Why are poor, white women dying younger than they used to?

White women who don’t graduate from high school have seen their life expectancy decline by five years over the past 18 years. As Monica Potts explores in a fascinating long-read at The American Prospect, that’s a big effing deal.

There are lots of racial, educational, and economic disparities when it comes life expectancy in this country (this biggest gap is between the most educated white and least educated black folks), but one thing that’s held true almost across the board is that it has been on the rise–drops in life expectancy are super rare. But women like Crystal Wilson of rural Arkansas–whose life and death Potts explores in the piece–are now unlikely to live as long as the generation that came before them.

Everything about Crystal’s life was ordinary, except for her death. She is one of a demographic—white women who don’t graduate from high school—whose life expectancy has declined dramatically over the past 18 years. These women can now expect to die five years earlier than the generation before them. It is an unheard-of drop for a wealthy country in the age of modern medicine. Throughout history, technological and scientific innovation have put death off longer and longer, but the benefits of those advances have not been shared equally, especially across the race and class divides that characterize 21st–century America. Lack of access to education, medical care, good wages, and healthy food isn’t just leaving the worst-off Americans behind. It’s killing them.

The journal Health Affairs reported the five-year drop in August. The article’s lead author, Jay Olshansky, who studies human longevity at the University of Illinois at Chicago, with a team of researchers looked at death rates for different groups from 1990 to 2008. White men without high-school diplomas had lost three years of life expectancy, but it was the decline for women like Crystal that made the study news. Previous studies had shown that the least-educated whites began dying younger in the 2000s, but only by about a year. Olshansky and his colleagues did something the other studies hadn’t: They isolated high-school dropouts and measured their outcomes instead of lumping them in with high-school graduates who did not go to college.

While there are no easy answers to explain the decline, Potts highlights research that found that whether women had a job was a major factor in how long they will live. In fact, it was only smoking and employment that seemed to matter. Women like Crystal–who went straight from high school to taking care of families of their own–missed out on the social networks, as well as that intangible “sense of purpose” that a job can offer. 

As Potts writes, “The rural South is a place that often wants to remain unchanged from the 1950s and 1960s, and its women are now dying as if they lived in that era, too.” Crystal’s relatives put it more simply still: they said she was “probably lonely.” And we know that loneliness kills.