Study: Weight Gain Most Prevalent Among Fat Friends

The company you keep may make you obese, according to a much-publicized study from researchers at Harvard Medical School and the University of California San Diego.

The data show that weight gain in a close friend or relative may significantly increase a person’s likelihood of becoming obese themselves, a conclusion that suggests anti-obesity efforts that focus on group programs—such as weight loss support groups or jogging clubs—will be more effective than individual-based interventions.

Researchers believe that the study’s results help explain a dramatic increase in obesity within the past 30 years, according to study co-author James H. Fowler ’92.

The study, which was published in last week’s New England Journal of Medicine, found that a mutual friend becoming obese correlated with a 171 percent increase in the probability of obesity in the other friend.

According to co-author and Harvard Professor of Medical Sociology Nicholas A. Christakis, the data suggest a need to revise upwards efficacy estimates of public health interventions to account for reverberating effects through social networks.

“Most evaluations of public policies that involve cost effectiveness compute the cost at an individual level and then calculate the benefits,” Christakis said. “If our findings are correct, a 20-pound weight loss in you might also induce weight loss...in your social network. There may be an additional 180-pound loss, so our cost effectiveness has gone up by a factor of ten. So this would mean we’d have to rethink how we calculate cost-effectiveness.”

The data also showed that the effect of a friend’s weight gain on a subject dropped dramatically when the friendship was not mutual—subjects that had an unrequited friendship with a person who became obese ran only a 57-percent greater risk of obesity, and there was no statistically significant change in the odds of becoming obese if the only the person gaining weight viewed the subject as a friend.

The study’s authors expressed great surprise at the intense press coverage their paper had garnered in a wide range of media outlets ranging from national and international newspapers to late-night TV. Stories about the study ran in the New York Times, Wall Street Journal, Los Angeles Times, the Washington Post, and was remarked about on “The Tonight Show with Jay Leno.”

The use of the word “contagious” in many of the story’s headlines to describe obesity’s ability to spread has also been the subject of much debate. The original paper never mentions the word, and many medical experts expressed their criticism to media outlets for their misuse of the term.

Christakis and Fowler arrived at their conclusions by analyzing data from the ongoing Framingham Heart Study, which began in 1948 as an observational study of Cardiovascular Disease in subjects from Framingham, Mass. and is now studying a third generation of subjects.

Both researchers plan to pursue the notion of the spread of disease through social networks and mentioned smoking and drinking as potential future subjects.

And when asked about other studies in the works, Christakis said he was already planning to enlist some Harvard help in working on related projects.

“A bunch of Harvard graduate students are going to be putting in a lot of elbow grease on this stuff,” he chuckled.

—Staff writer Clifford M. Marks can be reached at cmarks@fas.harvard.edu.