Warning: Graphic Politics

By Evan Falchuk

A friend sent me this interesting graph from the blog of the National Geographic.

You’ll have to click on it to see a bigger version.  It captures a lot of data very elegantly on a single graph–  Professor Tufte would love it.

What it shows is health care spending per person across a group of countries, along with life expectancies, average number of doctor visits per year, and whether a country has a system of universal health coverage.  Although putting all of this data on one graph is novel, the graph makes what by now is one of the oldest political arguments for reform – for all the money they United States spends on health care we don’t get a good deal.

So why blog about this graph?

Because it terrifically illustrates so many of the flawed assumptions about health care that have plagued the reform process.

Spending graph First, pretty much nothing correlates with anything.

Does spending more lead to longer life expectancies?  It doesn’t look like it – and it’s not just the United States.  Canada spends almost $4,000 per person but gets roughly the same life expectancy Spain which spends about 35% less.  Luxembourg spends one and half times as much as South Korea, but their life expectancies are the same.  Meanwhile, Austrians spend three times as much as Poles, and live longer, too.  Life expectancies do not appear to be a good metric for health care quality.

What about going to the doctor more often?  That doesn’t seem to make much of a difference, either.  Japanese seem to the doctor almost monthly, but live about as long as the Swiss, who go once a year.  Oh, and the Swiss still spend about 70% more than the Japanese on health care.  Australians and New Zealanders go to the doctor about as often as each other, but Australians live longer, and spend about a third more on health care.  It doesn’t seem like rates of doctor visits tell you much about health care quality, either.

So maybe there’s something about having a “universal health coverage provided by public and private insurers.”  Nope.  If  you live in Hungary you’re going to live a decade less than your fellow beneficiary of universal coverage in Sweden.  Even among the countries in the top chunk of life expectancy, their relative spending levels and rates of doctor visits are all over the place.

So what’s going on?

Americans are often criticized for being ignorant of the cultural differences between our country and others.  We are seen as either wrongly assuming that people from other countries think about the world just like we do, or, worse, that we aren’t as enlightened as people living in other countries.  This chart, from an American organization, is fascinating in that it combines seeming ignorance of differences among other countries with ignorance of the United States.

Here’s what I mean.  The division of the world into “red” and “blue” countries, based on the idea of “universal” coverage is simply false.  For example, like the countries themselves, the health care systems of Canada, Britain and Japan are very different from each other.  Yet this graph treats them as if they were essentially the same.

But it’s also incorrect about the United States.  It’s true that the United States doesn’t have a system of “universal coverage provided by public and private insurers,” but the “United States” is the wrong entity to look at.  Health care coverage is mainly a state, not federal question in America.  And if we look at the states, we see a different – and more complicated – picture.

For example, Massachusetts has a system of “universal coverage provided by public and private insurers.”  It enjoys a life expectancy at birth of  80.2 years.  Sounds like one of the “blue” countries.  But Massachusetts spends $7,000 per person per year on health care.  On this chart it would be a puzzling outlier.

Hawaii has an almost universal system, but spends far less than Massachusetts per person.  And yet it has the highest life expectancy in the country.  It, too, would be an outlier.    Meanwhile, “red” New York, where almost 15% of the population is uninsured, ranks sixth in the country in life expectancy and spends about $6,500 per person on health care.  Another outlier.

At a certain point, the existence of almost nothing but outliers ought to make you question your initial assumptions.

Not so with health care.   The story that the United States gets ripped off in its health care spending sticks because it is a convenient oversimplification.  It’s a catchy slogan to advance (or with other slogans, oppose) difficult to understand changes to a complex system.

But the problem with health care in America is we keep getting lured into assumptions that money is the right metric to measure the quality of medical care.  How much we spend, rather than whether patients are getting the right care.  Money, over medicine.

He puzzled and puzzled till his puzzler was sore. Then the Grinch thought of something he hadn’t before! Maybe Christmas, he thought, doesn’t come from a store. Maybe Christmas… perhaps… means a little bit more!

Merry Christmas.

UPDATE: Via Instapundit, Economist Dan Mitchell says the problem is the rise of third party payers for health care.  He includes his own graph that implies that the trouble is people are paying too little out-of-pocket for health care.  In this view, since people spending other people’s money don’t care how much they spend, high US expenses can be explained in this way.  It’s an interesting idea, except that if you look at other countries, like Canada, where the system is almost entirely third-party payer and has been for a long time, costs are far lower than the United States.  And in any case, the people who drive most of our health care costs are spending at levels that far exceed whatever deductibles or co-pays you could ever put on them.  It’s not about out of pocket costs.

If there’s a lesson to be taken from all of this it’s this: beware of charts promising simple explanations for health care costs.

But if you want to look at a chart that shows the biggest drivers of health care costs, try this the one in this post.  The misguided focus on saving money over good medicine means we get neither.  The surest way to save money is to make sure that patients get enough time with their doctors, who need to be able to use their training and judgment to help their patients make the right decisions.  It’s a global problem, but America suffers from the most exaggerated version of it.

UPDATE 2: Jon Peltier does a terrific analysis of the same data with alternative versions of the National Geographic chart.

