Recently an enterprising AP reporter, Stephen Ohlemacher, has made waves with a widely reprinted article. He claims that the USA has been sliding down the world rankings for life expectancy at birth, a key indicator of the health of its population. The USA now ranks 42nd out of 224 political entities in the US Census Bureau database, “down from 11th two decades earlier”.
I couldn’t find any recent Census Bureau publication on the subject, so Ohlemacher must be basing this on his own analysis. Annoyingly he hasn’t SFIK posted any spreadsheet on the Web. So to show the MSM how it can be done – and should be done routinely – I’ve tried to provide RBC readers with a reverse engineered sheet from downloads from the Census Bureau. Here it is. I have thrown in infant mortality as well.
My conclusion: the decline story is dodgy, but the data confirms the ongoing mediocrity of the US health care system compared to its peers.
The Census Bureau’s online database doesn’t go back more than 10 years so I had to use a 10-year rather than a 20-year comparison. My sheet gives the 2007 ranking as equal 44th, close to Ohlemacher’s. But the decline over ten years was only from 41st. (The article concedes that he wasn’t using a comparable list of countries for 1987). But how do you interpret the observation?
The first problem is that this set of countries includes swarms of micro-states like Kiribati and even tiny dependencies like St. Helena (population 6,563). This is good fun but pretty irrelevant, and allows wingnuts to nit-pick (“They want us to copy Andorra!”) A more sensible universe of comparison for the USA is the 30 member states of the OECD, the club of rich capitalist countries; the smallest member is Iceland, with 310,000 inhabitants already the size of a US city.
- Life expectancy: the USA ranks 19th out of 30, and
12four (Oops, sorry) years behind the leader, Japan.
- Infant mortality : the USA ranks 25 out of 30 at 6.4/ºº, over twice the rate of Japan and Sweden.
- There is no significant trend in the US rankings over the last decade.
So in spite of spending far more per head on health care than any other major country, America’s health care is stuck in the Bush league. Significantly, the US ranks lower than the UK on both metrics and for both dates, in spite of the notorious parsimony (until recently) of British socialised medicine.
It is striking how little convergence there has been over a decade in the absolute values. The worst countries typically added a year more to life expectancy than the best ones, but the latter are still improving. In infant mortality it’s hard to detect any narrowing of the gap except for Mexico and Turkey at the very bottom. At the top, Japan reduced infant mortality by a greater absolute amount than the US.
I think we should pay more attention to the infant mortality figure. Life expectancy is an objective but highly artificial indicator: calculate the death rates last year for every age slice of the population, and assume that a baby just born will be subject to those rates for every year of its life. The mean age of death that results from the simulation is the life expectancy at birth. So it captures a lot of past history, including the limitations of the medicine of the past, and not the lower risks the baby is likely to face. (So Ohlemacher is mistaken when he writes “A baby born in the United States in 2004 will live an average of 77.9 years”.) But if the baby faces poor care at birth, this will worsen her lifelong risks.
The real alarm bell for the USA from these statistics is the combination of its low infant mortality ranking, high absolute value, and its disappointing progress over time. The country is therefore very likely to slide further down the life expectancy table over the coming decades.