• How much does health care contribute to health?

    As Ezra Klein noted, The Bipartisan Policy Center included this infographic in their report on obesity and its economic consequences (PDF).

    Do you buy the numbers in this infographic? I don’t see any reason why we should. First of all, the report tells us that the source for the graphic is “Derived from information from the Boston Foundation (June 2007).” But there is no additional information saying what this means. Where did the Boston Foundation get the numbers in June of 2007? We can’t tell. We can’t even tell where to look.

    What I do know is that the figures in the “What Makes Us Healthy” side of the chart are pretty close to those offered by McGinnis et al.:

    Using the best available estimates, the impacts of various domains on early deaths in the U.S. distribute roughly as follows: genetic predispositions, about 30%; social circumstances, 15%; environmental exposures, 5%; behavioral patterns, 40%; and shortfalls in medical care, 10%.

    But these figures are about what kills us, not what makes us healthy. Is that what the infographic is telling us too? The near correspondence of the numbers to those in McGinnis et al. makes me wonder. In particular, notice that both the infographic and McGinnis et al. say that medical care plays only a 10% role in keeping us healthy (or killing us). About this, Booske et al. (PDF) write,

    The “long standing estimate” of 10% for medical care is actually based on “expert” estimates of the contribution of health care system deficiencies to total mortality (DHHS, 1980).

    Meanwhile, David Cutler and others suggest that health care is responsible for 50% of the gains in longevity over the past half century. Now, I have not read the underlying literature on this. I have not read DHHS (1980), nor can I find it. I will also confess that I have not read the full BPC report, which may be fantastic, notwithstanding my concerns about the infographic.

    Still, it seems to me there is no good reason to accept the 10% figure at the top of the left-hand side of the infographic. I’d like to know more how it got there. I’d like to hear the best argument as to why it’s correct.


    • I agree. It is difficult to even draw the line between what is a contribution of the health care system and what is not. Do you include vaccinations and if you do you consider if no one got vaccinations what would happen. If you did the number would be much higher than 10%.
      On the other hand it seems to me that the most affective health care is the cheap stuff like vaccinations.

    • What does it mean to say that “we spend 4%” on “healthy behaviors”? 4% of what? of “what we spend on being healthy”? What number is that? “Healthy behaviors” appears to include food (see the picture of the apple). Does “what we spend on behing healthy” include all food expenditures? Only expenditures on fruits? And what is the bicycle supposed to represent? Do expenditures on healthy behavior include sales of new bicycles? Pool memberships? What?

      The entire concept of the graphic seems incoherent to me. There’s a propoganda point being made, which may or may not correspond to something that’s more or less true = that we waste a lot of money in the health care system – but this graphic doesn’t demonstrate it.

    • “Where did the Boston Foundation get the numbers in June of 2007? We can’t tell. We can’t even tell where to look.”

      Perhaps Dr.Frakt should try using Google. The first link listed when I googled ‘Boston Foundation (June 2007)’ was to the executive summary of the report in question (found at http://www.tbf.org/uploadedFiles/BostonParadoxReport.pdf).

      The summary on page 8 says “The full report, with a detailed description of 30 indicators of health, health care and competitiveness is available from the Boston Foundation or New England Healthcare Institute, or through their respective websites, http://www.tbf.org and http://www.nehi.net.”

      • Seriously, knock it if the graphic’s not sourced, but it looks to be sourced to me… did he read the works cited? Certainly doesn’t sound like it.

      • Larry,

        Even if you go to the trouble of finding the full 208 page report (I believe it’s here, the fifth report with “BigReport” in the document title), the only time those figures are mentioned is on page 30.

        The graph used doesn’t seem to be cited in the endnotes, though it’s vaguely sourced from the Institute for the Future, UCSF/CDC. Searching IFTF’s site for “determinants of health” or “mismatch in spending” yields no useful results. So, the questions posed above are still wholly unanswered.

        I’d be quite curious if anyone were able to unearth the methodology. I remember seeing the IFTF graph in lecture a few years ago and always wondered what assumptions underlied the data.

        • Er, apologies. It appears I pasted the wrong URL when trying to link to the report (I just tried going back to read it). What I intended to link is below.


          or, directly to the PDF:

          • Adrianna, my apologies. I tried to find the underlying document on both the Institute for the Future an CDC websites and came up blank. It appears to be cited as “unpublished research,” and not available at either site. I would hope the authors at the Bipartisan Policy Center would be willing to provide it.

    • I wonder how they divide up public health from private health (i.e. medicine).

    • The “what we spend on being healthy” side makes little sense. Some portion of our spending on food, clothing, shelter, etc., all contribute to making us healthy, as does public spending on social security, unemployment insurance, supplemental nutrition, parks, social services, education, public transportation systems, environmental protection, occupational health and safety, clean water, etc,

      That said, the general point is correct that there is an imbalance in both public and private spending on health care. Not enough attention is paid to environmental factors and the social determinants of health. Programs focused on wellness and prevention are often held to a much higher bar than other interventions. No one asks if doing an MRI will save money, but that is frequently asked of behavioral interventions, wellness programs, public health programs and preventive services that may have a much larger bang for the buck in terms of health outcomes . The incentives go all the wrong way. This is worth attention–even if the graphics are nonsensical.

    • Thank you for pointing out how this new obsession with infographics can distort the facts.

    • I have a degree in public health, so I’m sympathetic to the claim that improvements to medical care aren’t the sole cause, and may even be a minority cause, of improvements in health.

      So, I bought the chart at first. And then I read this article, and yes, the chart’s assertions don’t seem on target.

      So, for me, it’s still about changing behavior, and then social/environmental factors, and then medical care.

      Ironically, the discipline of public health I work in is delivery of medical care.