Are you looking at the year the death happened, or the year it was registered?
We think there's a "cohort effect" with suicide deaths. So, all people born in decade X have a higher rate of death by suicide, and we see this as they get older. Here's an example from England.
> Possible contributing factors for the rise in suicide rates among middle-aged adults include the recent economic downturn (historically, suicide rates tend to correlate with business cycles, with higher rates observed during times of economic hardship) (6,7); a cohort effect, based on evidence that the "baby boomer" generation had unusually high suicide rates during their adolescent years (8); and a rise in intentional overdoses associated with the increase in availability of prescription opioids (1,2). Additional research is needed to understand the cause of the increase in age-adjusted suicide rates and why the extent of the increase varies across racial/ethnic populations.
I don’t think that really addresses my point though. My point is that comparing to 1999 might create a skewed statistical perspective because that seems to be around the time when Americans were engaging in the least self harm.
To clarify: I think alcohol can be a pernicious social and mental health problem.
But there’s also something weird with 1999. Suicides seemed lowest that year and I think that might relate to the drinking deaths as well.
Which makes comparing to that year a little dicey, because I think it creates a false sense of a trend that might not exist.