Not Stayin' Alive

This is not new news, but it's strong confirmation of earlier observations that have been somewhat controversial, and also bad news that the trend is continuing. That trend is declining life expectancy in the U.S. I'm not linking to the full report in JAMA because it's incredibly wonky and behind a paywall anyway, but rather to the associated editorial, which tells you what you need to know.

Before we get into the substance of this, let me explain the concept of life expectancy. I'll try to put this simply, but some people find it confusing. It's really a fictitious, though useful, construct. It isn't really a prediction of the future, but it tells us something about the present. Below is a portion of what's called a "life table." It shows the proportion of people in the U.S. who died in each 1-year age cohort in 2015.





As you can see, .006383, or a little more than 6/1,000 male babies died before their first birthday. (A simplifying assumption is that all deaths occur on June 30. Don't worry about it.) Of the remaining male babies, .000453, which only 4/5 out of 10,000, died before their second birthday. So it's somewhat dangerous to be an infant but once you get past that you're in pretty good shape until you get old. It isn't until age 81 that the probability of death for men goes up to the same level as the infant mortality rate.

So life expectancy at birth is the average (mean) age you would live if you were born in 2015 and experience the same probability of death each year as people who are alive in 2015, i.e. assuming nothing changes in the next 81 years or so. That obviously isn't very likely, so as I say, this is a fictitious concept; it's about the present, not the future. You might also notice that half the men are still alive at age 80 but male life expectancy in 2015 was actually a bit lower than that because that 50% represents the median, and as I say, life expectancy is presented as the mean, which was actually 76.4.* You will also notice that women live longer than men. Life expectancy for Black people of both sexes is lower than that for white people, but black women still live longer than white men.

Okay, so as the linked editorial explains, the decline, which began in 2014 but  followed an earlier plateau, is largely traceable to increased death rates for people in mid-life -- age 25-64, who during the post-war years had been quite unlikely to die. The increased death rate in those years is attributable largely to opioid overdose, alcoholism, and suicide, but not entirely. Obesity-related diseases also contribute. The industrial Midwest, Appalachia, and Northern New England were worst-affected, as were rural vs. urban areas.

Many people are labeling these "deaths of despair," and arguing that they are linked to the economic and social disruption associated with deindustrialization and the declining rural economy. That may be so, although the opioid epidemic is a more equal opportunity phenomenon. How the pain of these places has translated into changing political loyalties, however, is harder to explain. One thing is for sure -- it is not caused by immigration. It also is not caused by abortion, gay marriage, transgendered people, increasing numbers of minority elected officials, or gun safety legislation. I get that people and communities are hurting, but I don't know get why their anger is so misdirected.

* The relatively large number of people who die in infancy drags the mean down more than the small number of very long-lived people pulls it up. This will make sense to you if you have studied statistics, otherwise don't worry about it.