I think my memory may be going, but I don't remember writing about this important paper. It is: McAninch E, and Bianco AC in Annals of Internal Medicine, 2016: vol 164, pages50-56. doi:10.7326/M15-1799/
In this paper they show that there never have been satisfactorily controlled trials of T4 therapy, when this treatment was begun to be favoured over NDT long ago. And they also highlight the existence of the 10-15% of subjects who don't do well on T4 only. So I think it a bit rich for the thyroid establishment to carp about the lack of perceived benefit in trials on conjoined T4/T3 therapy when they haven't conducted proper trials on T4 alone.