The Silent Epidemic Killing White American Women

A momentous new report took up just six pages in a recent issue of the Proceedings of the National Academy of Sciences.

The title had no spoilers: "Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century." The authors were Anne Case of the Woodrow Wilson School of Public and International Affairs, and Angus Deaton of Princeton, the recent winner of the Nobel Prize in economics.

They presented a calm account of a social, political, and ethnic disaster that has struck the United States but no other industrial nation apart from the countries of the former Soviet bloc. In the 15 years between 1998 and 2013, about half a million Americans died years sooner than they should have.

They were middle-aged white people of European descent, the ethnic group and demographic that has ruled North America for close to 500 years. And while Case and Deaton didn't say as much, most of those half-million were likely women.

Case and Deaton argue that for the last half-century, Americans in general have lived longer and suffered less illness than they once did. So have people in other rich nations like Canada.

For some white Americans born between 1961 and 1970, however, something has gone wrong. They grew up in what should have been a wonderful time: the fall of the Soviet Union, the emergence of China as the world's economic powerhouse, the fading of the threat of a global nuclear war.

In 1998, however, some members of this age cohort began to sicken and die. It wasn't dramatic. The rest of the American population, like that of other rich countries, saw their life expectancy grow at two per cent a year -- including black Americans, who are still dying in greater absolute numbers than white Americans.

Dead when they should have lived

But this cohort's mortality began to rise at half a per cent a year. Not much, but it has continued year after year. By 2013, that meant 488,500 white Americans were dead when they should have lived.

"The post-1999 episode in midlife mortality in the United States is historically and geographically unique, at least since 1950," Case and Deaton say. During the same period, mortality in the same age cohort in six rich countries (and in American Hispanics), continued to fall. The same cohort of Canadians suffered less mortality than anyone but the Swedes and Australians.

What was the difference? Three causes: "Suicide, drug and alcohol poisoning... and chronic liver diseases and cirrhosis."

Other causes like diabetes barely rose above 10 deaths per 100,000. Lung cancer from smoking actually fell from over 30 to about 27 per 100,000. But chronic liver diseases rose sharply, as did suicides, and "poisonings" soared from perhaps two to 30 deaths per 100,000.

And who was dying? "The turnaround in mortality for white non-Hispanics was driven primarily by increasing death rates for those with a high school degree or less." Some post-secondary but no degree? Little improvement. A B.A. or more? "Death rates fall by 57 per 100,000."

"Although all three educational groups saw increases in mortality from suicide and poisonings," Case and Deaton say, "and an overall increase in external cause mortality, increases were largest for those with the least education."

In a recent commentary on Case and Deaton, Nobel Prize-winning economist Joseph Stiglitz argues that the crash of 2008-09 put many American families under intensified stress: they either lost everything or went deep into debt to hang on to their middle-class status.

But the stress had begun a decade earlier. Case and Deaton found suicides at the beginning of the period were higher in the South and West than in the Midwest or Northeast -- but everywhere, for every increase of one suicide per 100,000, two more died from drug and alcohol poisoning.

Even more striking, all white people between age 30 and 64 saw increases in mortality from the same causes; "the midlife group is different only in that the sum of these deaths is large enough that the common growth rate changes the direction of all-cause mortality."

Self-reported illness and pain also grew among white Americans, and so did the numbers reporting "serious psychological distress." This situation worsened by 2011 to 2013, and increasing numbers found that routine activities of daily life like walking a quarter-mile, climbing 10 steps, shopping and even socializing with friends had become more than "a little difficult. "The fraction reporting being unable to work doubled for white non-Hispanics aged 45-54 during this 15-[year] period."

'An epidemic of pain, suicide, and drug overdoses'

Case and Deaton call this an "epidemic of pain, suicide, and drug overdoses," and compare the toll to that inflicted by HIV/AIDS since 1981. Huge public-health efforts have managed to slow that epidemic, but the American public still has no real awareness of this new threat.

The report has predictably caused a stir, and one analyst, social-science blogger Andrew Gelman, has gone deeper into the numbers. He finds that the increase in deaths stopped in 2005: "Since 2005, mortality rates have increased among women in this group but not men."

