As Bart said, we should get some data to see if mobile use is actually predictive of question quality.
Assuming the data says it is, I think the easiest solution is to add mobile use to the question quality heuristic used to feed mobile posts to the Triage and Low Quality queues (for questions and answers respectively) and let our current bad-post-handling systems handle them. (First posts already go into the eponymous queue, though arguably it should be a "new users" queue that checks the first N posts for N > 1.)