But if I understand correctly, the point of bringing up the Oregon study was that access to medical care doesn't seem to have a lot to do with it. At least not the incrementally better access provided by Medicaid programs.

I would certainly speculate that there is a bundle of ideas that contribute to it. Ideas about personal decisions being impactful, leading to things like less smoking and better diet, and even better life situations (which can presumably lead to a different relationship with stress).

Also any disabled person (who probably has other health problems that reduce life expectancy) is by definition poor (poverty is defined as low market income).