The consistent standard error size might have more to do with...

The consistent standard error size might have more to do with uncontrolled variables within the system the data is being measured from. There are a lot of varables that affect severity and spread of acne including temperature, hormonal cycles, daily activities, skin cleanliness etc. I haven't read anything yet about the trials that state these variables were controlled somehow or that trial was split into groups with these variables common. It looks like the trials were designed to reflect real world conditions as much as possible so it could be expected you'd get standard deviations like those shown for each trial. A shift in the mean however like in the US combined and 5% BID results does suggest a systematic error due to a manufacturing flaw. As has been pointed out before there might be a correlation between these two results, not in labelling but in a manufacturing flaw that affected both. Only a guess, I'm sure once the company knows they'll tell us. Might be a while though as there'll be a lot of QA paper work to go through to figure it out. Apologies for bringing this up as i know many have moved on and want to move on there just seemed like a lot of confusion about the standard deviation.