Degree Name

Doctor of Philosophy


School of Psychology


Over the past two decades there has been substantial gain in knowledge regarding the biopsychosocial correlates of complex trauma experiences (e.g. Briere & Spinazzola, 2009; Cook et al., 2005; Ford & Courtois, 2009). There is a pressing need to integrate, translate, and apply this knowledge across at-risk and marginalised groups. The objective of this doctoral thesis was to provide new and original insights into the nature and impact of complex trauma within the context of ecological vulnerability; undertaking a person-centred examination of the homelessness vulnerable in Australia. A thesis by compilation format was adopted, comprising of a theoretical chapter (Chapter 2) and four interconnected empirical examinations (Chapters 3 – 6). Each of the four empirical chapters utilised data collected in the Journeys Home (JH) Study – a six-wave longitudinal examination of a nationally representative sample (N = 1682) of Australian individuals identified as homeless or at risk of becoming homeless (Wooden et al., 2012). Chapter 2 involved development of a novel resources-perspective framework, informed by a leading psychological traumatic stress theory, Conservation of Resources theory (Hobfoll, 1988; 1989; 1991). A key proposition of this framework was use of the concept of ‘risk factor caravans’ (Layne, Beck et al., 2009; Layne, Olsen et al., 2010; Layne, Briggs, & Courtois, 2014) to reflect the manner in which complex traumatic stress sequelae influence person-in-environment processes. Practical application of this novel theoretical framework was then demonstrated in the series of empirical examinations. Chapter 3 (Empirical 1) examined personal resource states and their associations with childhood complex trauma exposure and homelessness. Latent class analysis (LCA) revealed four distinct personal resource states: Disconnected (n = 129), Engaged (n = 544), Surviving (n = 131), and Thriving (n = 858). Compared to the Thriving reference class, cumulative trauma exposure was significantly predictive of likely classification into the personal resource states of Disconnected (OR = 1.27, 95% CI [1.17, 1.37]), Engaged (OR = 1.28, [1.22, 1.34]), or Surviving (OR = 1.39, [1.29, 1.50]). Chapter 4 (Empirical 2) examined associations between childhood trauma and risky alcohol consumption, and the implications for housing stability. Logistic regression modelling was used to examine associations between alcohol consumption and childhood abuse, socio-demographic factors, and changes in homelessness status. The results demonstrated that self-reported recall of childhood experiences of violence was significantly associated with Low Risk (OR = 1.54, 95% CI [1.01, 2.35]) and Risky (OR = 2.35, [1.41, 3.94]) drinking. Recall of childhood neglect was associated with a lower likelihood of Risky drinking (OR = 0.57, [0.34, 0.96]). Additionally, Risky drinkers were significantly more likely to remain homeless at follow-up (OR = 1.85, [1.16, 2.96]). Chapter 5 (Empirical 3) sought to model the ‘risk factor caravan’ concept and examined the nature of childhood complex trauma experiences using LCA. Six distinct childhood trauma history classes emerged Multiple (n = 262), Distal (n = 286), Proximal (n = 209), High Violence (n = 233), Indirect (n = 250) and Low (n = 423). Significant covariate differences between classes included: gender, biological relationship of primary carer at age 14 years, and time in foster care. Chapter 6 (Empirical 4) examined longitudinal patterns of psychological distress and associations with childhood complex trauma exposure. Growth mixture modelling revealed four distinct trajectories of psychological distress: Chronic (n = 335), Escalating (n = 93), Attenuating (n = 109), and Resistant (n = 957). Experiences of different types of trauma during childhood were associated with these psychological distress trajectories. In particular, adults experiencing chronic psychological distress were significantly more likely than those exhibiting distress resistance to have experienced multiple and varied childhood maltreatment, adjusted odds ratio (AOR) = 5.75, p = .002, 95% CI [2.37, 9.19]. Collectively, this body of work provides significant and original contributions to the traumatic stress domain. The proposed theoretical framework presents opportunity for translation of a resources perspective research approach to other vulnerable populations affected by complex trauma. Demonstration of the framework application using sophisticated statistical analysis techniques, gives new insights into the dynamic processes underpinning the relationship between complex trauma and homelessness. The value of person-centred longitudinal methodological approaches are highlighted. Evidence of psychological resilience in a large proportion of this sample population stands as a salient reminder of human capacities to survive and even thrive in the face of trauma and adversity. New insights present important clinical implications for potential avenues to improve access to and facilitation of social support services for individuals exposed to substantial ecological vulnerability and homelessness risk.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.