Want to lose weight? Drink (Good) Beer

I just could not resist writing about this when I stumbled upon the interesting factoid about beer and obesity:  States with the highest number of craft beers per 100,000 people also have lower obesity rates (for those of you statistically inclined, the correlation rate is -0.54).  This observation comes from The Atlantic Cities website, which did a report on the increase of craft beer breweries in the United States (there are more craft breweries now than any time since 1887).  What is even more interesting was the quick set of correlations they performed, including the one I just mentioned above.

Let’s work through this relationship. Below are the maps showing US Obesity rates in 2010 (from my previous post) and the map put together by The Atlantic Cities showing the number of craft breweries per 100,000 people, for each state.

First thing to note here, is that many of the the red states on the left (indicating states with adult obesity rates greater than 30%) overlap with the light purple states on the right (indicating that there is, on average, less than 1 craft brewery in the state for every 100,000 people).  I said mostly overlap, as there a few red states (like Michigan) that are a medium-purple too, which tells us that some heavily obese states  average between 2-3 craft breweries per 100,000 people.  Second, notice that dark purple states on the right (indicating, on average 3 or more craft breweries per 100,000 people) are also yellow states on the left (indicating adult obesity rates between 20-25%): Vermont, Montana, and Oregon So now I can add drinking my loveable “Old Chub” Scottish Ale as a good substitute for running home from work once a week!

Of course, I’m having fun – because obviously, drinking beer (even yummy craft brews) does not make you thin.  As we all know (and the The Atlantic Cities website is quick to point out) correlation only shows how well two things are related – not that one causes the other.   The question now is, are there any obvious characteristics that make Vermont, Montana, and Oregon polar opposites of places like Texas, Louisiana, Kentucky, and West Virginia, and could possibly be related to both obesity and craft breweries? 

I had two, and neither seem all that probable:  The first is mostly related to Montana, Oregon, and Vermont.  All three are what we could call ‘outdoorsy’ states, and are known for adventure tourism (like skiing, hiking, camping, etc).  So that might explain the low obesity rates – high exercise and activity. How does this relate to craft breweries?  My best guess is, with vacationing and adventure goes drinking and eating good food, and craft breweries would fair well in these states.  As far as I know, few people plan adventure-related vacations to places like Mississippi, West Virginia, or Texas (does anyone really plan a vacation to Texas?).

The second is better education, particularly in agriculture and sciences.  There is already a fairly decent correlation with overall education and decreases in obesity, and The Atlantic Cities website also showed a weak, positive correlation with craft brewery and adults with college degrees (0.32).  Ask any brewer, and he or she will tell you that making beer (or wine for that matter) is highly scientific, and requires lots of chemistry and math.  You need a population that is willing and able to use science everyday as part of its job.  The central US states do not fair well in reports on science and engineering education studies (Mississippi ranked last in all the states on the Science and Engineering Readiness Index of high school students).  So perhaps the underlying cause between craft brewery and obesity is better educational attainment.

As I said, neither are that great at explaining the correlation; if anyone has better ideas, fire away!

Finally, here are two more interesting correlations with craft breweries: Craft brewing is less likely in conservative states, with a modest negative correlation (-0.30) to 2008 John McCain votes (there was no statistically-significant association to Barack Obama votes); And craft brewing is more closely associated with higher levels of happiness and well-being (0.47).

Hurray Beer!


The New Defintion of a Red State

Let’s go back to 1985.  Not because of the awesome music (although, has anyone else noticed that Lady Gaga sounds so much like Madonna?  But I digress).  We’re going back to 1985 for that cutting-edge technology.

Computerized phone banks.

You see, in the days when The Cosby Show and Miami Vice ruled the airwaves, the high techno geeks were in the laboratory playing with the genetics of computer databases and fiber optics.  And what was born was the ability to have a computer automatically dial random numbers from a database.  Never again would a family make it through Sunday night dinner without a telemarketer calling to sell them long-distance packages.  But something really useful did come from this – the CDC’s Behavioral Risk Factor Surveillance System, or the BRFSS.

Despite the James Bond sounding name, the BRFSS was simply a system to call random phone numbers in the US, and then ask adults to answer question about their behavior and health.  Created by the CDC, the BRFSS collected data not just on how healthy we were (“Have you been diagnosed with diabetes?) but also about your behaviors (“How often do you wear your seatbelt?”).  The idea was, in order to be a healthy person you actually had to behave like a healthy person.  All the wishing in the world would not keep you from getting heart disease: Eating a healthy diet and not smoking would, though.

The BRFSS also collected gender, height, weight, age, and of course, where you lived (according to area code, which in the early 1980s was tied to your actual location, not the address of where you bought your cell phone).  And with this information epidemiologists at the CDC started tracking BMI, or Body Mass Index I know what you are going say – everyone hates the BMI, and that it is a poor tool for determining if someone is too fat.  But the BMI does do something really well:  It gives the relationship between height and weight as a single number.  And given that the BRFSS surveys all 50 states, and US territories, it is powerful.  More than 150,000 adults are surveyed each month, and from it we can track BMI in the United States, going as far back as 1985.

So, how many people in the United States had a BMI of greater than 30 (which qualifies as obese) in 1985?

Only a handful of states did the BRFSS in 1985, but from the data we have, all looks pretty good.  Most states had obesity rates of less than 10%, and only eight had rates between 10-15%.  Awesome, as they would have said back then.

So what happened between then and now?

Times were still lean, so to speak, in 1990.  By then almost all the states are collecting data, and obesity rates in all states are less than 15%.  Lots of pretty, light blue.

But something really dramatic happened by 2000.  Now we have all 50 states collecting data, and there is a huge pocket in the central and southern US where 1 out of every 4 US adult is obese.   Line up all your friends, start counting them, and it would go like this:  “One.  Two.  Three.  OBESE.  Five.  Six.  Seven.  OBESE.”  Way to go Ohio, Texas, Indiana, Illinois, West Virginia, Tennessee, Georgia…….  And aside from Colorado (the only light blue state left), the rest of the states have obesity rates of 15 – 20%.

And today (well, 2010 – the most recent data set we have)?  Hides your eyes, it is not pretty.

According to the CDC, over 30% of US adults are obese.  That is 1 in 3, people.  “One. Two. Obese.  Four.  Five.  Obese.  Seven.  Eight.  Obese.”  And notice there is no more blue.  The last ‘blue’ state was Colorado, and up until 2009, it was the only US state or territory that had an obesity rate of less than 20%.  Stop your bragging, Colorado.  You’re one of us now.

So what is the take home here?  Remember, the BMI has both height and weight in it.  So we could argue that only height is changing, and that is why the BMI is changing over time.  But since the BMI is your weight divided by your height-squared, that means the average height of all US adult has to shrink.  And we know that is not happening.

In my obesity class, I tell my students that BMI is not a perfect tool for determining if a single individual (you, or me, or your best friend) is too fat.  But it is the perfect tool for tracking what is happening, on average, in a population – like the United States.  And quite frankly, the United States is fat.  And getting fatter.