>>I do not know whether the subset was biased toward "surviors". My impression is it was not. Dew may know more about this. I think its a good question. <<
I'm not a statistician, but as I see it, for the subgroups of heparin vs non heparin to be valid there would have to be randomization pre treatment. If heparin vs not was up to the whim of the treating docs then you can't draw a valid conclusion, rather just speculation and a nudge to do futher studies. I think that was the case, correct me if I'm wrong.
Market Data 
Markets 
AI
