Obesity has reached epidemic proportions in America. More than 72 million American adults are
obese, according to estimates from the National Center for Health Statistics. But
obesity varies greatly by state. The map below, from the Centers from Disease Control (CDC), shows the obesity rate for the 50 states, measured as the share of people with a Body Mass Index (BMI) over 30 which the CDC classifies as "obese."
A week or so ago, I looked at the relationship between smoking and a variety of economic, social, and health factors. With a helpful analytical hand from Charlotta Mellander, we found that smoking was significantly correlated with obesity as well as being correlated with education levels, class structure, and other factors. So, we decided to take a quick look at the state-level factors that might be associated with obesity. We ran some simple correlations and scatter-plots between state obesity rates and these factors. As usual, we point out that correlation does not imply causality, but simply points to associations between variables. Still, a number of interesting things stand out.
It should come as little surprise that states with higher levels of obesity have significantly higher rates of death from cancer, heart disease, and cerebrovascular diseases like hypertension. There is a significant correlation between obesity and death rates from cancer (.7), heart disease (.7), and cerebrovascular disease (.7).
It might be, however, that states with greater percentages of obesity are those where people pay less attention to their health generally or are more likely to engage in risky behavior. And that's what we find at least in the case of smoking which correlates highly with state levels of obesity (.8).
Might obesity be related to states' broader social and psychological climates? To get at this, we looked at the relationship between obesity and a commonly used measure of subjective well-being or happiness developed by the Gallup Organization. Obesity is negatively associated with state happiness (with a correlation of -.6). Since these correlations only reflect associations between variables and not causality, it's hard to say whether this reflects the fact that happier people eat less, are healthier, or are less prone to obesity, or if unhappier people eat more, are unhealthier, or are somehow more prone to obesity, or if both obesity and happiness levels reflect something else. To get at this, we look at the associations between state obesity rates and social and demographic factors below.
Common sense would suggest that more affluent people would have lower levels of obesity and poorer ones higher, and we find such an association. Obesity is correlated with income levels (-.6) and more moderately so with economic output, measured as gross state product per capita (-.4).
One would think that states with greater concentrations of more highly educated people have lower levels of obesity, and that is what we find. States with higher levels of human capital, measured as the percentage of adults with a college degree, have lower levels of obesity (the correlation being -.8).
To what extent does obesity reflect the kind of work people do? We examine the relationships between obesity and three classes of jobs - creative/professional/knowledge jobs, blue-collar working class jobs, and standardized service class jobs like those in food processing and home health care. Obesity is strongly associated with the share of working class jobs (with a correlation of .7). Obesity is negatively correlated with the share of creative class jobs (-.6). Obesity is also negatively correlated with the share of service class jobs (-.4), though more moderately so.
Obesity is lower in states with higher concentrations of artists, musicians, and entertainers (with a correlation of-.6), those with larger concentrations of gays and lesbians (-.5), and immigrants (-.5). This likely reflects broader structural characteristics of those states, as more highly educated states also tend to be more tolerant and open to diversity.
Richard Florida
Richard,
This is a nice entry. However, I must politely note that your discussion of causality and correlation is confused. You write, "it's hard to say whether this reflects the fact that happier people eat less... or if unhappier people eat more... or if both obesity and happiness levels reflect something else". What you want to say is, "it's hard to say whether this reflects the fact that being happier makes people eat less... or if people who eat less become more happy... or if some other factor, or group of factors, cause changes in both obesity and happiness levels".
cheers,
Jeff
Yep, and we let the obese, red states dictate our health care reform. Lets pick the least capable among us to solve our toughest problems. No wonder we continue to fail. Illeterate, fat southerns run the nation while we make fun of educated liberal states with higher levels of education, lower abortion and divorce rates, and apparently, far better eating and exercize habits. Damn those liberals.
Excellent response to this article. It amazes me how one political party believes it's better to accept "dumbing down" with the likes of Sarah Palin, Joe the Plumber, Michele Bachmann, etc.
