Want to quit smoking? Make friends with a non-smoker. Research published in the New England Journal of Medicine last week examined the dynamics of social networks on quitting. And, no, “social network” does not mean Facebook or MySpace. It means the old-fashioned personal relationship.
The research, by Nicholas Christakis and James Fowler, follows another study they published earlier this year on how weight gain is affected by social networks. The strongest effect on quitting smoking was between spouses but was also quite strong between mutual friends, siblings, and co-workers (but only at small companies). One of the more intriguing outcomes was that someone can have an effect at three degrees of separation. That is, if you quit, social dynamics could result in someone you don’t even know quitting—someone who is a friend of a friend.
As the researchers recognize, deliberately using social networks and their dynamics could be a boon to public health. Unfortunately, the framework the researchers use when they reflect on how the dynamic works is very limiting. They and the media speculate on how friends who quit smoking shift norms so that those in the social network who continue to smoke experience a kind of moral force to quit.
Education level played a role in this dynamic. The total social network Christakis and Fowler examined had 27,000 members. Those with more education were more likely to quit smoking when a friend quit. The implication, they believe, is that better educated people are more sensitive to social norms—which strikes me as an odd explanation.
While this study is sociologically sophisticated, it is nevertheless psychologically and physiologically naïve. To quit smoking is a deliberate and often difficult act. As we know from psychologists, such those at the Center for Adaptive Behavior and Cognition at the Max Planck Institute for Human Development in Berlin, people do not make decisions on a cost-benefit analysis—even if it’s a calculation of who will be nice to them or how they can fit in better. Physiologically, the addictive properties of smoking affect people differently, with varying strengths according to each person’s unique biology and psychology. For many, overcoming the addiction requires more than a shift in social norms. They need support.
Instead of framing the social dynamics as a norms shift in a social network, it makes much better sense to frame it as people learning from each other. For me, the path that leads from shifts in norms has an ominous ring. The Christakis research took place over 37 years. As more people in the social network became non-smokers, the remaining smokers became more socially isolated—and on average smoked more. The dynamics in this social network quarantined non-conformists. Not a happy thought.
On the other hand, framing the dynamic of social networks as a learning process is naturally inclusive. Approaching this and other health problems as about learning rather than conforming also answers a question that is begged by the Christakis research: if it’s norms and social dynamics that change the smoker’s actions over time, how did it get started? If we use norms as the framework, it’s hard to say because norms by their nature are conservative and static. On the other hand, if we use learning as the framework, it simply points to people’s desire to learn, which is inherently dynamic and essentially about change.
So if you want to quit smoking, make friends with a non-smoker and learn what you can learn. In fact, if you want better health generally, make friends and learn what you can learn. Of such relationships are social networks made, best understood as mutual aid and support.
For more on this topic, see the section Social Environments in our book Too Much Medicine, Not Enough Health.