Role of Social Network Structure and Influence On Substance Use Among Homeless Youth

Background: Studies have demonstrated that drug use does not typically happen in isolation, but is initiated and reinforced through social-interactions and influence. Conceptually, network-structure and ensuing norms are especially salient in explaining drug use. Network analyses can be approached at two levels: egocentric (the direct ties of an index person with all of his or her network members), and sociometric networks (which refers to the complete set of relations between people in a population, both direct and indirect ties) (Neaigus, 1998). Both techniques have been used to understand drug use among housed youth. Work on homeless youth, however, has focused almost exclusively on dyadic or egocentric influences. Beyond the egocentric level, it is likely that peer influence also occurs at the level of the broader peer group, making it important to understand social network structure (such as network positions) and dynamics within groups (Lansford et al., 2009). Social network analysis with homeless youth has been conceptualized as a way of measuring homeless youth’s emersion into street culture (Whitbeck, Rose & Johnson, 2009)  and sociometric data helps us assess precisely how a youth is positioned vis-à-vis others in a network of other homeless youth. A lacuna of the literature on homeless youth networks is attention to how embeddedness within these social spaces shapes substance use. In this study, we assess the relative influence of network position and network norms on methamphetamine, heroin, and cocaine use among homeless youth.

Methods: Using Freeman’s Event Based Approach (EBA), 136 adolescents were recruited in 2008 at one drop-in agency serving homeless youth in Los Angeles, California. The EBA, which bounds individuals based on participation in a shared set of activities or events over time, seems the most applicable to homeless youth. The strategy uses accessing drop-in centers as an artificial “boundary” from which to recruit a population of youth. Participants provided self-reports of heroin, cocaine, and methamphetamine use. Youth nominated a full spectrum of network ties, including other youth at the agency. The structure of nominations among the 136 sampled youth were assessed with UCINET and visualized with Net-Draw. Logistic-regressions were used to assess associations among, substance use, adjacent peer substance use, and network position.

Results: Youth connected to more methamphetamine- or heroin-using peers were significantly more likely to use methamphetamine (OR=5.9) and heroin (OR=26.8). More importantly, structure was significantly associated with not only drug use, but also the choice of drugs. Youth in the core or center of the network were significantly more likely to use methamphetamine (OR=5.32), while youth who affiliated with larger networks were significantly less likely to use heroin (OR=0.34).

Implications: These results supported the general proposition that both peer and positional attributes affect substance use among homeless youth. Youth’s position in the network exposed them to norms supportive of specific illicit drugs. These results underscore the importance of tailoring interventions to reduce drug use at the network level and of recognizing drug use as not only a clinical problem but also a social problem.