Factors Associated with Post-Traumatic Stress Disorder Among Homeless Youths

Background/Purpose: For the nearly 2 million youths between 18-29 years of age who are homeless, experiencing victimization before becoming homeless and being subjected to traumatic events once homeless increases the likelihood for developing trauma-related symptoms and Posttraumatic Stress Disorder (PTSD). Not only is homelessness a traumatic event, but more than half of homeless youths have experienced child abuse and/or street victimization. As rates of PTSD and trauma symptoms are elevated among this population, this study examines the association of PTSD and trauma with child abuse (childhood physical, sexual, and emotional abuse), victimization once homeless (physical and sexual victimization once homeless), psychological factors (self-efficacy and social connectedness), and homeless lifestyle factors (length of homelessness and transience). Because previous research has suggested that traumatized youths with PTSD are more likely to suffer severe negative outcomes (e.g., suicide attempts) compared with traumatized youths without PTSD, this study distinguished homeless youths with and without PTSD.

Methods: Data were collected from a multi-site, cross-sectional study of 601 homeless youths, ages 18-25 years, who were seeking homeless services in Denver, Austin, and Los Angeles. Face-to-face quantitative interviews assessed demographics, whether participants met criteria for PTSD (Mini International Neuropsychiatry Interview), child abuse (Childhood Trauma Questionnaire), and physical and sexual victimization once homeless (Traumatic Events Questionnaire). Standardized measures of self-efficacy (Client Evaluation of Self and Treatment) and social connectedness (Social Connectedness Scale) were also included. Multi-nominal logistic regression was used to assess the odds ratios with 95% confidence intervals for the dependent variable of PTSD, Trauma without PTSD, and No Trauma.

Results: Two multi-nominal logistic regression models compared youth who met criteria for PTSD, those with no trauma, and those traumatized youth without PTSD (χ²(df=22) =127.68, p < .001). The results of the first model comparing youth with PTSD to those with no trauma showed that childhood physical abuse (OR=4.88, p < .01), childhood sexual abuse (OR=1.63, p < .01), physical victimization (OR=1.63, p < .05), sexual victimization (OR=2.22, p < .01), self-efficacy (OR=.84, p < .05), and transience (OR=.70, p < .01) were significant predictors. Results of the model of youths with PTSD compared with traumatized youths without PTSD found that a history of childhood sexual abuse (OR=1.50, p < .05), sexual victimization once homeless (OR=3.07, p < .001), self-efficacy (OR=.84, p < .01), and transience (OR=.70, p < .05) were significant correlates.

Conclusion and Implications: Results are promising for the field of homeless services as they provide insight into the correlates of PTSD among these youths. For this population, understanding experiences of both childhood trauma and victimization once homeless could aid practitioners to better recognize the important issues concerning trauma symptoms and focus on effective interventions for treating these highly vulnerable youths and for preventing further victimization. Findings also suggest that feelings of self-efficacy may buffer the development of PTSD. Although this study included homeless youths from disparate regions of the country, further research is needed to replicate these findings in predicting pathways leading to PTSD among highly traumatized and vulnerable homeless youths.