Futility analysis sample size


New Member
I have been asked to determine the sample size needed to demonstrate futility in a small study. I suggested going the "traditional route" and conducting an interim analysis at a pre-specified timepoint or sample size to determine whether or not to continue (likely via conditional power), but I'm wondering if there is a reasonable method for just providing the sample size a priori. The only thing I can think of is to determine sample size based on beta (Type II error), but I don't see how that would be different from determining sample size based on power.

Any suggestions welcome! Thank you in advance!