Bioequivalence and Bioavailability Forum

Dear HS,

Thank you for your reply!

❝ That’s the type of argument I definitely like the most. In line with “We have used bloodletting for 2000 years – can’t be wrong.” Quackery!

Indeed, such an argument gives me the shivers...

❝ Shouldn’t be such a big difference? See here and there.

As mentioned in the posts you refer to, differences should be marginal. In this case, however, differences are actually quite large in my opinion. Let me give you an example. For a two-stage crossover design both the CRO and I have calculated sample sizes with the following parameters:
alpha = 0.049
PE = 0.8969
CV = 0.1149

With both PowerTOST and FARTSSIE (and even StudySize) I come to a sample size of 16 (power=0.8474), while the CRO gives a sample size of 14 (power=0.820).

According to the file I received from the CRO, the Power procedure (Equivalence Test for Mean Ratio) was used with distribution set as lognormal and method as exact.

Can you explain these differences? Did the CRO use an incorrect procedure/syntax/option or anything? Or do you think that these differences are actually small?

❝ Well, you are the boss.

Thanks in advance!

Best regards,

Oiinkie