Why we gift the SEC states with so much power in national politics is beyond me. You are absolutely correct in your assessment.
Polaro, good point, I just really wish you would have spelled the work illiterate correctly...
as I spell what was meant to be "word" as "work"... noted. Perhaps you were drinking a bit of wine as I have been... ha!
You say rather off-handily that "Common sense would suggest that more affluent people would have lower levels of obesity and poorer ones higher, and we find such an association." I think that common sense would predict the opposite, after all, food isn't free. Unless one has uncommon sense, it's unlikely that you would explain poverty, obesity, criminality, and any others of the cluster of social ills as generated from a single phenomenon. It is quite logical to believe that people who make poor decisions (or where born incapable of making better decisions) are over-represented in many of these categories.
Food isn't free, but bad food (with lots of empty calories) is relatively inexpensive. And better quality food and a balanced diet do demand more of your pocketbook.
Also, finding time to exercise (a critical component in any kind of wellness/fitness) is far easier if you have disposable income and plenty of extra time on your hands. Quite a bit more difficult if one is working two low-paying jobs to support one's family and still only keeping your head above the poverty line.
So, yes, I would expect to see a correlation between health/fitness and income, with more obese and out-of-shape people appearing at the lower end of the income spectrum.
In cities the obesity rates are often considerably lower than surrounding areas, people drive less and walk more--often because a car can become a liability.
I agree, there are definitely lower obesity rates in pedestrian-oriented cities (NYC, Chicago, Boston, Philly, etc).
Where people walk more, they have exercise built-in to their daily routine (regardless of their overall diets or income levels).
@Polaro-- How rich! You criticize the illiterate while in the process misspell the word as "illeterate"! How clever!
Oo! And you wrote "Lets" when you meant to write "Let's".
Oo! And you misspelled the word "exercise," too! Surely that was on purpose? To be ironic and all?
Oo! And let's not miss an opportunity to slam overweight people, of course, and equate obesity with stupidity! Because of course, my being fat negates my 142 IQ and master's degree!
I had a laugh at the irony as well, but the fact that you've lowered yourself to critiquing his spelling/grammar would seem to indicate that you have no real counter to the actual *substance* of his comment?
An IQ of 142 is respectable, but not worth bragging about. Just imagine how smart you could have been if you'd lost a little weight...perhaps you'd have garnered a doctorate by now! ;) My recommendation would be to drop a few pounds and retest...you'll probably score higher if you increase the molality of your innate intelligence.
@Michael Baker-- Please don't misunderstand: I totally got what Polaro was saying, and for the most part agree with him. I guess what got me snarky was him being so snarky? What I mean is, I think he could have made his point just as well without being so mean-spirited ("Illeterate, fat southerns run the nation...."). And after 30 years of struggling with an eating disorder, I admit I'm sensitive to people who so casually throw "fat" in as part of an insult.
I shouldn't have lowered myself to his level.
And maybe you didn't need to be so sarcastic toward me?
Don't take my comments too harshly...I was just having a little fun.
Your post tangentially implied that IQ bore some relationship to weight. I just took it a step further and laughed at the concept of intelligence as a solute, i.e. the quotient in question was innate intelligence divided by a person's weight.
Lose some weight and presto! Stronger solution == higher IQ.
A hundred years ago in this country, it was indeed the rich who were by and large (pun intended!) the only ones who were fat. Why? Because they could afford to consume red meat.
Today, it is reversed, as we all know: the most calorically dense foods are often the cheapest, and in urban neighborhoods with no supermarket to speak of, residents find themselves surrounded by fast food outlets.
steve04074,
I differ on your definition of "urban." You define it as the low-income parts of a city. I define "urban" more broadly to include both more affluent areas, but restrict it to the city itself: e.g. if we're' talking about Washington, D.C. under this definition we're not concerned with Arlington, VA, we're only concerned with incorporated Washington D.C.
