Bioequivalence and Bioavailability Forum

❝ Does anybody understand / can explain to me what is meant by the last bullet point?

• 90% confidence interval:
  

  ❝ as one element of such justification, if applicable, the applicant could also take into consideration the width of the 90% confidence interval and the ratio to possibly justify that a false positive outcome due to ISR problems has a low probability.

Dear Helmut et al,

Maybe it's one of those riddles, wrapped in a mystery, inside an enigma.

My take may come across as a platitude, but it's not meant to so I'll go ahead with it anyway. The way I see it, and I'm not sure this is correct, is that if you happen to have a product with a narrow 90% CI (say 95 - 105), then you could opine that even if you have not performed the required ISR, it is unlikely that if you did and you found differences in your measurements that these differences would have pushed your study results outside the 80 - 125% CI. You would then conclude that is is unlikely the results of your study were the result of a false positive (i.e. it's a true positive). You could make a similar argumentation if your ratio was within 95 - 105%, I'd say. So, you are probable going to have better luck with studies with low intrasubject variability...unless the product has been shown to exhibit back conversion, which appears, in that case, everyone is screwed for those products.

Outlaw