Holy citations, Batman!

by Adrianna McIntyre

There’s been some kerfuffle of late about an infographic (below, click for larger) published by the Bipartisan Policy Center. It illustrates the apparent contradiction between determinants of health (behavior, environment, genetics, access to care) and how money is spent on those things.

click for larger version

The numbers are familiar to me—I actually memorized them once, for an exam. To be clear, these data are supposed to reflect causes of premature mortality, which is a more specific metric than the label used (what makes us healthy). Things that kill us early would have been appropriate, but probably didn’t match the tone the BPC was going for.

Jokes aside, I remember being puzzled by the notion that access to care only accounted for 10% of premature deaths. Wasn’t increasing access what the Affordable Care Act was all—well, largely—about? I also wondered how they attributed deaths (fractions of deaths?) to the various categories. Let’s say someone happens to: a) be a light smoker, b) have a family history of lung cancer, c) live with a heavy smoker, and d) have no health insurance. That person dies of untreated lung cancer. Was the death caused by behavior, genetics, environment, or access? I never did follow up with my professor on the nitty-gritty behind the numbers. So when Austin Frakt started questioning the 10% figure over at The Incidental Economist on Wednesday, I paid attention.

Don’t get me wrong; I fully appreciate infographics. I like infographics. I made this one here. I tried to finagle a flowchart on constitutionality here. If you want to chastise people who dumb down information into images better suited to the Twitter-driven news cycle, I’m probably as guilty as anyone (and less talented with Photoshop than most). I admit as much.

But I also like being able to look up the info behind the infographic. I tried that yesterday, after a commenter pointed out that the BPC had cited a 2007 study. Unfortunately, that was a dead end. I’m a talented Google-sleuth, but unpublished research is beyond even my reach.

And then I remembered that I probably still had my lecture slides. I did! On the slide in question, the lecturer put “Source: Adapted from CDC, 1979; IOM, 1988; and PHS, 1993.”

After puttering around the internet, I think the first is a reference to Healthy People: The Surgeon General’s Report on Health Promotion and Disease Prevention, the second a reference to The Future of Public Health, and finally, the third to The Core Function Project (gated). The first two are some 200+ pages, which I had neither the time nor patience to sift through. Happily, the Public Health Service document is shorter and makes reference to the numbers from the infographic on page 6, citing yet another report from 1982 (Institute of Medicine). I’ve been unable to access any version of that, but I did find this 2001 IOM report.

Most notably, under Definition and Measurement of Health (all emphasis mine):

Common epidemiologic measures, which emphasize morbidity and mortality, are incomplete. Morbidity data, for example, often omit information about mortality; mortality data typically do not include information about concurrent morbidity. Disease and disability affect multiple aspects of wellness. A comprehensive definition of health requires integration of broader concepts of morbidity and mortality.

And under Statement of Task:

An Institute of Medicine Committee on Health and Behavior: Research, Practice, and Policy was convened in 1998 to update the 1982 Health and Behavior report … the committee encountered limitations as to what it was able to address. [it] became clear that there are inadequate data to evaluate fully the cost-effectiveness of behavioral and psychosocial interventions in comparison with other ways of promoting health … Comparing behavioral and psychosocial interventions with other ways of promoting health on the basis of cost-effectiveness requires additional research.

The piece goes on to discuss more boundaries and limitations of the effort; I won’t bore you with the details here.

I don’t want to detract from BPC’s ultimate message—I’m sure there is solemn truth to the trends suggested by the graphic. Heart disease, cancer, and diabetes are among the leading causes of death in the U.S.;  all three can be readily associated with behaviors, genetics, and/or environment. Obesity is something that needs to be addressed, and sooner is better.

The veracity of the specific infographic itself is neither here nor there. What’s crazy is how ubiquitous the information presented is—and how hard it was to actually sleuth out, despite its ubiquity. I found multiple mentions of the figures and chased several dead-end reference trails before remembering my lecture slides. When I finally found something substantive, I learned that the data is up to 20 30+ years old and couldn’t be replicated.

What’s a wonk to do?

[h/t to contributor Karan Chhabra for the “wonkerfuffle” neologism. It’s wonderfuffle.]

Adrianna works in clinical research and will begin graduate studies at the University of Michigan this fall.
Follow her on Twitter @onceuponA.


One thought on “Holy citations, Batman!

  1. David Hutton says:

    Nice work sleuthing. I hope those weren’t my slides with poor citations!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: