by Peter Franklin
Wednesday, 7
April 2021
Chart
11:23

America’s disturbing death statistics

A new study shows that class, not race, is a bigger predictor of an early death
by Peter Franklin

Before Covid, there was another epidemic — the opioid epidemic.

In fact, as Anne Case and Angus Deaton of Princeton University have shown, the problem isn’t only the fatalities caused by opioid misuse, but also other forms of drug overdose, suicide and alcohol-related liver disease.

Described by Case and Deaton as “deaths of despair”, these scourges have had a measurable impact on mortality trends in America. Indeed, the problem was serious enough to send a century-long trend of ever-increasing lifespans into reverse.

It is not an equally distributed phenomenon. The rise in deaths of despair is overwhelmingly concentrated among poorer Americans. The size of this effect is shown in new research from Case and Deaton published in the Proceedings of the National Academy of Sciences.

The authors analyse figures for “expected years of life from 25 to 75”. In a population where everyone died aged 25, the expected years of life would be zero. If, on the other hand, everyone reached their 75th birthday then the figure would be 50. 

In the following chart, we see a long-term increase in expected years of life that peaks in the early 2010s and then, disturbingly, starts to decline. That’s bad enough, but the authors then present separate figures for people who have a four year degree (shown on the chart as “BA or more”) and for those who don’t (“No BA”):

Expected years of life from 25 to 75 by BA status. Credit: Proceedings of the National Academy of Sciences

As you can see, the whole of the decline is concentrated among the less educated part of the population. Among graduates, expected years of life continues on an upward trend. In other words, class-based inequalities are widening.

But what about racial inequalities? In another chart, the authors show separate trends for black and white Americans (also broken-down by education and sex):

Expected years of life from 25 to 75 by race, ethnicity, and sex.

This shows that racial inequalities certainly exist — with black Americans at a disadvantage to their white counterparts. However, the racial gap has narrowed considerably for both men and women over the last three decades — in clear contrast the widening class gap. That’s an irony given just how intensely focused US politics is on race right now. 

Of course, lifespan isn’t the only measure of inequality. Nor does any recent improvement in racial justice erase centuries of slavery, segregation and racism. 

Nevertheless, on this most basic measure of well-being, the class gap (as signified by level of formal education) is now much bigger than the race gap. As the authors put it: “those with a college diploma are more alike one another irrespective of race than they are like those of the same race who do not have a BA.”

It’s clear that something has gone very badly wrong for working-class Americans of whatever race. While Covid-19 is likely to bury this underlying trend under a flood of excess deaths, it doesn’t mean that America’s social emergency has gone away.

Join the discussion


  • I think that the above graphs are just trying to make an impact where there is no impact – playing with figures.
    The BA graph looks interesting but the whole of the difference between the categories is only about 2.5 years. Why is it surprising that people with BAs live 2.5 years longer? They are likely to have a better job, they think more, they maybe read a little, they talk about health. It is amazing, in fact, that the spread is only 2.5 years.
    For the hispanic/black/white, the education level needs to be factored in for it to be meaningful.

    The Hispanic result is interesting and is worthy of further study. For me, the annoying thing is the men vs women. All discussion we see is that women have a terrible life compared to men. What a con !!!!

  • One of the things I hate in medicine (I am a doctor) is this constant background hum about “the social determinants of health”. Those on the left-hand side of the political spectrum (ie: 90% of younger medical doctors) have a very simplistic understanding of the issue of how wealth relates to mortality. They think that if you give poor people more money, they would be healthier. This is “obvious” because richer people live longer.
    But for anyone with even a modicum of intelligence and objectivity, they realize that you need to correct for factors like smoking rates, substance abuse, eating habits, etc etc before you can tell if wealth is an INDEPENDENT predictor of mortality. And the answer seems to be – from what is available, as most of the research is biased towards “proving” that poor folk are victims – that wealth is a very minimal predictor.
    And even if one accepts that it is a minimal predictor, the question remains: What is the mechanism? How are one’s coronary arteries directly affected by the number of zeroes in one’s bank account? How is this possible? To me, the likelihood of this being the case is small. The most reasonable explanation I have ever heard is the chronic stress theory (poor people suffer more chronic day-to-day stress, which leads to higher baseline cortisol levels, which negatively impacts cardiovascular health amongst other systems). But even that is tenuous, as I meet a lot of very stressed and depressed rich folk, and many happy poor ones, and suicide rates don’t predict that well by social class.
    To me, the mortality gap between classes is much like the gender wage gap. It looks impressively large until one opens one’s brain to the obvious causes of this disparity, causes that are verboten to discuss as they destroy the “Victim” theory.

  • Looking at the data by race, the most interesting point is that hispanics have a life expectancy higher than whites by about the same margin of whites over blacks. So if America is institutionally racist, have we all gotten it wrong and it’s actually designed to favour hispanics rather than white or black people?

  • To get involved in the discussion and stay up to date, become a registered user.

    It's simple, quick and free.

    Sign me up