  • I totally disagree with yiou as there is no connection with the doctor's visits and good outcome.
  • VanessaOnTheRight
    Evan:
    Wonderful analysis! I heard some where (and am trying to look up the exact statistics to show my liberal friends) that some countries on the chart do not count miscarriages as an infant dying where we would. For example, the U S would consider a miscarriage at a certain number of weeks (let's just say 32) and a European country doesn't report it. Have you heard of that? I think that's one reason our infant mortality is higher. Another factor, of course, is that many of the European countries are very monolithic whereas the US has much more ethnic diversity so it's not an exact compare. Add to that we have a larger out-of-wedlock birthrate which again contributes to higher infant mortality. Please keep up the great analysis.
    Vanessa
  • Thanks, Vanessa. I'm not familiar with data on miscarriages. But it's correct, I think, that life expectancy is a function of overall public health, rather than an indicator of whether the health care delivery system is any good or not.

    Evan
  • As someone with a background in public health, I am unconvinced by the assumptions made in this graph. The countries with the highest life expectancy are not not necessarily the ones with the best doctor-patient relationships. They are reliably the ones with the strongest social safety nets, high literacy rates and high-performing educational metrics, lack of social conflict and/or the presence of cultural norms that promote healthy lifestyles.

    Medicine can play an important role in these things, but life expectancy and health and well-being overall comes from a much larger source.

    I agree with your deconstruction of the graph's popular interpretation, but I hesitate at your assertion of the alternative explanation. From my reading, it's the confounding factors that make all the difference here.
  • Thanks for this info
  • Hi Margaret, thanks for your comment.

    I agree with what you're saying, and didn't mean to suggest otherwise.

    Life expectancy - like the amount of money spent, or how often people go to the doctor - just isn't a proxy for the quality of medical care.

    The factors you describe so well drive things like life expectancy. Indeed, I linked to a chart of life expectancy by county in the US. You can see that the areas of the country that have the biggest problems with those issues tend to have the lowest life expectancy.

    Again, thanks for your thoughtful comment!

    Evan
  • Snowbat
    I disagree that there's no correlation between number of doctor visits and outcome. Did you notice that the thickest lines all slant upward (indicating more benefit per cost) while all the downward-slanting lines are thin? Some of the upward-slanting lines are also thin, like Mexico, so clearly there's more going on here than just number of doctor visits per year, but the graphic does seem to indicate that regular (prophylactic?) visits to a health professional can result in better long-term outcomes for less money. (Whether this is actually true is another question.)
  • I totally agree with you snowbat
  • Interesting question, I don't know the answer.

    The important issue for doctor visits, like many things, isn't quantity, it's quality.

    Average doctor visits in the US today feature face to face contact of 15 minutes or less - this is too short and both reflects and creates the fragmentation of care so prevalent in our system. Alas, this isn't just an American problem.
  • avaynshtok
    "First, pretty much nothing correlates with anything. Does spending more lead to longer life expectancies? It doesn’t look like it"

    Are you sure about this? Graphing life expectancy vs spending looks like this: http://drop.io/jruugl9/asset/spending-life-png

    Sure looks like correlation to me, w/ the US being a huge outlier. This makes intuitive sense if you look at the NG graph; pretty much all the countries w/ below average spending have below average life expectancies.

    Also, your point about states not being comparable is valid, but the HHS data you link to shows that even though per person spending varies by state, pretty much every single state would still be at the top of this graph for spending. So yes, while states aren't all the same, that doesn't make the US come out looking any better in this kind of analysis.
  • Hi,

    Thanks very much for commenting.

    I'm not sure your chart shows anything different than the original version (other than that you've removed the data on number of doctor visits).

    The United States spends way more than everyone else, while having life expectancies comparable to other developed nations.

    I get the desire to oversimplify health care. It's enormously complicated, and so it's appealing to look for simple umbrellas under which to put everything. But as a way of informing policy, it's a really bad idea, since it doesn't reflect the real world which policy will affect..

    In my post, I linked to a graph life expectancies across the US by county. You see that in some places, life expectancies are far lower than in others. There is a correlation with other data on rates of obesity, diabetes, smoking and other problems. Americans don't take very good care of themselves, particularly in those places.

    It's just one of the many important reasons why health care is so expensive in America. But in the rush to push through reform, any reform, policy makers are not addressing so many of the real drivers of health care cost.
  • MKirschMD
    As a physician, I cannot grasp the economic nuances of interstate health care and other national systems that you discuss in your post. Looking past some of the complexities you outline, it remains a startling statistic that we spend so much more per capita, and %GDP, than many other countries who have better statistics on longevity and infant mortality, two recognized measures of health. While we are doing so many things right, we are obviously doing many things wrong. My view from the trenches is that the system is wasting billions upon billions of dollars on unnecessary medical care. Some of this is defensive medicine. Some is demanded by patients and families. Some is for profit. Some is because physicians and the public love technology. And some is because we are in an era of medical excess; this is the way it’s often done today. I suspect that many of these forces are not present in other parts of the world. www.MDWhistleblower.blogspot.com
  • Thanks, Dr. Kirsch,

    It is striking that the US spends far more than other countries. The interesting question is "why," and it's not being asked very much in the reform debate.

    In fact, if the solutions proposed were based on an understanding of "why," they would look dramatically different than what has been proposed. You'd surely see more that would grapple with the issues you raise.

    At the end of the day, more people will end up insured under whatever passes, and that is a good thing. But it will also further entrench the existing system and the deep flaws in it that drive so much of our costs, and that's a very bad thing.

    Thanks again!

    Evan
blog comments powered by Disqus
  • "Medicine is learned by the bedside and not in the class room. Let not your conception of manifestations of disease come from work heard in the lecture room or read from the book: see and then research, compare and control. But see first."
    - Sir William Osler, MD
    The Father of Modern Medicine
  • Connect


    Via RSS


    On Twitter

    Subscribe via Email

  • Follow Us on YouTube:

  • Recent Posts

  • Recent Comments

  • Categories

  • Archives