You're looking at covid. Look at the 25 year period from .90 in 2000 up to 9.3 by 2008, then flat until covid, then a bump during covid times before lowering at 9.5

Sure you can say it's reduced from a peak which happened to coincide with an unprecedented pandemic. Wake me when it's back to the 9.3 norm of the 2010s let alone the 9 it was around 2000.

Now it's quite possible that fewer people are drinking at all, but of those who are drinking, they are drinking more.

Or it's quite possible that annecdotes about who you see in the gym and on instagram are just annecdotes

(for what it's worth I have seen data showing under 30s are drinking less in 2025 than under 30s drank in 2005, or 1985, etc, which would imply over 30s are drinking more to make up for it, but I don't have it to hand and it may not have been US specific)