Log in
Register|
Mohamed Yehia ★ Egypt, 2018-01-11 17:19 (2286 d 12:57 ago) Posting: # 18168 Views: 2,649 |
|
|
Hi All, Is it possible to make imbalanced Randomization sequence before entering the study? For example, using 33 subjects for 2 treatment, 2 sequence, 4 period cross over study. In my opinion, I don't accept this as despite the fact of the possibility of getting imbalanced sequence as a result of subject withdrawal, but I think that this does not mean that it is acceptable to perform imbalanced randomization from the beginning. I want your opinion asap. Thanks Edit: Relax; see also this post #9. [Helmut] |
|
ElMaestro ★★★ Denmark, 2018-01-11 18:10 (2286 d 12:07 ago) @ Mohamed Yehia Posting: # 18169 Views: 2,161 |
|
|
Hello Mohammed Yehia, ❝ Is it possible to make imbalanced Randomization sequence before entering the study? ❝ ❝ For example, using 33 subjects for 2 treatment, 2 sequence, 4 period cross over study. While it is a tradition to aim for balance I don't think there is a strong scientific argument against imbalance at the design stage. Many, many E+S trials are actually designed this way, with for example relatively few subjects allocated to placebo arms. If you have heard a regulator saying this isn't possible then it isn't possible but that would be dictated by taste, not science. You can achieve full type I error control and justify a power calculation. I am leaning towards thinking it can be done, but of course you may ask why anyone would aim for an imbalance of one. If it is simply a matter of "we can afford maximum 33 subjects" then you can surely say that scientifically you are squeezing the max out of the data by including as many subjects as possible, but power should still be decent whatever that means. If on the other hand you have failed to source enough Test or Ref units so that IMP availability creates a bottleneck then certainly the whole ordeal could be kind of dodgy from the outset perhaps and power could suffer a lot? So let us hear the actual background, please. — Pass or fail! ElMaestro |
|
Mohamed Yehia ★ Egypt, 2018-01-11 21:36 (2286 d 08:40 ago) @ ElMaestro Posting: # 18170 Views: 2,111 |
|
|
❝ While it is a tradition to aim for balance I don't think there is a strong scientific argument against imbalance at the design stage. ❝ I am leaning towards thinking it can be done, but of course you may ask why anyone would aim for an imbalance of one. […] ❝ If on the other hand you have failed to source enough Test or Ref units so that IMP availability creates a bottleneck then certainly the whole ordeal could be kind of dodgy from the outset perhaps and power could suffer a lot? ❝ So let us hear the actual background, please. Thanks a lot for your detailed reply. The actual background is that I only have 66 tablets of the reference product of a highly variable compound which means that i can only enter the study with maximum number of 33 volunteers. So, after making the sample size calculations I have found that it's better to conduct the study using the 2x2x4 approach over the 2x3x3 and despite the fact that it will not matter that much if we enter the study with 32 volunteers instead of 33 applying the 2x2x4 approach, but I am thinking about that there is a huge possibility that some.of the volunteers may be withdrawn due to any reasons. That's why I am trying to get benefit of each tablet of the reference. However, as I said before that i m not leaning towards doing what I stated above as in my opinion if I decided to use the 2x2x4 approach then I have to enter the study with even number in order to be balanced although I know that the study may end with odd number. But I believe that the design from the beginning should be applied to minimize the variations in the study as possible as can. So entering the study from the beginning with 16 volunteers with sequence 1 and 17 volunteers with sequence 2 would be a negative point in the design of the study. 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 #5! [Helmut] |
Ing. Helmut Schütz