Neighborhood disorder, peer network health, and substance use among young urban adolescents

The current study investigated the moderating effect of peer networks on neighborhood disorder's association with substance use in a sample of primarily African American urban adolescents.

A convenience sample of 248 adolescents was recruited from urban health care settings and followed for two years, assessing psychological, social, and geographic risk and protective characteristics. A subset of 106 substance using participants were used for the analyses. A moderation model was tested to determine if the influence of neighborhood disorder (percent vacant housing, assault index, percent single parent headed households, percent home owner occupied, percent below poverty line) on substance use was moderated by peer network health (sum of peer risk and protective behaviors).

Evidence for hypothesized peer network moderation was supported. A latent growth model found that peer network health is most strongly associated with lower baseline substance use for young adolescents residing in more disordered neighborhoods. Over the course of two years (ages approximately 14-16) this protective effect declines, and the decline is stronger for more disordered neighborhoods.

Understanding the longitudinal moderating effects of peer networks within high-risk urban settings is important to the development and testing of contextually sensitive peer-based interventions.

suggest that targeting the potential protective qualities of peer networks may be a promising approach for interventions seeking to reduce substance use, particularly among younger urban adolescents living in high-risk neighborhoods.