Sometimes researchers find nuggets of gold in the Framingham Heart Study that no one meant to put there.
Social scientists Dr. Nicholas Christakis of Harvard University and James Fowler of UC San Diego discovered a few such nuggets when they set out to study how people transmit health behaviors such as smoking and obesity to friends and acquaintances.
"We thought that we were going to have to do a completely new study," Fowler says. Not so, because Framingham has been collecting social networking information from its subjects since its inception. (The goal wasn't to analyze social effects, though -- it was to provide ways to track down participants for their next scheduled checkup.) Framingham had information on subjects' family members and close friends; their workplaces, which helped identify co-workers; and their addresses, which indicated who their neighbors were.
"We literally jumped for joy when we found all these records gathering dust in the basement at Framingham," Fowler says.
Because so many Framingham residents were part of the study, the researchers were able to link participants with just a few degrees of separation. The data, showing that friends can influence friends to smoke or be overweight, were published in the New England Journal of Medicine in 2007 and 2008.
The scientists also found that friends of happy people are more likely to be happy themselves. They reported their results in March in the British Medical Journal, and a similar study on loneliness will be published this summer in the Journal of Personality and Social Psychology.
The findings have implications for health interventions, Christakis says. For example, programs such as Weight Watchers and Alcoholics Anonymous harness the power of social influence to fight obesity and drinking.
"We need to be able to take the methods of Bernie Madoff and apply that to social networks," Fowler says.
"Create a Ponzi scheme of health."