Koo,

I think what Toinv is talking about the primary outcome meeting, or not meeting, the pre-specified threshhold for statistical significance. I don't know the BETonMACE statistical analysis plan, and I probably wouldn't understand it enough to explain it anyways. In most cases the threshhold for statistical significance is when one exceeds the 95% confidence interval, which is equivalent to a p-value of less than 0.05. 99% confidence interval is better than 95%, and p-value of 0.01 is better than 0.05. But really as long as BETonMACE meets the required threshhold, that is all that matters.

BDAZ