The spread of alcohol consumption behavior in a large social network.

Drinking in the Framingham Social Network in 2000This graph shows a sample the largest component of friends, spouses, and siblings at exam 7 (centered on the year 2000). There are 1,073 individuals shown. Each node represents a subject and its shape denotes gender (circles are male, squares are female). Lines between nodes indicate relationship (grey for siblings, green for friends and spouses). Node color denotes alcohol consumption behavior: red nodes indicate abstention, blue nodes indicate heavy drinking (more than one drink per day for women and more than two drinks per day for men), and yellow nodes indicate moderate intake between these two extremes. The graph suggests clustering in abstention and heavy alcohol consumption behavior, both of which are confirmed by statistical models discussed in the main text and presented in the supplement.

Relationship of Social and Geographic Distance to Heavy Drinking and Abstaining in Connected PersonsTop panels show the effect of social and geographic distance from heavy-drinking contacts on the probability that a principal is a heavy drinker in the Framingham Heart Study Social Network. Heavy drinking is defined in this case as having more than one drink per day for women and more than two drinks per day for men. The effects were derived by comparing the conditional probability of drinking in the observed network with an identical network (with topology preserved) in which the same number of heavy drinkers is randomly distributed. In the panel on the left, contact social distance refers to closest social distance (or degree of separation) between the contact and principal (e.g. direct friend = distance 1, friend's friend = distance 2, etc.). In the panel on the right, we ranked all physical distances between homes of directly connected principals and contacts (i.e., just those pairs at one degree of separation) and created six equally sized groups (1 = closest, 6 = farthest). The average distances for these six groups are: 1 = 0 miles; 2 = 0.26 miles; 3 = 1.5 miles; 4 = 3.4 miles; 5 = 9.3 miles; and 6 = 471 miles. There is no trend across physical distance. The bottom panels show the same kinds of analyses for the effect of abstaining contacts on the probability that principal abstains.

Effect of Heavy Drinking and Abstaining Contacts on Principals in the Framingham Social NetworkThis plot shows that future alcohol consumption behavior is positively associated with the fraction of friends and family who are heavy drinkers (red line) and negatively associated with the fraction of friends and family in the previous exam that are abstainers (blue line). Both lines based on bivariate LOESS regression, and dotted lines indicate 95% confidence intervals.

Contact Type and Drinking in the Framingham Social NetworkThis graph shows the change in principal alcohol consumption behavior given contact alcohol consumption behavior. The left panel focuses on heavy drinking (more than one drink per day for women and more than two drinks per day for men) and the right panel focuses on abstention. Estimates based on generalized estimating equation logit models of drinking on several different sub-samples of the Framingham Heart Study Social Network. The dependent variable in each model is principal drinking status and independent variables include lagged principal drinking status, contact drinking status, lagged contact drinking status, principal age, gender, and education, and fixed effects for each wave. Full models and equations are available in the appendix. Mean effect sizes and Bonferroni-corrected 95% confidence intervals were calculated by simulating first difference in contact contemporaneous drinking (changing from 0 to 1) using 1,000 randomly drawn sets of estimates from coefficient covariance matrix and assuming all other variables are held at their means.