RedHill Biopharma Ltd (RDHL)

[noretreat]

That fact is indeed good news...the great news is that they can read high efficacy among the test group even if placebo shows 20% positive. That has to mean they are seeing AT LEAST 35% efficacy among all participants and perhaps even 50%.

With 20% response among placebo group and 35% response overall, Efficacy among the test group would be 50%. With 160 test participants the confidence interval around the 50% number would be 12 points (3 standard errors). A slightly tighter interval would exist underlying the placebo group.

They almost have to be seeing numbers this good to cut the sample by that much. Otherwise to do so would be near insanity.