HIV Prevention in Large Networks of Homeless Youth

Objectives: Public health has become very interested in how social network ties impact health behaviors. Several studies have demonstrated that persons engage in behaviors like those of their network ties. The status or position of these behavioral clusters, however, has largely been ignored. To add to the body of evidence supporting the relationship between social network structures and health, this study examined interconnections among a sample of runaway and homeless youth (RHY) and how position in the larger, aggregated network structure was associated with HIV risk. Methods: Individual and social network data were collected from a sample of 136 RHY. A sociomatrix was generated based on youth nominating other youth in the sample. Network visualization was accomplished with a “spring embedder,” and k-cores were examined. Multivariate logistic regression models assessed associations between peripheral and non-peripheral network position and recent unprotected sex. Results: Small numbers of nominations at the individual level aggregated into a large social network with a visible core, periphery (including isolates), and small clusters. Youth who had been homeless for less than two years were more likely to belong to the periphery of the network (OR=3.6, p<.01) and youth at the periphery of the network were less likely to report unprotected sex (OR = 0.45, p<.05). Conclusions: Not only was HIV risk a function of risk-taking youth being connected to other risk-taking youth, but also where youth were located in the overall structure of the network. We provide suggestions for effective social network-based prevention programs for homeless youth.