U.S. Leads Rich Countries in Mortality Rate for 10-24 Year Olds

In a depressing example of poor health outcomes in the United States, The Economist (via The Incidental Economist) shows that for 10-24 year olds, the mortality rate (deaths per 100,000 population) is considerably worse in the U.S. than in comparable rich countries (24 other OECD members plus Cyprus, Malta, and Singapore).

At about 59 deaths per 100,000, the U.S. rate was double that of South Korea, Sweden, Germany, Switzerland, Japan and the Netherlands, and almost triple the rate of Singapore. As the graph below shows, the U.S. has the highest level of deaths from traffic accidents (though not much higher than Portugal, New Zealand, or Greece). However, as the story points out, the U.S. has by far the highest rate of deaths in this age group due to violence: 8.9 per 100,000, which appears from the graph to be about three times the rate for second-place Israel. The story notes that the violent death rate in Britain is only 1/18th that of the U.S. for this age group.

The U.S. is further back on the suicide rate, where New Zealand is worst at about 12 per 100,000, followed by Finland at approximately 11, and Japan and Ireland about 10.

For "other" causes of death (disease plus non-auto accidents; thanks to commenter Marc Brown at The Incidental Economist for clarification of this point), Portugal narrowly leads the United States.

What does this mean for policy? Certainly automobile use is not going to change much any time soon, absent huge increases in gas prices and better mass transit and high-speed rail. Of course, as Thomas Friedman has advocated, the U.S. could force the price of gas up by raising the gasoline tax by $1-2 per gallon, but no one in a position to make this even possible advocates it. America's gun policies, similarly, will not change any time soon. According to a survey by Hepburn et al. (2006), Americans currently own somewhere between 260 to 305 million firearms. If the Supreme Court upholds the constitutionality of the Affordable Care Act, health insurance coverage will increase, perhaps slightly lowering mortality rates for some diseases.

In the near term, then, it appears unlikely that the mortality rate will change very much. Sorry that there's nothing less depressing to report. I'd be interested to hear your views on these issues, since obviously I have just scratched the surface here.

Basics: America's Relative Decline in Health in One Table

I've reported before on how the U.S. has seen less life expectancy growth than other rich countries, while at the same time increasing health care spending at a more rapid rate. Another way to see America's relative decline in health outcomes is through the evolution of life expectancy compared to all rich countries over a longer period of time. The World Bank's World Development Indicators database goes back to 1960 and its most recent figures are for 2009. We can see, then, what has happened over a 49-year period.

The table below includes all members of the Organization for Economic Cooperation and Development, excluding former Communist countries (most of which were not independent in 1960) and Israel (for which World Bank data does not go back to 1960). Of the 27 OECD members included, only four (Denmark, Iceland, Netherlands, and Norway) have seen less growth in life expectancy since 1960, and Sweden's growth has been the same at 8.3 years. Only two OECD members today have a lower life expectancy than the U.S., and both (Mexico and Turkey) are much poorer than the U.S. South Korea and Chile, both developing countries in 1960, have now surpassed the U.S. in life expectancy.

While health outcomes have certainly improved over the last 50 years, we can see from the table just how small those gains are relative to what other countries have been able to achieve. And remember, these gains have come at much greater economic cost.

Country Name		2009	1960	Increase
Australia		81.5	70.8	10.7
Austria		80.1	68.6	11.5
Belgium		79.7	70.1	9.7
Canada		80.7	71.1	9.5
Chile		78.8	57.0	21.7
Denmark		78.6	72.2	6.4
Finland		79.7	68.8	10.9
France		81.1	69.9	11.2
Germany		79.8	69.6	10.2
Greece		80.2	68.7	11.5
Iceland		81.5	73.4	8.0
Ireland		79.5	69.7	9.8
Italy		81.4	69.1	12.3
Japan		82.9	67.7	15.3
Korea, Rep.		80.3	53.0	27.3
Luxembourg		80.1	68.3	11.8
Mexico		76.5	57.0	19.4
Netherlands		80.5	73.4	7.2
New Zealand		80.3	71.2	9.1
Norway		80.8	73.5	7.2
Portugal		78.7	63.0	15.7
Spain		81.5	69.1	12.4
Sweden		81.4	73.0	8.3
Switzerland		82.0	71.3	10.7
Turkey		73.4	48.3	25.2
United Kingdom		80.1	71.1	8.9
United States		78.1	69.8	8.3

