Like a Virus

by Russ Roberts on July 27, 2007

in Health

Gina Kolata writes on the front page of the New York Times:

can spread from person to person, much like a virus, researchers are
reporting today. When one person gains weight, close friends tend to
gain weight, too.

It’s much like a virus, you see. It’s not a virus. We know what a virus is. Obesity is not a virus. But it’s like a virus. It’s much like a virus. You see, the more it’s LIKE a virus, the more increasing obesity is like an epidemic rather than a failure of personal responsibility or merely a pleasant experience, say, of eating more ice cream and being a little less trim. The more it is like a virus, the less it is a personal choice, the more justified is government involvement on "public health" grounds. And in case you didn’t get the drift, check out the headline of the article:

Find Yourself Packing It On? Blame Friends

So blame your friends. Don’t blame yourself. Never blame yourself. After all, you’re standing in the path of a tsunami (scroll to the bottom.) There’s nothing you can do. But if we can just get more regulation to protect you from your friends, we can save you. We’ll start by saving your friends, first, of course. That will save you, eventually.

Here’s the summary of the study:

Their study, published in The New England Journal of Medicine,
involved a detailed analysis of a large social network of 12,067 people
who had been closely followed for 32 years, from 1971 to 2003.

investigators knew who was friends with whom as well as who was a
spouse or sibling or neighbor, and they knew how much each person
weighed at various times over three decades. That let them reconstruct
what happened over the years as individuals became obese. Did their
friends also become obese? Did family members? Or neighbors?

answer, the researchers report, was that people were most likely to
become obese when a friend became obese. That increased a person’s
chances of becoming obese by 57 percent. There was no effect when a
neighbor gained or lost weight, however, and family members had less
influence than friends.

It did not even matter if the friend was hundreds of miles away, the
influence remained. And the greatest influence of all was between close
mutual friends. There, if one became obese, the other had a 171 percent
increased chance of becoming obese, too.

You see it’s even worse than a real virus. It can spread over the phone or across the country covering hundreds of miles!

The real lesson here is that if you see your best friend gaining weight, stop being friends with your best friend. Dump your fat friends. You don’t want to catch the obesity "virus." In fact, make friends with people who are thinner than you. What a great study. All those people who judge people on their looks were right after all! It turns out that looking for thin, fashionable friends is actually good for you.

And it turns out the researchers actually have thought of this, though as you might expect, it isn’t a pleasant thought to have. The article in the Times continues:

If the new research is correct, it may say that something in the
environment seeded what some call an obesity epidemic, making a few
people gain weight. Then social networks let the obesity spread rapidly.

It may also mean that the way to avoid becoming fat is to avoid having fat friends.

is not the message they mean to convey, say the study investigators,
Dr. Christakis and his colleague, James H. Fowler, an associate
professor of political science at the University of California, San Diego.

do not want to lose a friend who becomes obese, Dr. Christakis said.
Friends are good for your overall health, he explained. So why not make
friends with a thin person, he suggested, and let the thin person’s
behavior influence you and your obese friend?

Beautiful isn’t it? If you have a fat friend, the two of you need to befriend a thin one. Or maybe two thin ones. After all, you risk exposing the new thin friend to the "virus." Obviously this is too risky. We need to quarantine fat people to protect the rest of us from the "epidemic."

At the bottom of the first page on the web version of the story, the author gives us a little more info about the magnitude of the changes we can expect from "exposure." Turns out it’s not quite as dramatic as it sounds:

On average, the investigators said, their rough calculations show that
a person who became obese gained 17 pounds and the newly obese person’s
friend gained five. But some gained less or did not gain weight at all,
while others gained much more. Those extra pounds were added onto the
natural increases in weight that occur when people get older.

What usually happened was that peoples’ weights got high enough to push
them over the boundary, a body mass index of 30, that divides
overweight and obese. (For example, a 6-foot-tall man who went from 220
pounds to 225 would go from being overweight to obese.)

And the last part of the article talks about how the study can never be replicated because it’s based on a one-in-a-lifetime data set, the Framingham Study. It turns out that the conclusions are based on the residents of a single town, Framingham, Massachusetts. I wonder if they controlled for time trends and economic factors correctly.

Be Sociable, Share!



58 comments    Share Share    Print    Email


Michael Blair July 27, 2007 at 2:19 pm

Y'know, come to think of it, I've gained about 2 lbs since I first found this site. You and Don are packing on pounds, aren't you?

