Do Your Friends Make you Fat?

James - I'm really interested in all of human nature. One of the aspects that I'm interested in is social networks. My colleague Nicholas Christakis and I have looked at the Framingham Heart Study in which we've mapped the full social network of 5000 people.

Chris - When you say social network, what does that actually mean?

James - That means friendships that means any kind of social relationship that you might have with a person. Some of these are family relationships like your parent, a sibling, your child. Some of these are relationships with people at work: your co-workers and also relationships with people you're nearby: your neighbours, next door neighbours and so on.

Chris - How do you dissect apart these social networks so you can work out who's friends with whom?

James - We got lucky with Framingham. In order to keep people coming back year after year, this is a very long-standing epidemiological study, we keep people coming back year after year we have detailed administrative records: not just on their family members but they also ask them to name a close friend who'll know where you are 2-4 years from now. We got a grant to actually turn all of that into electronic data so we have all the connections between one another in the study.

Chris - How did you interrogate that data and what did you find?

James - So we matched a lot of the data about health attributes to the social network and the very first place we looked was obesity. The reason why we started there is because that's a very easy thing to measure. Every time someone comes to the doctor every 2-4 years they get measured for height and they get on the scales. It's a very objective, easily measured attribute. We were very interested to see if something unexpected could flow through these social networks. What we found was that if your friend becomes obese it increases the likelihood that you'll become obese. This was true for friends, it was true for spouses and also for siblings.

Chris - I can understand this for family as people tend to eat together and therefore they tend to eat the same things in the same amounts. How do you explain friends?

James - Well, one other interesting finding that helps to explain it is that friends who lived hundreds of miles away appeared to have just as big an impact as friends who lived next door. One possibility if you found something like this would be that what's going on is you are eating together, you're drinking together and you're exercising together. The fact that the friends who lived so far away also had an effect means that what we thing is happening is the spread of social norms. This is the idea that meeting your friend once a year at Thanksgiving or something like that and you think, 'Hey, this person's put on weight - it's ok if I put on a little bit of weight.' Maybe they start talking about some exercise programme and you think maybe I should get to the gym more often and it changes your behaviour.

Chris - Someone wrote to me once and said that she pursues a relative diet where she brings in big bags of donuts and cakes for everyone in the office because relative to them, when they eat them all, she stays slim.

詹姆斯——没错!和一些更有趣的commentaries to our work appeared in cartoons. One of them was a Cathy cartoon in which the three friends were talking very nervously about the study and saying: 'This doesn't mean we have to stop being friends with each other.' Then when the waiter comes over they ask what they'll have to eat. Each points to the other saying: 'She'll have a salad and water.' I think there was this idea that people were worried from the study that people would take away some negative points from this. That you should get rid of your friends who are overweight. In fact the data suggests the opposite, that you only spread these kinds of behaviours through close social contacts: only close social contact actually makes us healthier. Really the best thing you can do is take control of your own behaviour or alternatively try to help your friends in taking control of their behaviour.

Chris - If it is a sort of social factor and the spreading of social norms in the way in which you say, if you look at things other than obesity do you see the same pattern emerging?

James - We do see them emerging with a number of other things that we do study. We had a follow-up study on smoking behaviour. There again we saw just the same that if your friends are smoking it increases the likelihood that you'll end up smoking. One interesting thing that we're starting to see in a variety of these studies is that it's not just your friend that matters, it's your friend's friend and even your friend's friend's friend. Although it's people that we don't know and have never met we can change their behaviour and it's going to ripple through the network and have an effect on us.

Chris - Do you think you could get sued one day for being fat and letting it rub off on your friends?

James - It's funny you say that. There's a programme in the United States called Boston Legal. Shortly after our paper came out last year this was a plotline that one of the main characters who's this sleazy character played by William Shatner, you know the old Captain Kirk. He fired one of his secretaries because she was overweight. She came back and sued him for firing her. His argument was well, here are these guys from Harvard, they've told you that this person's going to make me fat. I have a right to fire her for that reason.

Chris - The outcome?

James - I don't know because I didn't watch the programme! I don't know if she was found guilty. I do know that it raises this point that seemed to be in the press a lot when people were discussing our work which was, what do you do with this information? Unfortunately a lot of people said here are new criteria for getting rid of friends. I hope no one takes any of our research to mean that. Typically what we find is that every friend makes us better off.