. As I remember from the talk, his opinion was that the agency was also not able to deal with this complex statistical analysis… EMEA. The European Medicines Evaluation Agency. ❝ … and therefore the advice was… So they felt not to be “able to deal with this complex statistical analysis” and nevertheless gave an advice? Weird. When you know something, say what you know. ❝ … to consider the second stage group as a group effect and therefore only 90% CI's were calculated and not the 94.12%! That’s bizarre and a potential serious risk to public health, Imho. Btw, a 94.12% CI (αadj 0.0294) is not a solution for everything. Some people (including myself) think that already EMA’s standard ABEL inflates the patient’s risk (see this thread). Two stages‽ If somebody does know a procedure, please enlighten me. — 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 |