Al July 27, 2007 at 3:13 pm

Obviously, a quarantine is in order. We must ship all fat people to North Korea where there are no phones.

Or food for that matter.

Too Swiftian?

Rob Dawg July 27, 2007 at 3:52 pm

As one of the very very few remaining two digit numbered subjects of the study in question I find your "conclusions" both personally insulting and scientifically specious.

LisaMarie July 27, 2007 at 3:57 pm

Obviously the solution is for the government to regulate your friends. After all, we've tried letting people choose their own friends, and it's just not working. This state of anarchy where individuals choose their friends is accomplishing nothing but making us all fat. I can hardly wait to see the government apparatus that will carry out this plan.

Sam Grove July 27, 2007 at 6:58 pm

How old are these researchers?
I bet they are products of our public school system.

blink July 27, 2007 at 8:15 pm

We learn from the article that “people were most likely to become obese when a friend became obese.” As you explain, “obese” refers to a BMI threshold. Given this and the small magnitude of the “effect” in terms of pounds, the appropriate conclusion seems to be that individuals near the threshold of obesity CHOSE friends who are also near the threshold. Everything else could be explained by a small, general increase in body weight.

I have to admit, however, that this study has surely enriched my life – reading this post was absolutely hilarious.

SKPeterson July 27, 2007 at 8:38 pm

Although the analogy isn't perfect (bacterial v. viral infection), shouldn't we be devoting resources to the identification of "Obesity Mary": the one person who through the laws of six degrees of separation is responsible for something like 90% of all the obesity in America? By quarantining this person and putting them on a strict diet, exercise and an immediate liposuction and stomach reduction surgery, the obesity problem in the US should finally begin to abate.

speedmaster July 27, 2007 at 8:57 pm

So hanging out at Curves can actually make you fatter?! ;-)

Ray G July 27, 2007 at 9:08 pm

Wasn't it not too very long ago that bulemia and anorexia were an epidemic, supermodels were to be demonized for their withering influence on our youth?

Chris O'Leary July 27, 2007 at 10:02 pm

What could be at work is a phenomenon called Social Proof. This term, coined by social psychologist Robert Cialdini, says that people are heavily influenced by the actions of others; they create norms.

It could be that's what's going on here. People are following the norms that are set by their friends.

Ray G July 27, 2007 at 10:44 pm

I don't gain weight easily, but when my wife was pregnant with our first child, I gained more weight than I ever had in my life. And subsequently lost it. It wasn't sympathy, we just happened to be eating more. . .

lowcountryjoe July 27, 2007 at 10:50 pm

I'm considering launching a study of my own using a 'representative' sample size of 535 people out of an entire population. My theory that I'm going to test is this: if nearly all of sample eat at the trough will the cummulative weight of the population's wallets and purses decrease? If I can get my study funded with public money, at least my wallet won't be decreasing in weight no matter if the theory holds or not; I'll make sure of that.

cpurick July 28, 2007 at 8:36 am

Just out of curiosity, aren't "friends" defined as the people one socializes with — say, at dinner?

Jake Young July 28, 2007 at 11:50 am

The Framingham study population is an exceptionally well characterized population that has been studied since the 40s. It was primarily started to investigate risk factors for heart disease. I am certain that they have data on the econonomic factors for these people, though it would appear that some authors are not choosing to use it.

The suggestion, however, that it is a once in a lifetime data set and therefore cannot be replicated is not accurate. There are a couple other of these long-term longitudinal studies going already on such as in Busselton, Australia.

All told though, this paper is just bollocks. I can't believe anyone published this nonsense.

Jeff Smith July 28, 2007 at 1:13 pm

The problem here is not the fact that the data come from only one place, or that the sample size is not that big or that they did not include some particular covariate. It is that they do not have any good way to identify the causal effect they seek to estimate because friends are endogenously chosen rather than exogenously assigned.

Someone might helpfully point them (and the many credulous journalists who have reported this story) to the methodological discussions on peer effects in the economics literature.

Dr. Manski, call your office!

