Almost there...

Friday, February 20, 2009

Corrolation

SciAm posted this little article about a study that found a correlation (a really pretty strong correlation) between the likelihood of a stroke and the number of fast food restaurants in your neighborhood. Now when I started reading the article, I thought, "Wow, that's weird. I wonder why." And that's what all the scientists reading this are thinking too. And lots of folks are probably designing studies right now to figure out closer correlations, with the hope to eventually find something that points to causality. Is it that the availability leads to a higher likelihood of consumption, leading to a higher fat diet? Is it that there's something in the air pollution produced by the restaurants? Is it that folks in those sorts of neighborhoods are somehow less likely to get exercise? Is there something else these neighborhoods have in common? There's a whole boatload of "Why?" to go with a result like this.

But correlation is not causality at all. I get to the second to last paragraph where a restaurant industry trade group is slamming the study. This study is "seriously flawed," they say, the study shows "no correlation whatsoever between dining at chain restaurants and incidence of stroke." Um, okay, nor did it show a causal relationship between eating your food and having a stroke, but that doesn't make the study flawed. It just says there's a statistical relationship between the number of fast food places in the neighborhood and the incidence of stroke. Don't you want to know why? I do. And this opens up avenues for so much exploration of why, so why get your panties in a bunch over it? It could turn out to have nothing to do with fast food. Unless of course you already have other studies showing the relationship between dining at fast food restaurants and risk of stroke (or heart disease, or something else), and you don't want anyone going and poking their nose in there to look more closely. Or maybe you already have a study about the smoke coming off your grills and deep fryers? I don't know, but your reaction makes me suspicious.

Meanwhile, will we start seeing realtors producing neighborhood maps showing the incidence of fast food restaurants in the neighborhood the way we do now with little maps of schools and their test scores? That could go both ways actually, because I suspect some people love their MickyD's enough to select a neighborhood based on how easy it is to get a Happy Meal on the way home. Also ironic, their definition of what makes a place have a lot of fast food means that Berkeley, haven to granola-eating hippies and foodies, has one of the higher incidences of fast food restaurants around, though I defy you to find a Burger King or Carl's Jr. in the city limits.

5 Comments:

  • It's an interesting article. It is just a correlation, but it would be interesting to see why. I wonder what would be found if they compared # of fast food restaurants with neighborhoods of the same density but less fast food places. # of fast food places correlates with density of buildings in general (you don't have 5 Mickey Ds right next to each other in rural areas) and high concentrations tend to have greater amount of air pollution to begin with, which could lead to higher stroke incidents. Though the grease from the fat fryers getting into the air sounds like a possible theory as well.

    By Blogger Chrisfs, at 4:13 PM  

  • When I think Fast Food, I think Drive Through. When I think Drive Through, I think automobile dependent culture. When I think automobile dependent culture, I think about fewer opportunities to walk from place to place, which means less exercise, which contributes to heart disease, etc. But then how do you test for that?

    By Blogger John, at 2:55 PM  

  • Actually got around to reading the article. The definition of fest food restaurant seems to be
    two of the following,
    "but the restaurants had to have at least two of the following four features: offering a takeout service, having limited or no waiting staff, requiring customers to pay before giving them food, or having pre-prepared food."
    So even places that don't fry food at all and offer healthy food could qualify as a fast food restaurant under this definition. That points to a more general factor of urban areas. Something like pollution.

    Thanks for posting these occasionally. They are neat.

    By Blogger Chrisfs, at 2:56 PM  

  • Interesting SciAm post about the importance of play.

    http://www.sciam.com/article.cfm?id=the-serious-need-for-play

    By Blogger Chrisfs, at 3:42 PM  

  • This comment has been removed by the author.

    By Blogger Chrisfs, at 11:27 PM  

Post a Comment

<< Home