I have been an optimist and as Dew said my view could be biased because I am involved in the love affair with DNDN (moneywise and scientifically) - my numbers are as follows:
1. If IMPACT is more like 9901 than 9902a, then interim power is at least 90%
2. IF it is more like 9902a than 9901, then the interim power is no more than 20%
3. If IMPACT tracks the integrated 9901/9902a, then interim power is at least 50%.
I am assuming interim event goal has been substantially increased from 164 to 230, and that DNDN advanced final timeline with detailed in-house simulation, an action that any statisticians that are worth their salts should have been able to do, and have assured themselves the reduced event goal reduction would not harm final analysis.
