This really isn't an issue for IMPACT as modeled by D9901+02a. With OBF, the nominal alpha for the final is always above .04 regardless of where the interim is. As such, the power for the final look at 360 would always be over 90%.
Apparently I was not clear in my point (which was, admitedly, a subtle one). To restate it differently - my point was that as the interim moves out towards the final, the total alpha may go up. But more and more of the power of the trial is in the interim, and less and less is in the final. In the extreme, imagine that the interim was moved out to final event count minus 10 events, and the alpha assigned to the interim was 0.045. The alpha assignable to the final is going to be about another 0.045 - but the chance that that alpha would ever be tripped is very small because almost all the successful trials would get stat sig on the interim. Very few would fail the interim and get the final - even though the final has an alpha that is quite large.
News 
Market Data 
Discover 
AI
