The problem I have with this reasoning is that just about everybody has hearing loss in those upper frequencies, so testing and finding hearing loss there doesn't really tell us all that much (although, I'd still prefer it if they did). You will find a minefield of people with varying degrees of hearing loss (both in a traditional audiogram up to 8 kHz and those who only show upper losses), and those who have tinnitus as a symptom.One problem is the rudimentary audiogram which often only goes up to 8 kHz. Extended high-frequency audiogram can show hearing loss in those upper frequencies but this test is rarely carried out. If you get perfect results from that too, you can still have hidden hearing loss.
You are only getting an extra octave's worth of information for the average person by including the higher frequencies (depending on age, of course) because most 30 odd year-olds can only hear up to around 16 kHz. It should be expected that everyone will have damage of some kind, though, unless they are small children or something.
The crux of this argument cannot be settled until the hypothesis is put to the test, and that can't happen until someone manages to restore hearing loss so we can see what happens. Both sides have well thought out ideas of what could be happening to cause it.