Method: The 100 youth in the sample were from the socio-metric network data derived from the Homeless Youth Network Survey in 2008 at a drop-in center in Los Angeles. Network data was collected in a face-to-face interview conducted by a trained interviewer, soliciting network ties using a 17 category name generator (e.g. “friends”, “family”, “people you kick it with”). A sociomatrix was created linking participants in the sample. A directed tie from participant i to participant j was recorded if participant i nominated participant j in his/her personal network. Matches were based on: name, alias, ethnicity, gender, approximate age, and agency attendance. This technique allows us to assess associations of actual behaviors among social network peers rather than relying on perceptions of peer behaviors as reported by other youth. Concurrency was assessed by number of sexual partners in the past 90 days. Participant demographics, risk behaviors (e.g., unprotected sex and methamphetamine use), and social network properties (e.g., peer involvement in concurrency, peer's practice of unprotected sex and peer's methamphetamine use) were included in the model. Bivariate and multivariate linear regressions were conducted. Due to the extensive literature on homeless youth's sexual risk, we used one-tailed tests rather than two-tailed tests when assessing the impact of network characteristics on concurrency.
Results: The prevalence rate of concurrency among youths in this study was 37 percent. For each participant, more than one third of their social network peers had used methamphetamine, one out of two did not use condom at their last sexual intercourse, and over one third of them had been involved in concurrency. At the bivariate level, being a minority, personal methamphetamine use, social network peers' involvement in concurrency and methamphetamine use were found to be significantly associated with concurrency (p<.05, one-tailed tests). In the multivariate model, personal methamphetamine use was still found strong predictor to participants' involvement in concurrency, while controlling other variables (P<.05, two-tailed testes).
Discussion: The study has important implications for designing interventions for homeless youth. Compared to the general population, more homeless youth are involved in concurrency. Personal methamphetamine use was identified in this study as a significant association with concurrent sexual partnerships. To our knowledge, this is the only study that looks at methamphetamine use among homeless youth and its impact on their concurrency. This finding suggests targeting on methamphetamine-using homeless youth can be an important strategy for reducing concurrency.
