Toward Understanding Homelessness: The 2007 National Symposium on Homelessness Research. Changing Homeless and Mainstream Service Systems: Essential Approaches to Ending Homelessness

By:
Martha R. Burt, PhD, The Urban Institute, Washington, DC
Brooke E. Spellman, MA, Abt Associates Inc., Peoria, IL

Abstract

Martha Burt and Brooke Spellman focus on how federal policy and state and local action have stimulated the development of homeless assistance networks and how those networks are evolving to address ending homelessness. While little formal research has been done on this subject, the authors present frameworks for assessing system change as well as describe promising practices from the field. They describe factors that may influence the success of change efforts, including the local and state context, the interest and commitment of stakeholders, the scope of desired change, the governance and management structure for change, and the intended process for change. They also review mechanisms that help make change happen by reorienting local Continuums of Care, matching clients and services, retooling funding approaches, and using data to track implementation and outcomes.

Introduction

In 1998, when the U.S. Departments of Health and Human Services (HHS) and Housing and Urban Development (HUD) sponsored the first National Symposium on Homelessness Research, the focus was primarily on describing the array of approaches to helping homeless people that had been developed during the previous decade, and how they worked. Few in the field had begun to address how a community might end homelessness.

Much has changed since that time. New programmatic approaches have evolved (e.g., Safe Havens), but more important, federal policy has stimulated the development of homeless assistance networks and systems. In 1995, after seven years of distributing McKinney Act Supportive Housing Program (SHP) funds through annual national competitions, HUD implemented the competitive continuum-of-care (CoC) approach for deciding who receives SHP support for transitional and permanent supportive housing. A CoC is a local or regional system for helping people who are homeless or at imminent risk of homelessness by providing housing and services appropriate to the whole range of homeless needs in the community  from homelessness prevention to emergency shelter to permanent housing. Each year, HUD develops and publishes preliminary estimates of how much SHP funding each eligible community in the country would receive if it wrote a qualifying CoC grant application. To qualify, communities have to show that they have assessed needs and existing resources and identified gaps, and that the resources they are requesting from HUD will help fill the gaps that the community has identified as top priority. The entire process stimulated a new kind of communication among relevant programs and agencies, often leading to increased cooperation and program innovations and moving many communities closer to having a real system rather than a set of independently operating programs (Burt et al., 2002).

The processes of community organizing developed through the CoC process received a substantial boost beginning in 2000, when the National Alliance to End Homelessness (NAEH) developed and disseminated a plan to end homelessness for the whole nation in 10 years (NAEH, 2000). This plan incorporated a major shift in orientation and emphasis, from managing homelessness to ending it. This shift has been significant enough to be dubbed a paradigm shift in the nations approach to homelessness (Burt et al., 2004), because it entails important new ways of thinking about homelessness and concomitant changes in who should be served, what approaches should be used, and how resources should be allocated.

This shift, and the expectation that it could succeed, was made on the basis of research evidence. Three pieces of information from research helped advocates make a convincing case that shifting the goal from managing homelessness to ending it was both the right thing to do and something that was possible to achieve:

1. A finite group of homeless people on which to focus was identified. Research by Kuhn and Culhane (1998) documented subgroups among homeless people characterized by transitional, episodic, and chronic patterns of homelessness. These researchers also documented the fact that the 10 to 15 percent of people with a chronic pattern of homelessness consumed half or more of system resources  in this case, shelter bed-nights  making them a very expensive group to continue serving in emergency shelter while not resolving their homelessness. A reliable estimate of homelessness nationwide based on the 1996 National Survey of Homeless Assistance Providers and Clients (Burt, Aron, & Lee, 2001) then made it possible to estimate the number of chronically homeless people  150,000 to 250,000  a number that proved to be small enough for policymakers to believe that a solution was possible.
2. An effective service approach was identified. Evidence accumulated that permanent supportive housing (PSH) worked to keep many formerly homeless people housed (Shern et al., 1997; Tsemberis & Eisenberg, 2000)  especially those who were chronically homeless and had appeared to be most resistant to leaving homelessness.
3. The solution was economically worthwhile. Research showed that PSH is cost-effective  that it compares favorably in cost to outlays for public crisis and emergency services used by long-term homeless people, but with a better outcome: ending their homelessness (Culhane, Metraux, & Hadley, 2002; Rosenheck, Kasprow et al., 2003).

