The Spread of Obesity: A Dietary Problem or Can You Blame Your Network?
http://www.nejm.org/doi/full/10.1056/NEJMsa066082
Before going into the content of the paper, I want to note that you can expand the article found through the link above to see the full results (if you are interested) or you can just look at the more succinct results that are not expanded. I also highly recommend looking at the animations in the article to see the spread of obesity in the Framingham network over a 32 year span and also how network links are dynamic over time and not static.
The waistlines of Americans is growing in what has been deemed an obesity epidemic. The airwaves are filled not only with media stories depicting the ever growing obese population in the U.S., but also countless commercials on how best to personally lose weight. What if it isn’t only a personal choice? What if obesity could spread like a disease in a network, creating a chain reaction of growing nodes (in this case people)?
The study linked to (above) looked at a network of about 12,000 nodes in Framingham over a 32-year period and tracked the spread of obesity in the network. Using BMI (Body Mass Index) they found that having an immediate neighbor in the network (meaning they have a direct tie to the person) who was obese made the person 57% more likely to become obese. With adult siblings they found that a person was 40% more likely to become obese if another sibling did and they observed a similar probability if a spouse became obese. In addition, they found that the effect was much stronger across the same gender, so a female increases the chances another female will become obese more than another male and vice versa. Although the authors do not specify a certain causal factor that attributes to this phenomenon (they do not describe the mechanism behind this spread) the effects of the network can clearly be seen in the heightened probability of obesity when a person is connected to other obese people. The spread of obesity along social paths (the ties in this network represent social relations between two nodes) is especially poignant when you watch the animation of the network over the 32 years of the study.
This article shows the effect networks can have on individuals within it and furthermore, the effect people can have on their neighbors. They even found that a person three degrees away from another person increases the probability that that person will become obese. That is as if your friend’s friend’s friend became obese and as a result you began to gain weight (remember, the authors do not specify the causal mechanism here). The authors found that the nature of a friendship mattered. For instance, mutual friends (where each person considered the other a friend) had much higher probabilities of obesity than directed friendships (where one person thought the other was a friend, but the other did not consider them a friend).
In class we have been learning about the network effects in terms of the users of social networking sites, users of certain technologies etc… but in this article the effects are of a different kind. Instead of nodes (people) benefitting from other nodes using a product, the nodes are being directly affected by their neighbors in the network through the paths between them (in this case a social relationship). It is possible to see that the network has a certain power over how much the nodes in it weigh and the individual nodes have some say over how much the nodes connected to it (up to three degrees away) weigh. In this way, the network shapes the individual nodes and the nodes shape the entire network in a reciprocating cycle.
The problem of obesity is surrounded by advertisements for self-improvement, but maybe the problem lies in the people you have around you and the ways in which they affect you. They could change your norms concerning food consumption and exercise (e.g. having dessert every night is acceptable) or simply being surrounded by more obese people might change the way you view obesity (e.g. a little bit of chubbiness becomes normal so you are okay with being a little chubby). These are possible mechanisms behind the effect seen at a network level in Framingham, but regardless of the mechanics behind the spread of obesity, it can be seen that it is a social contagion that can spread through a network like information, a disease, or a fashion trend.
Although obesity is certainly very much guided by the individual eating and living patterns of a person, this article demonstrates the immense influence a person’s social network can have over them in terms of weight. Most people think of obesity as a personal problem, but knowing that your weight gain could cause a friend (or a friend’s friend’s friend) to gain weight might change the way in which we approach maintaining our bodies. If you believe your weight gain will not only adversely affect you, but also your spouse, your siblings, your friends and everyone connected to them, you might think twice about getting an extra helping of dessert or skipping your morning jog.
Take a look at the following study that shows the “Three-degrees-of-influence rule of social contagion” study is not statistically sound.
http://www.laboratoryequipment.com/news-Well-Known-Obesity-Study-is-Flawed-082611.aspx?xmlmenuid=51
In their original paper, Christakis and Fowler claimed to have provided evidence of a “three-degrees-of-influence rule of social contagion” within networks such as families and friend groups where obesity characteristics could be transmitted socially. Mathematics professor Russell Lyons’ research, “The Spread of Evidence-Poor Medicine via Flawed Social-Network Analysis,” published in the journal Statistics, Politics, and Policy, questions the conclusions made by Nicholas Christakis of Harvard and James Fowler of the Univ. of California, San Diego. When Lyons reviewed the evidence, he found not only a lack of statistical significance in the findings, but also that both the researchers and the reviewers did not realize that the statistical procedures Christakis and Fowler had used were inapplicable. Their methods, Lyons found, were fundamentally flawed.
Great Post. lnformative and thought provoking. You have touched on some very useful points.Thank you so much for sharing such a great content.
Lose Weight
The best way to weight loss is to watch what you eat, when and you eat. In order to lose wight, healthy eating and exercise are very important. To keep your weight off you’ve got to eat healthly, eat right and exercise regularly.
Lose weight