Longitudinal impact of housing on homeless adolescents mental health and substance use

Introduction: Homeless youth have greater substance use and mental health problems than housed youth. Recent longitudinal analysis on chronically homeless youth found no causal pathway between the three factors of housing, substance use, and mental health. We hypothesize that such pathways exist, and that housing has a positive impact on mental health and substance use among runaway and “newly” homeless youth. Methods: 254 homeless adolescents were recruited using representative sampling from homeless agencies and street venues. Participants were followed every six months for three years. Demographic data was collected, along with screeners rating substance use and dependency, and the BSI was used to assess mental health status. Multi-level models full maximum likelihood estimation were used to estimate the relationship between housing status, substance use, and mental health outcomes over time. Results: In the model, housing status was included as a time-varying covariate with stable housing as the reference group while controlling for personal level variables, such as gender, ethnicity and LGB, and other time-varying covariates, such as substance use and dependence. Both the initial level (intercept) and the linear effect of time (slope) were specified as random effects (allowed to vary across individuals). Multilevel modeling allows us to make casual inferences on how housing status impacted mental health and substance use for recent runaways. Conclusion: Prior longitudinal studies have examined chronically homeless youth while this is the only longitudinal analysis of newly homeless youth. Housing did have a positive impact on mental health and substance use for newly homeless youth.