Download files
Abstract
Homelessness is a globally persistent social and public health concern. Individuals who experience homelessness are not, however, a homogeneous group. Two subgroups with particularly poor individual and social outcomes are homeless people who inject drugs and rough sleepers, yet limited research has examined their needs. These subgroups, which have been characterised as ‘hard-to-house’, may be suitable for supported housing programs such as Housing First. However, there has been little evaluation on the impact of Australian adaptations of the Housing First model on individual outcomes, and whether there are differential impacts based on the type of Housing First configuration (i.e., scattered site or congregated site). This thesis addressed these identified gaps through an examination of the prevalence and correlates of homelessness among people who inject drugs; an examination of homeless individuals based on the length of their rough sleeping history; and a 12-month longitudinal evaluation of outcomes associated with scattered site and congregated site Housing First programs in Sydney, Australia.
A number of important findings emerged from this thesis. Firstly, almost one-quarter of people who inject drugs were currently homeless and self-reported poorer mental health, including schizophrenia, and a greater history of incarceration than their stably housed counterparts. Secondly, chronic rough sleepers were more likely to have a substance use disorder and experience disabling blackouts, fits, or losses of consciousness than homeless individuals who reported shorter histories of rough sleeping. Finally, the 12-month Housing First evaluation revealed that the provision of housing with attached support, regardless of configuration, resulted in improved quality of life and reduced psychological distress. Contrasting findings were identified for criminal justice system engagement, as the number of self-reported criminal justice system contacts decreased among scattered site participants but increased among congregated site participants over 12 months.
Policy and research implications are discussed in terms of service provision. Given that individuals who experience homelessness are not homogeneous, and that configurations of Housing First programs may be associated with differential impacts on individual outcomes, further evidence is needed to identify which configurations of supported housing will be most beneficial for specific homeless subgroups.