He specifically referred to the 3 day NEWS2 parameter. If we only met stat significance in day 3, I would assume it's because many patients showed very quick recovery.
Is it possible that success at 3 days makes the 7 and 14 timeframes hard to meet p value on?