I think this is implied in this paper by Aaronson (http://www.scottaaronson.com/papers/bqpph.pdf) but I am not sure.
Begin with $NP \subseteq BQP$ (*)
$\Sigma_{2}^{P} = NP^{NP} \subseteq BQP^{BQP} = BQP$
$\Sigma_{i+1}^{P} = NP^{\Sigma_{i}^{P}} \subseteq BQP^{BQP} = BQP$
With this inductive step we have $PH \subseteq BQP$
The only part of this proof I doubt is the step $NP^{NP} \subseteq BQP^{BQP}$ because I am unsure of some oracle rules.
My assumption is that if $A \subseteq B$ then:
$A^{C}\subseteq B^{C}$
and
$C^{A}\subseteq C^{B}$
with no assumptions about $C$, are these correct?
I feel like this would be a well-known result if it were true without any additional assumptions, so my belief is that I must need to make an additional assumption at some point that I have missed.