Look at the standard error, page-4

It doesn't matter if the US vehicle was contaminated with CBD as the US vehicle values lie within the Australian vehicle SE range (i.e. the differences between the USA and Australia cohorts can be just due to measurement error and noise). There is no need for any other explanation. You don't need a fancy explanation for tossing 3 heads in a row with a coin toss - after all chance is much more likely than the coin has heads on both sides.

Permetrex may well be an active treatment, but I suspect that most of the change is just due to regression to the mean and better skin washing. The more important thing for a shareholder of BOT is there is no evidence that BTX 1503 is doing anything at all from the data presented to date by the company. If it is doing nothing it is not surprising the safety profile is so good.

Once again I do want to point out that I am reaching this conclusion based on the data released by the company and so it may well be incorrect if the data is incorrect. I really need to see the raw data to be more confident, but in its absence we all have to work with what we have.