Brad July 28, 2007 at 2:31 pm

My theory is that the authors know this is BS. What they are actually sneakily trying to demonstrate is Seth Godin's notion of an idea virus. They have succeeded. Scott Adams notes this study on his Dilbert Blog today:

LowcountryJoe July 28, 2007 at 3:32 pm

An idea virus? That must be like the many hundreds of e-mails that I receive that have either a chain-letter theme, instruct me on which days to boycott gas stations, cite a conspiracy, or tell me which products are hazadous.

Darren July 28, 2007 at 4:27 pm

Ah yes, idea viruses (memes). I'm pretty sure worship of the state is a meme. Anyone think the government might launch an effort to stop the spread of that one?

K July 28, 2007 at 7:03 pm

LisaMarie is right in saying the government should choose your friends, the old way is just not working.

Agreed. The government will make my friends diet when I want to lose weight. And make them exercise to keep me toned up.

They must also start taking music lessons and studying Spanish.

Lee Kelly July 29, 2007 at 10:55 am

I have no friends.

What are the implications for my weight?

Marianne Coleman July 29, 2007 at 2:16 pm

If you like cartoons, the NYT-Week in review section-pg 2 – 7/29/07 just for fun…!

Jerry July 30, 2007 at 9:39 am

K – you will live a full and happy life, unencumbered by the issues and weight-gain your friends (if you had any) would place upon you.

Good game.

Jerry July 30, 2007 at 9:40 am

^^^ Lee, even.

Honestly, this article is total stupidity though. I'm surprised that any reputable journal would even consider it, let alone actually publish it :|

Jon July 30, 2007 at 9:45 am

Oh oh! I'm 6' and 145lbs. Be friends with me!

This could be a great marketing scheme, can I now charge my fat friends for my friendship? "Pay to hang out with me and lose weight!"

Absolutely absurd.

Rob Dawg July 30, 2007 at 4:53 pm

The Framingham [Heart S]tudy population is an exceptionally well characterized population that has been studied since the 40s. It was primarily started to investigate risk factors for heart disease. I am certain that they have data on the econonomic factors for these people, though it would appear that some authors are not choosing to use it.

My late grandparents were #s 2 & 3. My aunts and uncles still participate. My generation (3rd) transects all continents and range in age over 40 years. These surveys have incorporated my progeny. I regularly fill out surveys and periodically and at my own expense participate at great expense in certified health screenings of detail and extant most cannot imagine. Prof. B's cavalier dismissal of the value of these efforts suggests to me that I should drop out and thus lose to humanity the value of 4 generations of longitudinal data. Prof B has additionally inspiried me to investigate a potential free market solution. Should I contact the other participants and collectively bargain for a share of the proceeds? Free market Prof. B. Hey, we could even go "Disney" and retroactively sue for past use. Good job Prof. B. Long live individual self interest. Oh wait…

J July 31, 2007 at 12:58 pm

I refer you to the lyrics of a rap song I heard a while ago that discusses how ridiculous it is to refer to an "obesity epidemic." It's by Lazyboy and the song is called "underwear goes inside the pants."

"Americans, let's face it: We've been a spoiled country for a long time.
Do you know what the number one health risk in America is?
Obesity. They say we're in the middle of an obesity epidemic.
An epidemic like it is polio. Like we'll be telling our grand kids about it one day.
The Great Obesity Epidemic of 2004.
"How'd you get through it grandpa?"
"Oh, it was horrible Johnny, there was cheesecake and pork chops everywhere."

MiltonFriedman August 2, 2007 at 2:39 am

Maybe solidarity amongst the the proletariats is not such a good thing.

lose 10 pound September 18, 2007 at 9:47 pm

So true. I never like desert, but since I got marry I tended to eat more desert, it was because my husband is a desert lover. That's how we both gained weight! It is all his false!
Now we are trying to lose weight together and desert is out of the list!

sleepingbeauty March 1, 2008 at 2:59 pm

Obesity has a lot of negative effects on an individual. For one, being obese causes a person to feel easily tired. Doing everyday body motions like walking, tying one’s shoe lace and sitting down – as well as standing up – is very challenging if not difficult.

there are ways to battle obesity. The goal of someone who is suffering from obesity and who is trying to lose weight should not be to lose weight. The main objective should be: to eat healthy, to always think healthy and be healthy.

Practicing eating foods that are good for the health is the best way to start. Having a regular time for exercise is another. It does not have to be a rigorous program in the gym. One could easily start with walking at least thirty minutes a day, everyday.

Previous post:

Next post: