the life expectancy of male workers retiring at 65 had risen six years in the top half of the income distribution but only 1.3 years in the bottom half over the previous three decades.
However, deep in the article they also note that the behavior of the poor and the rich among the old are quite different. Comparing a poor Florida county with an adjacent rich one (Putnam vs. St. Johns), Fletcher notes:
Adults also smoke at nearly double the rate they do in St. Johns, and they are far more likely to be obese and far less likely to be physically active.
He quotes doctors who say that the difference in life expectancy shows differences in health insurance and health care availability. Yet it is clear that a big part of the difference is due to self-chosen behavior, by both the rich and poor populations.
The centrist position here is that both factors are relevant.
But I think at this point we can say that the many bad effects of smoking, at least, are the smokers' own dumb fault, and not primarily to be addressed by health insurance and having more doctors. And in coming years I think we will reach the same conclusion about most causes of obesity and lack of exercise.