By 2001, ending chronic homelessness in 10 years had become a goal of the present federal administration. The federal Interagency Council on Homelessness was revitalized in 2001, and federal agencies mobilized to do their share. Federal agencies worked together to organize Policy Academies to help states begin planning to end homelessness. Nine Policy Academies and one National Learning Meeting were held between November 2001 and November 2005, attracting teams of representatives from the mainstream state agencies whose resources and energies would have to be committed if the goal of ending homelessness were to be achieved. Every state and two territories sent teams of state agency representatives to at least one of the five Policy Academies focused on ending chronic homelessness for individuals or the four Policy Academies focused on ending family homelessness. Almost every state created its own interagency council or task force on homelessness, and as of fall 2006, 13 states had adopted 10-year plans to end homelessness or chronic homelessness (Cunningham et al., 2006). Eight of these states were among the 17 that attended two Policy Academies, another indication of their commitment to do something serious about ending homelessness. Some attended two Policy Academies on ending chronic homelessness, while others attended one on ending chronic homelessness and another on ending family homelessness. The Policy Academies themselves, plus follow-up and technical assistance activities, laid the groundwork for mainstream state agencies, with their extensive resources, to become involved in state and local efforts to end homelessness.

HUD embraced the federal goal to end chronic homelessness by establishing a Government Performance and Results Act objective on homelessness, against which the Department is rated annually. Objective C.3. reads: End chronic homelessness and move homeless families and individuals to permanent housing. To support departmental progress on this objective, HUD used tools such as the competitive CoC grant process to support local change. By 2001, the vast majority of cities, counties, and states were organized into one of the more than 450 continuums of care that HUD stimulated through the annual CoC funding process (Burt et al., 2002). HUD began prompting communities throughout the country to adopt the federal goal as their own by requiring a section on plans for ending chronic homelessness and another on addressing other homelessness as part of annual applications. An increasing number of state and local governments have joined the federal government in formally committing themselves to ending chronic homelessness in 10 years. The majority have gone further, taking on the broader task of ending all homelessness. In the National Alliances analysis of the 90 10-year plans that are complete and have been accepted as state or local policy, 66 percent have the goal of ending all homelessness, with the remaining 34 percent focusing only on ending chronic homelessness (Cunningham et al., 2006).

We recount this history because it is directly pertinent to our task in this paper. A community can offer homeless assistance services for decades without needing, or getting, system change. System change can begin within the homeless assistance system, but the goal of ending either chronic or all homelessness will most likely also require commitment from mainstream public agencies. These agencies, be they city, county, state, or private, include mental health, substance abuse, welfare, health, child welfare, workforce development, criminal justice, and above all, subsidized housing and community development. Moreover, mainstream agency involvement must be active, as these systems themselves need to change if the goal of ending homelessness is to be reached. No community is likely to end either chronic or all homelessness without mapping out a multi-year strategy and moving toward it systematically. The resources and actions of mainstream service agencies are essential to the success of any such strategy. So system change  how to do it, how to know youve done it, and how to show that the changed system is succeeding in ending homelessness  has moved to the forefront of attention.

This paper looks at the process of system change and presents some lessons learned from pioneers in the effort to end homelessness that can be applied more broadly. The paper will also note early successes of system change related to the goal of ending homelessness. The paper does not discuss what an effective system to end homelessness should look like, for two primary reasons. First, system change efforts are still in early stages and we have much to learn before we can draw such conclusions. Second, a good argument can be made that the configuration of a changed system to end homelessness must be defined by local decision makers. Other papers in this Symposium may tell us what works for specific populations, and local decision makers may pick and choose among the best. But the balance of system elements will still depend on local factors.

Synthesis of Research Literature

System change has interested people in many disciplines, in part because it is by all accounts so hard to do and hard to sustain. Corporations and businesses care about system change because a poorly functioning corporate system means lower profits. The most successful approaches to assuring improved educational outcomes for the most disadvantaged children rely on changing educational systems, from individual schools to whole districts, through comprehensive school reforms (Borman et al., 2003). Helping the most disadvantaged and hardest-to-serve welfare recipients to get and retain jobs has required system changes involving welfare and workforce development agencies, and sometimes mental health, substance abuse, and other agencies (Martinson & Holcomb, 2002). Children- and family-serving agencies have long sought system change to increase the effectiveness of service delivery systems (Burt, Resnick, & Novick 1998; Melaville & Blank, 1991). Is it any wonder that in the homelessness arena we also find ourselves in need of guidance to move systems toward greater responsiveness?

