Here is a link to the actual paper: http://journals.plos.org/plosone/article?id=10.1371/journal....

They recommend 4,000 IUs of vitamin D as the maximum safe dose, not because higher doses were found to be less helpful but because 4,000 was the most helpful and also the highest amount they actually tested. It is possible (but untested) that 6,000 and 8,000 IUs could have been even better.

However, it does appear that at 16 weeks both the groups getting 2,000 and 4,000 IUs of vitamin D leveled out at the same blood serum level of vitamin D.