The spread of obesity in a social network.

To the Editor: In the report on the spread of obesity in a social network, by Christakis and Fow­ ler, and the accompanying editorial by Barabási (July 26 issue),1,2 variables of social and economic status (SES) are not mentioned. Social networks conceal a high degree of social homogeneity — that is, people are more likely to network with people at similar income and educational levels — and we suspect that they may operate differently in dif­ ferent contexts. For example, the way a social net­ work affects the incidence of obesity in a commu­ nity where most people have a low educational level and cannot afford a healthy diet is likely to differ from the way it affects an affluent and well­ informed community. In 2000, obesity in the Unit­ ed States was reported to be highest among black women and people who had not completed high school,3 and among black women, a low socioeco­ nomic position predicts an early onset of obesi­ ty.4 SES variables should be included in any inter­ pretive model, and public health policies for the prevention of obesity should take into account the SES characteristics of the communities they are supposed to benefit.5