Life-course predictors of homelessness from adolescence into adulthood: A population-based cohort study

Introduction: Internationally, the prevalence of young adult homelessness is concerning. Few data on life-course predictors from longitudinal studies exist, limiting our capacity to inform prevention strategies at the population-level. Methods: Data were drawn from a state representative population-based sample of young adults from Victoria, Australia participating in the International Youth Development Study (IYDS; N = 927, 54% female). Participants were recruited in state-representative secondary school samples at Grade 7 (age 13, 2002), with follow-up in Grades 9 (age 15) and 11 (age 17) and at ages 21, 23 and 25. Using longitudinal path modelling, we conducted a series of analyses testing life-course predictors of young adult homelessness across multiple socializing contexts, and the interrelationships among them. Results: The rate of young adult homelessness was 5.5%. Path modelling showed higher levels of family conflict at ages 13 and 15 uniquely predicted homelessness by age 25. This effect remained after accounting for other risk factors in peer-group (e.g., interactions with antisocial peers), school (e.g., low academic performance), and community contexts (e.g., low neighborhood attachment). Peer drug use and interaction with antisocial peers at age 15 mediated the association between family conflict at age 13 and homelessness by age 25. Conclusions: Findings point to the vulnerability of early adolescents to family conflict. This vulnerability heightens risk for young adult homelessness. Findings strengthen the case for both primary prevention programs that build healthy relationships between family members from early on in adolescence and for investment in homelessness prevention at key developmental periods.