>DNDN – I would expect the HR to go up now with the tail events coming in.<
The non-constant HR is only part of the story. The other part of the story is Program-Survival Bias.
It’s now obvious that the HR observed in 9901 and 9902a was biased high relative to the true underlying HR in this patient pool, just as PSB said it should be.
“The efficient-market hypothesis may be the foremost piece of B.S. ever promulgated in any area of human knowledge!”