Our assignment is to summarize research knowledge about changing community systems into configurations that promote the goal of ending homelessness. It is important to note at the outset that significantly less relevant literature exists on this topic than on others at the Symposium. A Google search of system change + homelessness produces 66,000 items, but only a handful are research  most of the rest are plans, or advice. The research on which we base much of this paper comes from HUD-sponsored projects on community-wide strategies to end chronic street homelessness (Burt et al., 2004) and prevent homelessness (Burt, Pearson, & Montgomery, 2005); a Corporation for Supportive Housing evaluation of a project called Taking Health Care Home (THCH) that is designed to change community systems to promote development of permanent supportive housing (Burt & Anderson, 2006); and two research syntheses offering blueprints for changing systems, one by HHSs Center for Mental Health Services (2003) and one by the Corporation for Supportive Housing (Greiff, Proscio, & Wilkins, 2003) that has been the basis for many presentations at Policy Academies. Most of the research has focused on permanent supportive housing to end homelessness for persons with disabilities who have been homeless for a long time. Yet system change efforts related to homelessness reach well beyond this population and these interventions. Therefore, we also incorporate examples from our own experience working with communities to make change happen.

We will address several aspects of system change based on research and written reports that have become available since 1998: (1) documenting system change itself and how it has been brought about, (2) documenting the effects of such change on preventing and ending homelessness, and (3) describing how communities have used a variety of databases and feedback mechanisms to give themselves the information they need to set targets and keep themselves on track to meet them. We will not be able to recommend best practices substantiated by a strong evidence base, but we will be able to present approaches and practices that are widely recommended and seem to be promising.

How Shall We Describe Systems and System Change?

The literature offers a number of schemes for describing systems and system change. We use two in this paper. The first focuses on signs that systems have changed, and the second focuses on the types of relationships among agencies that characterize systems at different stages of integration. Both schemes were used in the THCH evaluation (Burt and Anderson, 2006) to describe the changes occurring in the study communities.

Laying a New Foundation (Greiff, Proscio, & Wilkins, 2003, p. 7) identifies five signs by which one can recognize system change when it is complete, or nearly complete; change should be clear in all five areas (text in brackets [ ] is the present authors):

• A change in power:  There are designated positions  people with formal authority  responsible for the new activity (not just committed or skillful individuals who happen to care about it).
• A change in money:  Routine funding is earmarked for the new activity in a new way  or, failing that, there is a pattern of recurring special funding on which most actors in the system can rely. [This could be new money, a shift in existing funding, or new priorities and criteria for accessing existing money].
• A change in habits:  Participants in a system interact with each other to carry out the new activity as part of their normal routine  not just in response to a special initiative, demonstration, or project. If top-level authorities have to command such interactions to take place, then the system has not absorbed them, and thus has not yet changed. [Service delivery improvements fit in here, ranging from referral hotlines and simplified application procedures, through case-by-case provider sharing of resources, up to and including services integration (through multi-agency teams, co-location, and the like) or systems integration (such as universal applications, merged funding streams, multi-agency goal-setting and follow-through)].
• A change in technology or skills:  There is a growing cadre of skilled practitioners at most or all levels in the delivery chain, practicing methods that were not previously common or considered desirable. These practitioners are now expert in the skills that the new system demands and have set a standard for effective delivery of the new systems intended results.
• A change in ideas or values:  There is a new definition of performance or success, and often a new understanding of the people to be served and the problem to be solved [i.e., new goals]. The new definition and understanding are commonly held among most or all actors in the system, such that they are no longer in great dispute. [For instance, a whole CoC could reorient itself toward ending homelessness, or at least toward ending chronic homelessness. Either of these events would be system change if followed by actual changes in behavior to assure movement toward the goal.]

Since people who are homeless interact with many systems, including homeless-specific agencies and the health, mental health, corrections, child welfare and foster care, public benefits, employment, and housing systems (as documented by Culhane et al., 2002 and Koegel et al., 2004, among others), achieving integration of these systems can make a significant difference in the manner and speed with which a households homelessness is resolved. Services and systems may be integrated to varying degrees, making it more or less simple to get individuals the range of services they need or to end homelessness through the combined, concerted, organized, and strategic actions of many different actors (Cocozza et al., 2000; Provan & Milward, 1995; Randolph et al., 2002). Services integration refers to the ability of a community to get any individual or family the services it needs, especially when the needs span two or more service systems. Services integration may be accomplished in a number of ways  a common approach is the multi-agency casework team, whose members are able to marshal the resources of their respective departments efficiently and effectively to help individual clients. Systems integration refers to changes in two or more service systems that reorient the systems activities toward more efficient and effective achievement of common goals  goals that may be new or long-standing.

