In addition to trukterious' point, which I agree with, I would echo that the idea that the RDA is just wrong is also a distinct possibility. Take a look at this: http://blogs.creighton.edu/heaney/2015/02/13/the-iom-miscalc... Ponder what it implies about the other RDA values. And look at the date. It wasn't that long ago.

I just pulled a fairly recent bottle of CostCo Vitamin D out here, and it's still using a value of 800IU as the RDA.