Apr 17, 2011

I made a graph

I made this graph in order to try to answer the question:  Is a public health system more expensive than a private one?  The x-axis is government expenditure on health as a percentage of total health expenditure.  The y-axis is total health expenditure, as a percentage of GDP.

The orange diamond represents the United States.  Data is from 2007 and compiled from the World Health Organization 2010 report.  The solid blue line is a linear fit.

The countries shown are from the top 27 countries by GDP per capita. Not shown are those countries which spend less than 4% GDP on health:  Singapore, and the Islamic/Arabic monarchies Qatar, Brunei, United Arab Emirates, and Kuwait.  Also not shown are countries which receive external resources for health:  Israel.

List of countries shown:  Luxembourg, Norway, United States, Switzerland, Netherlands, Australia, Austria, Canada, Ireland, Sweden, Iceland, Denmark, Belgium, Germany, United Kingdom, Finland, France, Japan, Republic of Korea, Spain, Italy.

Thoughts?

13 comments:

  1. The cultures of these countries are very different than those of the U.S. with much more homogenous populations.

    ReplyDelete
  2. And much smaller too.

    ReplyDelete
  3. i really do think the rather diverse and much greater population is what makes a huge difference. Due to our size gives some areas a much greater cost and gives some health care suppliers chance to skrew you over.

    ReplyDelete
  4. First, suppose we ignore the data point representing the U.S. Doesn't there still seem to be a correlation between increased public health spending, and lower overall health spending? Is there any evidence to back up the claim that public health systems are more expensive than private ones?

    Second, which kind of diversity are you talking about? Linguistically, the U.S. is not as diverse as Canada, Spain, Luxembourg, or Belgium. I would imagine linguistic diversity would be a major headache for an efficient health system. Why would racial diversity affect health system efficiency?

    ReplyDelete
  5. List of countries life expectancy
    31 Luxembourg 78.7
    13 Norway 80.2
    36 United States 78.3
    4 Switzerland 81.7
    16 Netherlands 79.8
    5 Australia 81.2
    16 Austria 79.8
    11 Canada 80.7
    29 Ireland 78.9
    7 Sweden 80.9
    3 Iceland 81.8

    ReplyDelete
  6. Hey Anonymous, what does your life expectancy data indicate about the relative costs of a public vs. a private health system?

    ReplyDelete
  7. Freyguy,

    That graph is definitely very interesting. The United States pays way too much for health care and without getting the added benefit that should expected from that.

    There is another graph that I would like to see and it is healthcare expenditure vs. average population Body Mass Index. The American lifestyle is in general not healthy and I think adds a lot of cost to our system. This would assume that Americans are much more obese than the other countries on that graph, which seems to be true, but I don't know for sure.

    The other graph that I think would shed some light onto our problem with healthcare spending is a graph of percentage of high-cost tests/procedures vs. total healthcare expenditure. One of the problems with our system is that the payment is better for those things and preventative medicine is reimbursed very poorly. An analogy is that we don't pay for the oil changes but rather for when the entire engine needs to be replaced.

    At the end of this post, I realize that I have not really agreed or disagreed with your post per se. However, I definitely agree that something needs to be done.

    -Colin

    ReplyDelete
  8. I'm also curious what Anonymous is arguing in terms of life expectancy. Is it that the lower life expectancy means longer living nations are healthier and therefore have less overall health care costs? Or does it mean that for all our health care spending we still have lower life expectancies in this country? Although, I'm not entirely sure, it seems that the fact that Medicare spends 1/4 of it's budget on end of life care shouldn't be affected by age of mortality within that range. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1282187/

    We are suffering from more and more chronic health problems at earlier ages. I agree with Colin's engine analogy. Perhaps more spending on preventive measurements would be beneficial.

    ReplyDelete
  9. On a side note very soon being adopted by OSUMC is the Accountable Health Care Organization. While I know a lot about medicine, I am still learning about health care finance so for what it's worth my understanding of ACO's are this. Doctor's are given a set amount of Medicare money given the needs of the population of patients they are treating. If they go over they don't get reimbursed, if they go under they keep a portion of the profits. The idea is too push evidence based medicine and cut down on unnecessary tests. While this is similar to an HMO, it keeps patients from being locked into a health care system, and it helps reward quality of care by the physicians. Part of the problem with this is a head CT in a patient with a headache that shows a hemorrhage is good medicine, a head CT in a patient with a headache that shows nothing is bad medicine. I'm still learning about this and for anyone interested I believe there will be a big article in the upcoming New England Journal of Medicine and for Colin the Annals of EM.

    ReplyDelete
  10. You, as a scientist, know that isolating one parameter in a set of variable parameters is difficult. This is the case with life expectancy and healthcare. When one considers such other variables in the equation, such as homicide rate, vehicular accident rate, obesity rate and drug addiction, a different picture may emerge......and does.

    ReplyDelete
  11. So you claim. But I didn't graph life expectancy, I graphed costs ...

    ReplyDelete
  12. But if high costs deliver more of what people are asking from healthcare, is high cost necessarily a bad thing, or rather a choice?

    ReplyDelete
  13. If high costs are delivering better health care, then I agree it is not necessarily a bad thing, it is a choice. However, that's a big "if". If high costs are not associated with better health care in the 21 countries surveyed, then high cost in this sample is necessarily a bad thing.

    Is there evidence that the countries that spend less are actually getting less? Or are the countries that spend less simply getting more bang for their buck?

    I graphed life expectancy vs. spending -- there's no correlation. Clearly spending more does not imply covering more people -- the U.S. spends the most, yet paradoxically covers the fewest (though that may change with the new regulations). The U.S. tends to have fewer doctors, nurses, and hospital beds per capita. Not more. According to the Institute of Medicine and the Journal of the American Medical Association, as of year 2000 in-hospital medical errors caused ~100,000 preventable deaths per year in the U.S. The New England J. of Med. reported Americans receive 55% of recommended medical care, including vaccines for people at risk, routine screenings, blood testing for people with diabetes, etc. Quite remarkable.

    The only arguable evidence I can find that spending more = getting more is in relation to cancer survival. There is some correlation here, but its significance is disputed due to over-diagnosis (people can "survive" what were likely benign, asymptomatic cancers).

    So we seem to be spending quite a lot in order to cover fewer people, have fewer doctors, and kill lots of patients. What am I missing?

    As the Journal of the Am. Med. Assn. pointed out: ""Americans enjoy the best health care system in the world." So says President Bush, and many physicians agree with this claim. But frequent repetition doesn't make it true."

    ReplyDelete

Tell us what you think!