it was interesting to me to see that if they started out with a 50mg treatment and it was lowered to 30mg....
That is the main type of bias the ITT analysis is intended to address, and I think they did that. ITT is not intended to try to see if the drug worked in people in whom you have no follow-up data to actually determine this as some have suggested. AS Doc noted, there are some statistical techniques that can essentially calculate what the results would have been if there had been no missing data.
News 
Market Data 
Discover 