The first author (Burt & Anderson 2006; Burt et al., 2000) has used a five-level scheme to describe integration stages  isolation, communication, coordination, collaboration, and coordinated community response. These stages can represent the initial status of a potential system and the relationship of its component parts, and also the movement toward changes that are likely to end homelessness.

One can use the integration stages described below to benchmark a communitys progress from a situation in which none of the important parties even communicates, up to a point at which all relevant agencies and some or all of their levels (line worker, manager, CEO) accept a new goal, efficiently and effectively develop and administer new resources, and/or work at a level of services integration best suited to resolving the situation of homelessness for the largest number of people in the shortest period of time. The framework also recognizes the possibility of regression from one stage to previous ones if prevailing factors work against integration. Brief descriptions of these integration stages follow:

• Isolation  recognition of the need to communicate about the issues that require a system solution is lacking, as is any attempt to communicate. Even worse than isolation is hostile communication, suspicion, and distrust. This was the situation in many communities at the time that HUD instituted the continuum-of-care application process. It still prevails in some communities as the reality of relationships between homeless assistance providers and government funding agencies.
• Communication  talking to each other and sharing information in a friendly, helpful way is the first, most necessary, step. Communication must inform participants what their counterparts in other agencies do, the resources they have available to them, and the types of services they can offer. Communication may happen between front-line workers (e.g., a mental health worker and a housing developer), middle-level workers, and/or among agency leadership. It may occur among these personnel in two systems, three systems, and so on up to all the systems in a community. In many communities the parties who need to work together to create a coordinated system to end homelessness have not reached even this first stage. Everyone operates in isolation in hostile interactions that do not advance understanding or assistance for homeless people or the possibilities of preventing homelessness. Even when people know each other and sit on the same committees and task forces, they still may not communicate enough to share an understanding of the role each could play in ending homelessness. This latter situation is the norm in most communities  people know each other but have not really gotten down to the hard work of listening to and hearing each other.
• Coordination  staff from different agencies work together on a case-by-case basis and may even do cross-training to appreciate each others roles and responsibilities. Again, coordination or cooperation may happen among front-line workers or middle-level workers, and/or involve policy commitments for whole agencies by agency leadership. It may occur among these personnel in two systems, three systems, and so on up to all the systems in a jurisdiction.

  Coordination may also be services integration. Multi-agency teams that help specific individuals obtain appropriate services are examples of coordination, as are multi-service centers where a homeless person can connect with many different agencies but there is no overall case coordination. However, at this stage, no significant changes have occurred in the services each agency offers or how the agencies do business. Coordination does not involve major changes in eligibility, procedures, or priorities of any cooperating agency. It merely means they agree not to get in each others way and agree to offer the services they have available when it is appropriate to do so, albeit sometimes in new locations or through new mechanisms such as a multi-agency team. It does not entail any significant rethinking of agency goals or approaches.

• Collaboration  collaboration adds the element of joint analysis, planning, and accommodation to the base of communication and coordination, toward the end of systems integration. Collaborative arrangements include joint work to develop shared goals, followed by protocols for each agency that let each agency do its work in a way that complements and supports the work done by another agency. Collaboration may occur between two or more agencies or systems, and usually does involve system change to varying degrees.

  Collaboration cannot happen without the commitment of the powers-that-be. In this respect it differs from communication and coordination. If agency leadership is not on board supporting and enforcing adherence to new policies and protocols, then collaboration is not taking place (although coordination may still occur at lower levels of organizations). Because collaboration entails organizational commitments, not just personal ones, when the people who have developed personal connections across agencies leave their position, others will be assigned to take their place. They will be charged with a similar expectation to pursue a coordinated response and will receive whatever training and orientation is needed to make this happen. Collaboration in this sense can be seen in many examples given throughout this paper, including Connecticuts three waves of integrated state funding for PSH, the ways the Massachusetts Department of Mental Health has developed partnerships to produce PSH, Minnesotas 10-year plan to end homelessness, Portland/Multnomah County, Oregons three-way funding structure for PSH, Seattle/King County, Washingtons funders group, and Columbus/Franklin County, Ohios Rebuilding Lives initiative.

