I think the concern is that you could still be getting pharmacodynamic effects at 1mg, so the control wouldn’t really be a control. It’s entirely feasible that

I think the concern is that you could still be getting pharmacodynamic effects at 1mg, so the control wouldn’t really be a control. It’s entirely feasible that you could get good results and limited side effects at dose x, and poor results overwhelmed by side effects at dose 25x.