Introduction:
Recently, “falling US life expectancy” or “US life expectancy falling behind other wealthy countries” has been a popular topic. Charts like this one have been circulated widely:
A few months ago I asked the following question on Twitter: What explains graphs like this? Especially since many of the common explanations, like the United States having for-profit health care and lots of guns and meat and cars and suburbs etc, applied well before 1990 or 1995?
Another common narrative is to emphasize drug overdoses, suicide, and liver problems as major factors, under the alliterative catch-all of “deaths of despair”. I feel like I’ve seen a lot of articles with headlines like “deaths of despair explain one-third of the life expectancy gap”, before I wondered what explained the remaining two-thirds.
Either way, the contributions of different causes of death to graphs like that aren’t always quantified clearly. The goal of this series of articles will be to try to think through the question, as an admited beginner. In this first article, I’m going to start by trying to define the question as clearly and precisely as possible.
What Should The “Other Rich Countries” Be, Exactly?:
Let’s start by comparing all countries by “life expectancy”. The following graph shows the life expectancy of countries and territories, in 1990 and 2017, from the Global Health Data Exchange (which will be the main data source for these articles; subsequently “GHDx”).
That’s a bit crowded, and it’s a bit hard to see how off-trend the United States is or isn’t. But remember, the question is specifically about United States life expectancy compared to that of “Other Rich Countries”.
What should the “Other Rich Countries” be, exactly? This is the first step to making the original question more precise. When I first Tweeted about this, there was some concern that the chart in the introduction could be misleading if it included a changing mix of countries and/or countries with unusually rapid “catch-up growth”.
To address this concern, I want to make a fixed set of “Other Rich Countries” that were “similarly relatively rich” in 1990 and in 2017. The World Bank has GDP per capita in PPP (international) dollars for most countries for 1990 and 2017. (Note that the scales are not identical here, since my goal was to show relative order in both years.) Portugal seems like a reasonable “natural break”, with the countries above Portugal in either 1990 or 2017 more-or-less constant, and more-or-less in the same relative order.
I am somewhat arbitrarily excluding “petrostates” (because I suspect GDP conveys less information about public living standard) and territories (because they feel like “policy special cases”). Of the countries that have listed life expectancies from GHDx but lack GDP numbers from the World Bank, I believe only Bermuda and Andorra would have made my arbitrary cutoff.
That leaves my set of “Other Rich Countries” as: Andorra, Australia, Austria, Belgium, Bermuda, Canada, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Iceland, Ireland, Israel, Italy, Japan, Luxembourg, Netherlands, New Zealand, Norway, Portugal, Singapore, Spain, Sweden, Switzerland, the United Kingdom; along with the United States. This seems like a pretty reasonable set. (It’s basically Western Europe, the Anglosphere, and Singapore/Japan/Israel/Cyprus.) No offense is intended to any excluded countries.
What does the life expectancy chart look like zoomed-in on “Other Rich Countries”?
That’s a bit more clear-cut. The United States is indeed off of the relatively tight trendline of “Other Rich Countries”.
To check this, I used the “Fergany method” to calculate the annual life expectancy for this “Other Rich Countries” aggregate. (This is probably a significantly simpler method than was actually used by the GHDx but it should be fine for illustrative purposes.) Their life expectancy improved by about ~5.6 years from 1990 to 2017, from 76.7 to 82.3; United States life expectancy improved by only ~3 years, from 75.6 to 78.6.
(Note that the United States is not alone on its own trendline, Costa Rica and Mexico and Malaysia and Panama and such are on a similar trendline, but it is alone among “Other Rich Countries”.)
The original question can therefore be phrased more precisely as:
“Why did United States life expectancy improve by 2–3 years less than the life expectancy of Other Rich Countries [as defined above] from 1990 to 2017?”.
In other words, the question we’re asking is about differences in rates of change of life expectancies — it’s kind of inherently a second-order or third-order question. Even in 1990 the United States was on the low end of “Other Rich Countries” by life expectancy. The difference isn’t new, at least not with these countries and not over this time scale. What’s new is the change in the difference, that the United State improved more slowly than “Other Rich Countries”.
Finally, is this driven by one or two countries, and/or by the particular choice of years? I don’t think it is. The following graph shows life expectancy for the United States and each of the “Other Rich Countries”, along with my computation of the life expectancy of “Other Rich Countries” taken as a single population:
Pretty much every “other rich country” than the United States had life expectancy improve more quickly than in the United States, and the life expectancy gap between the United States and the aggregate of “Other Rich Countries” grew over most of this time period, although it certainly accelerated over the last decade or so.
Conclusion:
Putting all of this together, we can see that, compared to a fixed set of “other rich countries”, United States life expectancy indeed fell off trend, rising more slowly and even falling recently. This trend doesn’t seem to rely on a particular choice of years or countries.
Before we try to explain that further— wait, what even is “life expectancy”, anyway? See Part Two.
Notes And Sources: Public health data from:
Global Burden of Disease Collaborative Network.
Global Burden of Disease Study 2017 (GBD 2017) Results.
Seattle, United States: Institute for Health Metrics and Evaluation (IHME), 2018.
Available from http://ghdx.healthdata.org/gbd-results-tool.
GDP numbers from the World Bank as I said.
