But the problem is you're basing that on a stat in a report that even the authors of the report state is not reliable and have explicitly stated should not be used to draw the conclusion you have.
As I said in an earlier post that study had unknown, ie: unrecorded, at 38% and investigation by...