Mortality Among High-School Educated Whites in the U.S.: An Anthropological View

Fig. 1. All-cause mortality, ages 45–54 for US White non-Hispanics (USW), US Hispanics (USH), and six comparison countries: France (FRA), Germany (GER), the United Kingdom (UK), Canada (CAN), Australia (AUS), and Sweden (SWE).

Every so often a piece of research comes along that is a real game-changer—it literally shakes the earth under your feet.  I had that experience about a year ago when Anne Case and Angus Deaton, two economists, published an analysis of recent mortality trends in the United States.  If you electronically search “mortality trends” for the U.S., you will see that, overall, mortality rates are declining, as they are for Canada and Western Europe.  What Case and Deaton did was to separate out mortality rates for non-Hispanic Whites.  Starting about the year 2000, rather than continuing to decline like everyone else’s, mortality for this group bucked the trend and started to climb.  When causes of mortality were examined, deaths from lung cancer were declining, from diabetes were stable, and for three causes were climbing, dramatically.  These were chronic liver disease, suicide, and what Case and Deaton refer to, by default since that’s what the feds say, “poisonings” (read: unintentional drug overdose).

I’ve taught epidemiology off-and-on for a long time, and mortality rates just don’t jump around like this, unless, that is, something catastrophic is happening.  An example of a catastrophic event leading to high mortality rates was the fall of the former Soviet Union.  In the decade following that political upheaval mortality—especially male mortality—climbed, fueled by a potent combination of vodka and cigarette smoke.

A further component to their findings was that the changes in mortality rates were highest among non-Hispanic Whites who had a high-school education or less.  In 1999, rates of death from “poisonings” were 4 times higher for people with a high-school education or less than they were for people with a college degree.  In 2013 those death rates were 7.2 times higher for the less-well educated versus the well-educated.

I felt so compelled by this evidence that I dropped what I was doing in my classes—one on cognitive anthropology and one on the history of anthropological theory—and taught the Case and Deaton paper.  Even though it caught a lot of attention in the national press, at least for awhile, I was afraid that it would escape the notice of many of my students and, furthermore, that they might not really appreciate the magnitude of the results.

The pattern of results suggests that non-college educated Whites are experiencing some kind of profound stress and that in response they are self-medicating with alcohol (hence chronic liver disease) or with prescription opioid pain medication (with its attendant risks of overdosing), and that they are responding with major depression and the associated risk of suicide.  In their interpretation, Case and Deaton emphasized the stress of economic insecurity for working class Whites, noting that widening inequality might account for the trend.  This would seem to me to affect other population groups—like African Americans—even more than non-college educated Whites, yet the trend toward higher mortality from these causes is not observed in other population groups.  Case and Deaton also suggest that it might be specifically the transition in retirement programs from guaranteed benefit plans to defined-contribution plans, with their associated stock market risk.  In this interpretation, looking forward to an uncertain and possibly impoverished future is the source of stress.

For obvious reasons—and I’m talking about the election season in which we find ourselves—I have continued to think about this research, given the prominent place that Whites with a high school or lower education seem to be playing in support of one of the major candidates.  Is it, to quote a political strategist from a past campaign associated with the other major candidate, “the economy, stupid!”  Or, is it that, and something more?

As I considered the findings, I was reminded of an old paper by James P. Henry and John C. Cassel from the American Journal of Epidemiology in 1969.  They examined cross-cultural data on age and blood pressure, noting that, while many physicians believed the rise of blood pressure with age to be “natural,” it was in fact “cultural.”  In many communities around the world, especially those that had yet to be drawn very closely into capitalist, market economies, there was little evidence of an increase of blood pressure with age.  In what were called (back in the day) “modern” communities, blood pressure rose with age.

To explain these results, they drew on a process that Cassel had been thinking about for some time, namely, the inconsistencies and incongruities that can accompany profound culture change.  Cassel’s preferred research strategy had been to follow migrants into a new setting, where he predicted that the incongruity between the culture they arrived with, and the culture of their majority host community, created a period of stressful and taxing adaptation, as the migrants tried to adjust to their new setting.  The end result of this stressful adjustment, especially if it was not especially successful, was an increased risk of disease.  Henry and Cassel suggested that the same process could be occurring across the life-span of an individual, arguing that in the modern world, with the ever increasing pace of social change, an individual is born into and socialized in one culture, yet ends up living in another, as the world changes around him or her.

This strikes me as an eminently plausible interpretation for the Case and Deaton findings.  Non-college educated Whites are indeed facing economic stresses, but the broader cultural changes they are experiencing are even more profound.  And what can more effectively and graphically communicate to them that the world around them has changed than the fact that they will shortly trade their first African American president for their first female president?  (I trust Sam Wang and his Princeton Election Consortium.)

Cassel drew heavily on culture theory for his insightful interpretation of epidemiologic data.  Case and Deaton’s findings suggest that those insights are still relevant.


Biocultural Systematics is written by members of the University of Alabama Biocultural Medical Anthropology program.

Bill Dressler, a professor in the department, has conducted research in social epidemiology in Brazil and the U.S.

Leave a Reply

Your email address will not be published. Required fields are marked *