>It's only over-prescription if the prevalence hasn't changed.

Actually you've missed one other classification.

Was it previously under diagnosed?

So we have underdiagnosis, increase in base rate, increase in fraudulent diagnosis. At least anecdotally based on the older people in my family that seem to snore hard and have poor sleep habits, underdiagnosis seems pretty likely to me, but I guess we'd need to see studies of randomly selected patients to know for sure.