Bioequivalence and Bioavailability Forum

Dear Helmut!

❝ below the code for all designs covered in PowerTOST(), conventional (unscaled) ABE. You have to paste the functions only once to the R-console and call the function Sens() with values defined in the end or call it directly, e.g.,

❝ Sens(CV=0.24, GMR=1.07, pwr.target=0.8, pwr.min=0.7, des="3x6x3")

THX for that code. Maybe there is a better solution than pasting the function into the R-console: Incorporating it into PowerTOST . May need some additional effort for checking the input and some effort for documenting.
May need also some thinking about the return value(s) of the function.

BTW: What did you mean by your sentence in your post above "This little exercise is not a substitute for the sensitivity analysis which should be performed in study planning ..."? Sens() not useful in real life?

❝ Note that the code tries to keep the degree of imbalance as close as possible between sequences. In real life other combinations may occur. As already noted by Detlew in this post the more im­bal­anced a study is, the lower the power.*

This is surely a minor shortcoming of your code. But if we assume drop-outs by random its a reasonable attempt I think in the planning phase where we don't have 'real life' data. And at least I have no "better" idea.

❝ ...

❝ @R-Freaks: The construction of imbalanced sequences is both lengthy and clumsy. Any suggestions are welcome.

Will try it if time allows .
Idea: use number of sequences = steps from known.designs() and make a general solution of the n's in sequence groups - looping over sequences.

❝

• If a sequence is completely lost, power2.TOST() – correctly? – gives power=0. See these two examples:
  

  ❝ power2.TOST(CV=0.2, n=c(1, 2, 3, 3, 3, 3), design="3x6x3")


  ❝ power2.TOST(CV=0.2, n=c(0, 3, 3, 3, 3, 3), design="3x6x3")

If this is truly correct? Don't think so.
Simply the design changes to another one. Maybe also to one not implemented.
In your example it changes to "3x5x3". Not in known.designs()!
This can't handled properly in PowerTOST at moment. No df's, no design constant ...
Eventually power2.TOST() should throw an error/warning if it encounters some n's = zero.