The article implied a causal relationship (blue partisanship causes higher life expectancy) but overlooks relevant changes in partisan alignment. For example the article shows that life expectancy in the tidewater region is higher than yankeedom. But Virginia was solidly red from 1950 until 2008. By contrast, Appalachia was mostly blue from the FDR era until 2000. Look at the counties that voted for Carter twice and then for Clinton.

Meanwhile, most of these life expectancy patterns existed before these political realignments. For example, West Virginia had low life expectancy in 1990 as well, despite voting for Dukakis by almost the same margin as his home state of Massachusetts.

Even just looking at the current statistics it doesn’t map cleanly to partisanship: https://www.cdc.gov/nchs/pressroom/sosmap/life_expectancy/li.... Idaho is about as deep red as it gets. It voted for Trump 2:1 and hasn’t voted democrat since LBJ. So is Utah. But both have high life expectancy, similar to Oregon or Vermont.

Also, the whole exercise is meaningless without adjusting for race. Race gaps are bigger than state gaps. California does artificially well because Hispanics and Asians live longer than whites. Using 2012 data, whites in California have a lower life expectancy than whites in Idaho or Iowa: https://www.livescience.com/18835-race-lifespan-states.html