"Actually what we see," Gelman says, "is an increasing mortality among women aged 52 and younger -- nothing special about the 45-54 group, and nothing much consistently going on among men."

Other researchers confirm the problem for women. A 2012 study found that white women without a high school diploma "lost five years of life between 1990 and 2008," by which time equally educated black women were outliving them. Earlier this year the Urban Institute published similar findings.

This seems counterintuitive. Perhaps it's easier to accept the idea of uneducated white men hitting middle age and finding themselves going nowhere. We can understand them hitting the bottle and the Oxycontin, and if those don't finish them off, hundreds of millions of firearms are close at hand.

But if middle-aged white American women are drinking, smoking, drugging, or shooting themselves to death, what are we to understand?

The cohort effect

Dr. Shannon Monatt, a demographer and rural sociologist at The Pennsylvania State University, said she thinks Gelman's analysis is solid.

"I wonder if this is a cohort effect," she wrote in an email, "something to do with the social, economic, and political conditions in the years they were born and grew up. This would be the group of women who were born between 1961 and 1970, reaching their teens between 1974 and 1983, and reaching early adulthood (graduating high school, heading off to college, starting families) between 1979 and 1988.

"So what made things different for women versus men during these periods? And how was what women were experiencing during these developmental periods different from the cohorts of women who came before them? A couple of thoughts, but I am still not sure how I think these relate to rising mortality.

"This was the cohort of women born during the sexual and social revolution. They were young children or just being born at the height of the civil rights/women's rights era. This was a time of great optimism, enthusiasm, and hope for women (including the mothers of this cohort).

"They were teens and reaching adulthood during pretty bad economic periods (mid 1970s-1980s) and may have ended up in family situations or career that are not what they expected when they were children. This was a generation of girls who were born during the start of second wave feminism. Perhaps they expected to see more progress for women than what we have seen.

"Their period of early adulthood was when women's labor force participation really skyrocketed. So a greater proportion of this cohort had to deal with the competing and stressful demands of work and family during a period in which men were still viewed as the primary breadwinner and women were still viewed as responsible for raising the children and keeping the house. Stress often leads to unhealthy coping mechanisms, including poor diet, excess alcohol consumption, smoking, and more recently, opiate abuse."

Your life depends on your class

As discouraging as these studies are, they largely vindicate a century of public health research that shows your very life depends on the class you're born into. Every social class is healthier and less stressed than the class below it, and sicker and more stressed than the class above it. Mobility between classes is far easier downward than upward.

It's not that poor, uneducated people don't know how to eat properly or keep themselves clean. They usually do; they're just too stressed to take care of themselves. As researchers like Dr. Richard Wilkinson have shown in book after book, income inequality within nations creates stressors on the relatively poor (even if they're far better off than the proverbial Bangladeshi peasant).

It's not even a purely capitalist problem. In his 1996 book Unhealthy Societies, Wilkinson says life expectancies and public health improved remarkably in postwar Europe's communist nations -- when incomes were reasonably equal. But by the mid-1970s the party elites (the nomenklatura) in eastern Europe and the Soviet Union began to enjoy a far higher standard of living, with higher incomes and privileged access to stores where Western goods were available.

The result was a general decline in public health, with sharp rises in homicides, chronic liver disease, and cirrhosis. The satellite nations abandoned communism by 1990, and by 1991 the Soviet Union itself was gone. The Russian oligarchs arose, income disparities widened dramatically, and male life expectancies plummeted as a generation of impoverished Russians drank and drugged themselves to death.

Now we can see that something very similar is happening in the apparent winner of the Cold War. When you feel that you're at a terrible disadvantage in your own society, you're stressed and you tend to self-medicate. It might be alcohol, or tobacco, or opioids -- or all of them. It might be beating people in your own family, just for a glimmer of a sense of control over something in your life, or to claim a scrap of what feels for a moment like respect.

If the men in this cohort aren't dying prematurely, then the effects of inequality have been literally fatal for the women.

Canada's mortality rates may be consolingly low, but for some Canadians they have always been high -- especially for aboriginal Canadians. Their incomes are a fraction of other Canadians', and their incarceration rates are much higher than others'.

In both countries, far more research needs to be done to understand -- and remedy -- the deadly consequences of inequality.