To the three stages that promote better services and supports for homeless people, we add a last stage, which is collaboration involving all of the critical and most of the desirable systems and actors in a community. This type of response has sometimes been called a coordinated community response (CCR), and we adopt that terminology here to distinguish this type of community-wide collaboration with the long-range goal of ending homelessness from collaboration among two or three agencies. Coordinated community response is system change and integration, going beyond collaboration in several directions.

• First, all of the systems in a community essential to preventing and ending homelessness must be involved. This includes homeless assistance providers and agencies providing housing subsidies, and also those promoting the development of affordable and special needs housing. It includes agencies that fund supportive services, most frequently mental health and substance abuse agencies, but also employment and health agencies, and others offering services that may be needed to address the underlying factors that contributed to homelessness. It includes agencies such as law enforcement and corrections, mental hospitals and private psychiatric units, and other institutions discharging vulnerable people with disabilities who are at risk of homelessness and need appropriate housing. It often involves the business community, which is heavily impacted by street homelessness. Ideally, others will also be involved, including representatives of local elected bodies, funder representatives, and consumer representatives.
• Second, CCR involves a mechanism for seeing that individual clients or households receive the services they need  that is, it integrates services, through one or more of several mechanisms. The result of this streamlined service delivery at the client level should be improved client outcomes as well as more efficient and effective use of resources. In the context of addressing chronic homelessness, service integration involves connecting services and housing to help clients with long-term homelessness and one or more disabling conditions to find and keep housing and reduce use of expensive emergency public services. An important finding of the Access to Community Care and Effective Services and Supports (ACCESS) demonstration, which may seem obvious in hindsight but was not actually anticipated, was that people got housed only when the housing agencies were at the table (Rosenheck et al., 1998, 2001, 2003b). In the context of preventing or ending family homelessness, weekly cross-system case management meetings and pooled resources among homeless intake, child welfare, and income maintenance agencies may be used to move families coming into shelter rapidly back into housing or even to keep children with their parent in permanent housing instead of allowing the family to become homeless and removing the children to foster care. By working together and developing the mechanisms to respond to their clients housing crises before a household becomes literally homeless, providers can intentionally serve all clients rather than opportunistically serve only those who come to them while others fall through the cracks.
• Third, CCR entails a functioning feedback mechanism. In many communities this is a monthly (or more frequent) meeting of those most actively involved in developing appropriate interventions or smoothing bureaucratic pathways. (This function should be different from a direct service meeting to facilitate matching clients with services and housing units, even though both meetings may involve the same players.) Some communities have also found that forcing themselves to collect data on their progress and then to review the data at the monthly meetings shows them what they have achieved, helps them identify and resolve bottlenecks, and provides a powerful positive incentive.
• Fourth, CCR includes an ongoing mechanism for thinking about what comes next, asking what needs to be done, how best to accomplish it, and, finally, what needs to change for the goals to be accomplished. This mechanism can take one or more forms, such as task force or council, regular stakeholder meetings, and quarterly retreats. Whatever the mechanism, it must translate into shared decision-making and strategic planning at multiple levels as well as the expectation that each part of the system will modify its own activities to support and complement the work of the other parts.
• Fifth, it is a great deal easier to maintain the first four elements of a CCR if someone is being paid to serve as coordinator to organize and staff the interagency working groups and committees necessary to accomplish community-wide goals.
• Finally, a coordinated community response is never a done deal. If it is really doing everything expected, including identifying remaining gaps and continuing to seek ways to improve the system, it continues to evolve. We do not attempt to assess communities discussed in this paper using this framework except in a few examples, but changes from one stage to another should be obvious from community changes described below. The evaluation section of the paper discusses how the framework can be used to measure the impact of system change efforts as they mature and evaluations are formalized.

It is most fruitful to use this scheme to characterize movement and change rather than a steady state or a comprehensive overview. We follow this principle in Exhibit 1 below, where we give brief examples of movement from one level to another, focusing sometimes on relatively narrow but still challenging integration efforts such as that of the Skid Row Homeless Healthcare Initiative in Los Angeles and sometimes on the broadest possible efforts to mobilize all elements of a community to address the ultimate goal of ending homelessness.

