Sustaining exits from long-term homelessness

30 Apr 2015

Executive Summary: Journey to Social Inclusion was a three year pilot that provided intensive support designed to break the cycle of long-term homelessness. This report presents the social and economic outcomes of the J2SI pilot. It covers the three years the trial ran, and the 12 month period following the completion of the trial. The evaluation used a randomised controlled trial to track and compare the outcomes of the J2SI participants (Group J) with those of an equivalent group of long-term homeless people (Group E) who were supported by existing services. After 48 months 67% of the original participants remained involved in the trial.

The evidence suggests that J2SI had a positive impact on the lives of most participants, over time and relative to the control group. The physical health of Group J improved, with the proportion reporting no bodily pain increasing from 27% to 45% over the four year period. In other measures of bodily pain Group J also report larger gains over time and relative to Group E. Although there is some variation in the use of health services with both groups showing greater improvements in some areas relative to the other group, the most important pattern to note is that Group J’s average use of emergency hospital services and their average number of days hospitalised in general hospitals and psychiatric units declined by about 80% over the 48 month period. Group E’s need for emergency hospital treatment increased by 21%. While the average number of days Group E spent in hospital declined by about one third, the reduction is considerably less than observed in Group J. These results represent a substantial health care impact and suggest that an intervention comprising of stable housing and intensive case management can reduce the public burden associated with the over-utilisation of health services. The report also shows improvements over time and relative to Group E in the use of welfare and homelessness services, and the amount of time incarcerated.

In the absence of the J2SI Program, Group J’s use of welfare service has started to rise. Finally, there was only a modest improvement in the extent to which the participants’ felt connected to and supported by the community over the four years. However, the trend was always in a positive direction, and in the final survey J2SI participants recorded their highest scores on both social support and social acceptance indexes.

There are signs, however that the impact of J2SI is declining in some areas. After three years 85% of J2SI participants were housed compared to 41% of those who were receiving existing services. In the 12 months following its closure the proportion of J2SI participants who were housed dropped by 10 percentage points to 75%. Although this was substantially higher than at baseline, compares favourably with international studies, and is still 17 percentage points higher than that reported by the control group (58%), it was the first ‘substantial’ decline we have observed over the four year period. Similarly, while the emotional health of the J2SI participants improved and they report lower levels of stress, anxiety and depression after four years compared to where they were at the start of the trial, the results are not much different to those reported by Group E. Similarly, although there had been substantial gains in the labour force participation rate during the trial, these gains were not sustained when J2SI closed. Throughout the trial we found little change in the substance use behaviour of the participants, although this is a common finding in studies evaluating service impacts among the long-term homeless. Finally, we found the short-term economic benefit to be modest but trending upwards, with a return of 0.25 for every dollar invested. Taking into account lives saved over a 10 year time frame the economic benefit was more substantial, with a $1.32 return for every dollar invested.

J2SI shows that breaking the cycle of long-term homelessness is possible and that intensive support coupled with stable housing can reduce demand on expensive health, justice and welfare services. However, some of the successes and improvements were not sustained and some individuals subsequently became homeless once J2SI support ended. About one quarter of the participants would have benefitted from ongoing support. The agencies they had been referred to after J2SI ended failed to provide the level and breadth of support necessary for ongoing stability. This presents two challenges for policy makers. First, many services, particularly clinical health services, are configured in such a way as to exclude the long-term homeless for the very issues they seek assistance with. Second, policy makers need to explicitly acknowledge that a small minority of homeless people require ongoing and indefinite support.

Among those who had succeeded in maintaining their housing some had made a successful journey out of homelessness and were continuing to move on with their lives. Equally, it was clear that having a home did not lead to social acceptance and social inclusion for everyone. With limited employment options, few social networks outside of the homeless population, and few alternative social activities, opportunities for social inclusion remain limited for many individuals whose experience of social exclusion is both wide and deep. In this context programs designed to permanently end long-term homelessness such as J2SI need to temper their expectations and accept that years immersed in homelessness not only have physical and emotional effects, but long-term social and economic effects as well. Further, despite sharing many similar characteristics and experiences the long-term homeless are a heterogeneous group. Recognising similarities and differences among the long-term homeless is important in terms of designing appropriate interventions – whereas some participants would benefit from ongoing support, for others the support requirements were less. Much has been learnt from the J2SI pilot – what works, what doesn’t, for whom and why. The key now is to translate these findings into a more balanced policy framework that is sensitive to the challenges and complexities of permanently ending long-term homelessness.

Publication Details
Published year only: 
Geographic Coverage