• Income related inequalities in health care

    Before he went on vacation, Austin pointed me to an OECD working paper, entitled “Income-Related Inequalities in Health Service Utilisation in 19 OECD Countries, 2008-2009.”  I’ve found it to be quite interesting. There’s one figure I’d like to highlight for all of you:

    What you are looking at is the needs-adjusted probability of having had a doctor visit in the last twelve months. For each country, the probability is shown for people in the lowest quintile of income, average income, and the highest quintile of income.

    I’m going to ignore Denmark for the rest of this discussion, because for some reason the probability for that country was in the last 3 months, versus 12 months for every other country.

    The first thing to note is that the average rate of a visit to the doctor varies among all these countries from a high of 91% in France to a low of 68% in the United States. Think about that the next time someone tells you how our problem is that we consume too much health care.

    The second thing to note is how much variation there is between those at the upper and lower end of the economic spectrum. In the UK, for instance, there is almost no difference in utilization between the rich and the poor. All see the doctor equally. In most other countries, though, there is some inequality based on income.

    None as great as the United States, though. The difference between the probability of seeing the doctor for the poor and wealthy is greater in the US than in any of the other measured countries.

    People like to believe that we don’t ration care in the US. We do. More than just about any other country, we ration by cost.


    • Typo in the second paragraph – twelve months, not years.

      Interesting data, thanks for the tip!

    • This diagram says it all.
      Even our rich have poor access and if you are not rich you are out of luck.
      Our corporatocracy has destroyed our country.

    • It’s crazy and sad and bewildering that the US, which was to some degree designed as a critique of the injustices embedded in the social organization of those ancien regime nations, is now, with respect to the equality of its own citizens, sputtering along behind them, particularly the country which our founders originally separated from.

      • True, but in reality the US has always defined injustice in a different and more limited way than most Europeans and other modern developed nations.

        The US tends to focus on the civil rights that were emphasized by 18th century thinkers — freedom of speech, religion, assembly, and in general freedom from political tyranny. Modern developed countries other than the US recognize those rights as well, but place a much stronger emphasis than us on economic freedoms — food, shelter, clothing, education, and health care. They are and were more heavily influenced by social and economic thinkers who came into prominence in the 19th century.

        The US has come to the idea of addressing those economic freedoms late and weakly by international standards, and now stands in sharp contrast to most other states. The dismal state of our health care system, the weakness of our educational system, the high levels of economic inequality and poor levels of economic mobility, our weak workplace safety, consumer protections, and environmental safety, and the high levels of real poverty and poor government response to poverty all are examples of how we compare poorly to most other countries where attention to those things is part of the consensus political position.

        This chart is just one more example, but there are many other things that would show the same skew.

    • In a study like this, devil’s in the details, and in particular the needs adjustment is tricky.

      How do you disentangle physician visits and patient need? Do we really have good survey data for morbidities for each of these countries sorted by income quintile that doesn’t rely on healthcare access? It has to be independent of healthcare access, because this study is otherwise just running in circles.

      Do you go around calling people and seeing if they have diabetes? How will they know if they have diabetes if they haven’t visited a doctor in two years?

    • It looks like another case where looking at the top quintile is not an adequate way to understand life for the rich (in the US). The top quintile still mostly rely on health insurance from work with co-pays, deductables, fear of coverage denial and the rest. To find the healthcare habits of the rich you need a cut off high enough to account for people who can pay out of pocket for care. Doubt even the top 1% covers that, more likely top 0.1%.

      Still an excellent chart, but analysts have to start understanding the most significant US inequality doesn’t show up in quintiles or even deciles.

    • You’re welcome for the ‘tip’…by the bye; many European countries that thought they had solved the problem of inequality in access to care are discovering growing gaps between income groups, regions, and ethnic and racial groups. For the last category there are real problems in grasping the dimensions of the problem due to political and cultural tabus.

    • The whole paper is worth a read, though it is considerably less exciting if you just want an excuse to bash the US, which does OK on some measures. (No, I don’t think the US has the best HC in the world….). It ends with a brief survey of insurance systems across the OECD, which I appreciated.

    • I am confused, because I have seen rates of doctor visits per year that are far higher than these. Here’s one example:


      This purports to come from OECD data. They are roughly 10-fold higher than the figures in your graph.

      Any idea what’s going on?

    • I don’t think this rules out the over-utilization argument. I would like to see this data compared to the primary care physician:specialist ratio in each country. We are top-heavy with specialists. Extensive data from the Dartmouth Atlas shows that more specialists/area usually doesn’t mean healthier people, and sometimes it means the opposite (http://www.dartmouthatlas.org/keyissues/issue.aspx?con=2937).

      This could also be indicative of a different view of health care. We view things as an “oh crap” system, where we get care when something goes wrong (http://www.chron.com/opinion/outlook/article/Texas-has-top-medical-centers-but-provides-poor-2174885.php). Other countries could have a more public health/prevention focused system.

      Also, AHRQ has shown that the sickest 5% of patients consume about 50% of healthcare resources, while the bottom 50% consume almost 3% ). That’s a very small segment receiving most of the care. Are they receiving high quality care? Or just lots of care?

      I like the post, and I do think that access to healthcare is a problem, but I don’t think you can just say this explains away overconsumption.

    • I like this post, these are interesting data, but I don’t see that you’ve convincingly demolished the argument that we consume too much health care. First of all, these data do not account for the 45 million Americans who are uninsured and consume half the health care of the insured. That’s one of the reasons our physician visit rate is low. The other reason has already been discussed by Justin above: we are top heavy with specialists and light on PCPs.

      Second, just because the French go to the doctor all the time doesn’t mean they a) need to see the doctor all the time, and b) are healthier because of it. Who says the French rate is the right rate of visits?

      That said, there is clearly a lot of underuse of effective, needed care in the U.S. This problem coexists with the overuse of ineffective, unnecessary care, particularly invasive procedures, imaging tests, and end of life rescue care.

      • My intent wasn’t to demolish that argument. I said to think about it the next time someone says (with no data) that we consume too much. I have other posts which show that utilization has been going down even as spending has gone up.