Biotech Values

[iwfal]

ONTY

- First, there's an inadequate understanding on this board of what "overpowering" means.... and in the context of HRs, it means that low p-values don't necessarily translate to particularly low HRs.

I think "overwhelming efficacy" is the term that y'all were claiming as proof that the p value had to be lower than OBF. But you are missing the fact that "overwhelming efficacy" are just the buzzwords used to describe a stop at any interim. You can actually see that terminology used in official guidelines and textbooks as the description for any stop at an interim. And the reasoning is obvious - that with fewer events and a smaller alpha a stop at an interim is, by definition, a lot better than the HR the trial could have actually seen and still been stat sig. E.g.:

http://www.drugabuse.gov/funding/clinical-research/draft-guidelines-establishing-operating-data-safety-monitoring-board

PS As I noted on the other board - of 11 Big Pharma onc trials I hunted down, with absolutely no ambiguity in the protocol, 5 had interims, and 2 of those 5 used an interim that, had they stopped at the interim, would have been far from spectacular.