This is an important distinction because in urban areas many people have no difficulty getting to grocery stores while others struggle--income is probably somewhat of predictor for obesity in cities, while in suburbs income is useless as a predictor.
@Polaro,
Actually, NY and NJ have the highest abortion rates in the country and abortion rates generally correlate with so-called progressive voting patterns.
Poor fat Steve...yeah, it does. Typing too fast and not proofing my stuff is a problem. You, however, skipped all the content, and made fun of typos, but not the message. Good for you, self-professed, fat Steve. Keep up that (sic) master's degree intellect. Oh, did I make fun of your typo? My bad.
NY and NJ do not have the highest percentage abortion rates in the country based on ay statistics I have seen. Proof?
@Polaro-- What word is it that you're contending that I misspelled?
master's degree
n. An academic degree conferred by a college or university upon those who complete at least one year of prescribed study beyond the bachelor's degree.
The American Heritage® Dictionary of the English Language, Fourth Edition
Copyright © 2009 by Houghton Mifflin Company.
Published by Houghton Mifflin Company. All rights reserved.
@Polaro,
See the 2005 data on abortion rates reported by the Guttmacher Institute on page 10 of the report:
http://www.guttmacher.org/pubs/journals/4000608.pdf
For analysis of the data, see:
http://blog.american.com/?p=7192
HTH
Richard,
Interesting trends you've pointed out though I'm not particularly sure what you were hoping to point out. The most intriguing thing about the CDC data is that in 1989, the first year this data was collected, not a SINGLE state had an obesity rate above 15%. Yet today the "thinnest" state, Colorado is the only one still in the 15-19% category.
Thus while its interesting to point out that health problems are escalating in some places more rapidly than others, it is important to keep in mind that Obesity is a genuine epidemic that is effecting the ENTIRE COUNTRY regardless of education, income, profession, or local diversity. I don't mean to criticize your work, but the folly of research like this is that it encourages people to adopt an "us vs. them" mentality (demonstrated by posters like Polaro) when really every single person in this country needs to hear the public health message.
MN Patel, MD
Good point. You can see the change over time here:
http://edition.cnn.com/SPECIALS/2007/fit.nation/obesity.map/
Terrible color scheme on plot.Dark blue (colorado 15-19%) is adjacent to light brown (20-24%). In no way are these colors adjacent.
Totally agree. Communicating complex datasets visually is a real art form. Check out http://www.informationisbeautiful.net/ for excellent examples of this sort of thing.
Hm... New Jersey has a high abortion rate. It also has among the highest income level, HS graduation, college attendance, lowest infant mortality, etc. Maybe not be forced to have unwanted children is a good thing?
If you take a look at the data in the Nov 25 2:37 post you'll see that there is no direct relationship between high abortion rates and wealth. The locality with the highest abortion rate by far is D.C. (54.2). New Jersey's rate (34.3) is significantly higher than Pennsylvania (13.8) or Iowa (10.6), or even Massachusetts (19.9) to pick a few examples. Consider also that New Jersey was a relatively wealthy state for most of the nation's history, well before the legalization of abortion.
Different ethnic groups have different rates of obesity. Not to take that into consideration when speculating about this is ridiculous. What's the correlation of the percentage of African Americans and Mexican Americans to obesity by state?
The geography of obesity is nuanced. In major cities (not suburbs), income is often a predictor of obesity. as those who can afford to live near grocery stores will. However, many city-dwellers find alternatives to driving given how messy parking and driving gets--people move more: e.g. bike commuting, or just walking places in the course of a day. As far as the suburbs are concerned, income is not as good of a predictor, and the issues surrounding obesity there are somewhat different.
When remembering the maxim about lies, damn lies and statistics, one should also remember the First Political Law of Cause and Correlation:
The First Political Law of Correlation, Cause and Coincidence
Correlation in Support is Cause.
Correlation in Opposition is Coincidence.