[timr]

Before looking for cause, you always look for plausible co-variates. Here, there are a ton: red counties tend to be rural. Rural places have huge problems with health care access, obesity, addiction, poverty, and many other issues that predict poor health outcome.

[kibwen]

The study takes ruralness into account, shown in this graph: https://www.bmj.com/content/bmj/377/bmj-2021-069308/F4.mediu...

[timr]

That doesn't "take it into account"...it just plots them, based on which presidential candidate the county voted for. And by looking at the plot, it's clear that age-adjusted mortality rates are almost perfectly aligned for rural democratic and republican areas until 2009, when they drop suddenly for the democratic areas. Then they both continue tracking in parallel. Why is that? Can't tell. Sure is suspicious. Could it be that a bunch of counties flipped in 2008?

I skimmed this paper to see if the authors did any of the facepalm-obvious things you'd do to control for covariates if you were serious about doing an observational analysis (e.g. multi-factor regression), but saw nothing. They're literally just looking at differences in slopes in the lines:

> Trends in mortality were examined to identify changes in slope using Joinpoint Regression Program version 4.8.0.1, which models consecutive linear segments on a log scale, connected by joinpoints, and can measure when slopes of annual percentage change (APC) undergo a statistically significant change

This is a...well, let's just call it non-conventional...analysis method.

[orwin]

> Rural places have huge problems with health care access, obesity, addiction, poverty, and many other issues that predict poor health outcome.

So, yes, probably, but. I went in a rural area in West Virginia for some personal reasons. And i mean rural. One hours and a half from Charleston, and half an hour of the nearest town (population: 728). I think i met all the democrat voters there(four dozen or so :P). Mountain guides, kayak/raft guides, musicians(a lot), old hippies living off the land, a lily farmer, a lot of organic farmers too (And the first goat farmer who convinced me than US cheese is at least as good as french/italian goat cheese). Some of them without running water, almost half of them off grid. I think the "fatest" one was barely over my weight, so between 27 and 29 BMI. I also went to the organic farmer market in Hungtington (Clearly not republican leaning, as the sticker against mountain top mining, pro-unions and "Freedom Industries = Lexycon, protect our water" could show) and to a city music festival (were i met all those communist hippies again). Frankly, the overall health and shape of those people, i thought i was back in europe.

Poverty and health care access, yes. Addiction: not that much. Some are alcoholics, i think i met one meth user, but no one tried to sell to us, and their social life is way too full to have time for this shit. Obesity: clearly not. Less than in Paris.

Also, these people were clearly aware of the Elk river incident (it was four years after, still had banners and stickers warning about water quality) and did flush/replace their pipes, and could very well be the only ones who did.

[Spivak]

That would only explain a point in time state of affairs but not why they observed the effect widen along party lines. Their theory was that blue areas expanded medicare and red areas didn't but there could be others.

[heavyset_go]

Medicaid was expanded to cover those with low incomes in some states, not Medicare.