Replies to post #223669 on Biotech Values

[biomaven0]

I suspect that they didn't properly align the HR confidence interval with the required 0.02 significance level.

That mismatch sometimes seems to be a thing:

It seems to be acceptable or has been accepted in publications to present a confidence interval that does not match the corresponding significance level or alpha level.

http://onbiostatistics.blogspot.com/2016/11/

[DewDiligence]

Elaborating on what Peter posted, if MRK has shown the 97.7% CI for the HR of final OS (consistent with its alpha of 0.023)—instead of the 95% CI—the upper bound would indeed have been >1.0.