Better health and ageing for all Australians

Homelessness and Mental Health Linkages: Review of National and International Literature

1.7 Evaluation of intersectoral interventions for people who are homeless and living with a mental illness

Up to Publications

| TOC |next page

People who have a mental illness and are homeless are not a homogeneous group – they are likely to have a variety of complex needs that require a range of services. These services may include housing, substance abuse treatment, health care and income support or job training. This level of complexity requires a multifaceted response, and no single agency can meet all these needs at present. Most services have evolved to cater to a particular need such as housing, or drug and alcohol rehabilitation, or mental health, or primary health care. Each sector has evolved independently of the other, with its own funding source, service eligibility requirements, geographical boundaries, treatment or service philosophies and administration policies. In order to gain access to each of these systems, the homeless person is required to negotiate various forms, eligibility requirements, transport arrangements and waiting lists (Randolph, Blaskinsky, Leginski, Parker & Goldman, 1997). While there are many descriptions of 'best practice', there are few well-designed evaluations of programs that use intersectoral approaches in responding to the needs of people who are homeless and living with a mental illness.

In order to address a lack of integration and inter-agency cooperation, a number of ACCESS demonstration projects have been funded in the USA with the aim of preventing homelessness among the mentally ill (Randolph, Blasinsky, Morrissey, et al., 2002). These demonstration projects have developed various models of inter-agency cooperation and different housing programs to meet the needs of the homeless and mentally ill and are among the first longitudinal, experimentally designed studies of housing and service interventions for this population. These programs concluded that homeless adults who have severe mental illnesses, often thought to be beyond the reach of existing outreach services, are willing to accept psychiatric treatment and can remain in community-based housing with appropriate support.

Key findings included:

  • Homeless people with severe mental illnesses will use accessible, relevant community psychiatric treatment services.
  • Residential stability is an attainable goal for most people with severe mental illness.
  • Formerly homeless people with severe mental illnesses are an important resource.
  • Substance abuse is a major factor in homelessness among people with severe mental illnesses.
  • Housing stability, appropriate psychiatric treatment, and increased income lead to an improved quality of life.
  • Consumer needs and self-reports about preferences should be considered when making decisions about housing.
The policy implications identified by these ACCESS demonstration projects included:
  1. Service systems must be integrated at all levels to remove barriers and promote efficient use of services: psychiatric services with housing, social services, substance abuse treatment and the criminal justice system.
  2. Substance abuse treatment must be an integral part of comprehensive psychiatric services for people with severe mental illnesses.
  3. A range of housing options is required. Independent living with the availability of support services is both possible and preferred by most people with mental illnesses. Because this does not meet the needs or preferences of everyone, however, other choices may have to be considered.
  4. Preventive health care and education are critical, especially relating to the risks of HIV/AIDS, tuberculosis and smoking.
  5. Longer-term follow-up studies should focus on how to sustain early gains. (Herrman, 1999).
A follow-up study of these projects examining inter-agency collaboration was conducted by Morrissey and colleagues (Morrissey, Calloway, Johnson & Ullman, 1997). One of the main findings of agencies receiving ACCESS funding was that they were poorly integrated and fragmented, resulting in a perception that they were inaccessible to people who were homeless and suffered severe mental illness. Inter-agency links most often took the form of information sharing or client referral. Only one per cent of agencies were involved in joint funding arrangements. Most agencies had their own funding source and were not dependent on each other. As a consequence, agencies acted more autonomously in their inter-agency relationships and pursued their own interests rather than collective goals.

The ACCESS agencies were better connected to the services in their immediate locality, compared to other agencies in the collaborative network. As this is a longitudinal study, an evaluation of the number and type of linkages between agencies can be made. It may be that close links between a small group of service providers may serve client needs better than efforts to link up an entire service network. Alternatively, integrated services within one agency may outperform those involved in multiple service links.

A recent evaluation of the ACCESS program found that all the demonstration sites delivered improvements in terms of client outcomes. Services that formed part of the experimental sites, that is, had contacts with other services and exchanged client information, did deliver better housing outcomes for clients than those that were not integrated. However system integration made no difference to clinical outcomes, service use or perceived quality of life (Rosenheck, Lam, Morrissey, et al, 2002; Goldman, Morrissey, Rosenheck et al., 2002). Another ACCESS demonstration project found that service providers could effectively link people who have a mental illness and are homeless into community mental health services. A three-year longitudinal study found that community mental health services could engage hard-to-reach homeless people and that this reduced the number of days hospitalised. The effect continued after the intervention was terminated (Rothbard, Min, Kuno, & Wong, 2004).

A collaboration between the Centre for Mental Health Services and the Centre for Substance Abuse Treatment in the United States was initiated in response to the high proportion of people who have mental illness and drug abuse and are homeless. The two organisations came together in a jointly funded and administered agreement to document and evaluate the effectiveness of their homeless intervention strategies. To facilitate this process a cross evaluation steering committee was formed with management from both centres present. Responsibilities included development of common data measures, design of cross-site analysis, and policies on data sharing (Rosenheck, Resnick, & Morissey, 2003). Recent evaluations have demonstrated that clients that were independently housed showed a greater decrease in their alcohol consumption and greater decrease in their drug use compared with clients in unstable housing (Mares & Rosenheck, 2004).

| TOC |next page