Social Networks: A Hypothesized Mediator in the Association Between Incarceration and Substance Use Among Homeless Youth with Incarceration Experience

Background and Purpose:

Homeless youth with incarceration experience have higher rates of substance use than their counterparts without incarceration experience. Previous research on the social networks of homeless youth suggested that social network composition can be highly influential on their substance use. However, few studies have explored the association of incarceration and social network composition, and the indirect associations of incarceration on substance use through social networks among this population. To inform substance use reduction among homeless youth, we examined the nature of their social network composition and explored whether social network composition mediates the association between incarceration and substance use with homeless youth in Los Angeles, California.

Method:

A sample of 938 homelessness youth was recruited from three drop-in centers in two distinct neighborhoods between 2011 and 2013. Computerized self-administered surveys and social network interviews were used to collect individual and social network attributes of participants. Participants self-reported their incarceration histories and substance use (including marijuana, methamphetamine, cocaine, heroin and injecting drug use) in the past 30 days. Social network attributes included proportions of network members who use substances. Path analysis was conducted to investigate whether incarceration have indirect impacts on substance use through the social network composition of homeless youth.

Results:

Findings indicate that about 65.11 percent of participants have histories of incarceration and more than three quarters of participants used marijuana in the past 30 days, while a relatively small proportion of participants used methamphetamine (25.50 percent), cocaine (15.68 percent), heroin (10.24 percent) or injected drugs (9.23 percent). Participants with incarceration experience tend to have a greater proportion of network members who use methamphetamine (p<0.001), heroin (p=0.007) and injecting drugs (p=0.005). Moreover, participants with incarceration experience are have a greater likelihood to use marijuana (p<0.001), methamphetamine (p=0.002), cocaine (p=0.030) and injecting drugs (p=0.011). Path analysis revealed that there is an indirect effect of incarceration on methamphetamine use, such that methamphetamine use is partially mediated by the proportion of methamphetamine-using network members (p<0.001) and through the proportion of injecting network members (p=0.022). Likewise, there is an indirect effect of incarceration on heroin use, which is partially mediated by the proportion of heroin using network members (p=0.012) and through the proportion of injecting network members (p=0.016). In addition, there is an indirect effect of incarceration on injecting drug use that is partially mediated by the proportion of injecting alters (p=0.011).

Conclusions and Implications:

Homeless youth with incarceration experience have greater engagement with substance using peers, and substance-using social network mediates some of the association between incarceration and substance use. To reduce substance use behaviors among homeless youth, interventions need to take into consideration individual life experiences and the characteristics of their social environments. Specifically, the social work profession need to contribute to the development of customized approaches for this specific population that recognize their criminal justice involvement and their social network engagement.