U.S. Health Care #1? Color Me Dubious, Too

Aaron Carroll has a good catch on The Hill's "Healthwatch" blog giving an uncritical eye to a recent study by a British conservative thinktank, the Legatum Institute. Its 2011 "Prosperity Index" ranks the United States #1 in the world for health. Not only that, the U.S. is far ahead of #2 Switzerland, with an index of 3.54 vs. 3.03, over 16% higher. Problem is, the U.S. scores only so-so on what both Carroll and I would consider the most important variables, healthy life expectancy (27th) and infant mortality (36th), does pretty well on some measures (tuberculosis, sanitation) and poorly on others (respiratory diseases), and only ranks #1 on one of the index components: health care spending. Since the U.S. has been seeing worsening bang for our health care spending bucks, this does not compute.

As Carroll notes, it is not clear how you weight these variables in such a way that the U.S. could possibly come out #1, and the explanation gives no real clue of the relative weights, just longer and shorter lines that are supposed to tell us the relative weights without any actual numbers. More confusing still (and I think Carroll misses this), the health variables have weight in both "income" and "well-being" regression analyses. While health spending per capita is more highly weighted than life expectancy and infant mortality on the "well-being" side, as Carroll notes, on the "income" side the reverse is true. Not that the weights, whatever they really are, make sense.

Searching the prosperity.com website for the promised detail on the methodology, I was unable to find any rationale for the choice of variables (let alone their weights) beyond the bland claim that the eight sub-indices were based "on decades of empirical and theoretical research done by established experts." That certainly clears it up. More curiously still, in response to why it uses such a complicated methodology, the FAQs answer, "The Prosperity Index is an assessment of the DRIVERS of prosperity. We do not seek to identify which countries are prosperous or not." I would have thought the latter was the whole point, given that they have already imposed a concept of "prosperity" that is 50% income and 50% "well-being." Compare this with the UN's Human Development Index, which is 1/3 income, 1/3 health, and 1/3 education.

So I'll email the folks at prosperity.com and see if I can find out the answers to these methodological questions. Meanwhile, like Aaron Carroll, color me dubious, too.

US health care bang-for-buck declining over time and relative to other rich countries

Lane Kenworthy has a great post (http://lanekenworthy.net/2011/07/10/americas-inefficient-health-care-system-another-look/, h/t Matthew Yglesias) comparing health care spending and life expectancy. He goes beyond a simple scattergram of rich countries on the two variables to look at what has happened over time in 20 OECD member states. It plots how each country's health care spending per capita and life expectancy have increased over time.

Notice how divergent the U.S. curve is. Not only does it have lower life expectancy than the other 19 countries graphed, it is gaining less in life expectancy for each dollar per capita of increased health care spending. As Kenworthy notes, this is especially true "after the early 1980s when we reached expenditures of about $2,500 per person (in 2005 dollars) and life expectancy of around 74-75 years." Since then, spending in the U.S. has more than doubled to over $6,000 (constant 2005 dollars), but we have only gained four years or so in life expectancy. Japan only spends about $2,500 per capita today, yet has achieved a life expectancy of about 83 years.

 Kenworthy points out that using changes rather than levels of life expectancy and spending lets us factor out some of the differences between countries, citing higher U.S. murder rates, obesity rates, and geographic dispersion. I'm not sure these are such big drivers of differences in life expectancy; one major factor is the difference in infant mortality rates. According to the CIA World Factbook's 2011 estimates (https://www.cia.gov/library/publications/the-world-factbook/rankorder/2091rank.html), the U.S. has an infant mortality rate of 6.06 per 1,000 live births; the other 19 countries are all below the U.S., with Sweden and Japan having rates of less than half that. Moreover, geographic dispersion does not seem to make much difference within the United States: Of the top 20 states for life expectancy (http://en.wikipedia.org/wiki/List_of_U.S._states_by_life_expectancy), nine of them are also among the 20 least densely populated states (http://en.wikipedia.org/wiki/List_of_U.S._states_by_population_density), suggesting that if there is a correlation, it is likely to be small.

That said, Kenworthy's chart gives us a striking illustration of the inefficiency of the U.S. health system and how its relative efficiency has declined over time.