Thanks tada.

My goal was to get some sort of reference statistically on the required test vs control difference and then importantly to so see where that sits in a statistical framework. However, this is a surrogate model at best. If it is completely off base I hope some readers/posters will set me straight.

I am assuming that at the SSA the committee will have biostaticians use the actual "time to first event" raw data to determine sample size requirements. I would assume that the panel would also examine the statistical requirements at the sub-group level e.g. Crestor vs Lipitor, etc.

So if my analysis is a reasonable surrogate model based on 8 MACE events/100 patient years then we should be hoping that the rate in the test group around 6.5% and the control group around 9.5% to hit the minimum RRR of about 30% and 99% confidence level.

If the results are in the test = 7% and control = 9% then we are in an "iffy" situation  with an RRR of 22.2 and I would hope that the committee would examine the Lipitor and Crestor sub groups.

If the test results are test =6% MACE and control =10% then the RRR = 40% and the results are significant at the 99.9% confidence level. This would be wonderful.

Anyway, this is all hypothetical because we don't know and won't know anything about test vs control until top lines are released. I hope the top lines are handled with great care and are not just superficial. For example if the results are "iffy" with Lipitor but significantly positive with Crestor I hope this is presented.

GLTA

Toinv