Factors Affecting the Likely Success of System Change Efforts

From the review of existing research and observations of local community practice, the authors have identified five major factors that affect system change:

• context of the local community and the state,
• interest and commitment of key stakeholders,
• scope of desired system change,
• governance and management structure for system change, and the
• identified process of system change.

The mix of factors will vary from one community to the next; thus, the pattern and success of system change will also vary. The review of current research suggests that no one factor of system change is more important than another, but there do appear to be cumulative impacts. That is, having multiple factors in place, such as strong state agency support and a dedicated staff member managing the system change activities, may help overcome obstacles to system change. Conversely, the absence of two or more factors may significantly hamper progress toward system change. Presumably, the more complete, strategic, and well-executed the process, the faster the goals will be realized and the greater will be the magnitude of the results.

Context of the Local Community and the State

The starting frame of reference of the local community will impact the speed of change and may affect a communitys ability to mobilize stakeholders. The community leaders driving change will need to assess the current stage of the system (isolation, communication, coordination, collaboration, or coordinated community response). Readiness for change is affected by the occurrence of trigger events that mobilize community support, whether providers are content or dissatisfied with their current methods of addressing homelessness, prevailing philosophies and level of investment related to the current system, availability of data to compel change, the economic and social climate that may affect a communitys ability or willingness to redirect resources to address homelessness, and commitments of those who control major resources beyond the community itself. For instance, if public agencies and homeless assistance providers alike acknowledge that current approaches are not effective in addressing homelessness, there is a shared context for discussing possible solutions that will probably involve system change. If no trigger events have provided the impetus for change or if no data show that the current system is failing, then many may resist change altogether. If agencies are already collaborating, then they already have skills and experience working together on community solutions and will have an easier time taking the next step (Pindus et al., 2000; Martinson & Holcomb, 2002).

Trigger Events and Paradigm Shifts

A recent HUD-sponsored study examined seven communities making progress in ending chronic street homelessness (Burt et al., 2004).^[1]  The study identified the importance of a trigger event in mobilizing significant commitment to developing new approaches to ending homelessness for this most resistant segment of the homeless population. In most of the communities visited, a trigger event galvanized the observed approach. In Columbus and San Diego, the event was the desire to develop a part of downtown that had a high concentration of street homeless people. The business leaders who wanted the development became committed to assuring that it did not happen unless plans were in place to serve and house the homeless people it would displace. In Philadelphia and Birmingham the trigger event was a proposed anti-homeless city ordinance. Consumer and service provider protests in Los Angeles, Philadelphia, and San Diego stimulated responses in those communities, and an invitation to develop a pilot program for a new funding source prompted the Los Angeles County Sheriff and Mental Health Department to work together for the first time to create an integrated services program for homeless people with mental illness who were leaving the jail with no place to go. The two communities that already had strong organizational structures and leadership (Columbus and Philadelphia) were able to capitalize on these trigger events with relative ease and speed. But it is important to note that several communities and public agencies that did not have an organized leadership structure or well-developed public agency involvement and investment before the trigger event (for example, San Diego and two programs in Los Angeles) were able to use the event to re-examine their situation, decide to take action, organize themselves, mobilize resources, and make and carry out plans for approaches to address and reduce chronic street homelessness. Thus, these communities were able to turn these trigger events to their advantage and gain commitments to new goals and new resources, rather than allowing the event to worsen the circumstances of street homeless people. The event itself is often perceived locally as a watershed moment  the catalyst that began the process that resulted in the current commitment to reduce or end chronic street homelessness.

Frustration and Philosophy

If homeless assistance providers feel their current approaches are working to end homelessness, they may resist efforts at system change. Alternatively, some providers and some public agencies committed to a high-demand approach will not consider certain models that have been shown to work with very service-resistant homeless people but which conflict with their philosophical viewpoints. But if providers themselves are frustrated with the current models or feel that the current system is ineffective in engaging certain populations, they are more likely to welcome change. For instance, in Chicago it was the providers, not the city, that came back from national conferences saying what were doing is not working, we have to do something different and lobbied until they got a 10-year planning process under way. The process of change does not stop with a recognition that something different is needed, and even high frustration levels do not necessarily lead to change without other ingredients also being present. For instance, providers may not have the knowledge or organizational capacity to move to new approaches even if they want to, and funders may be stymied by contractual processes that are hard to shift or change. Later in this paper we give examples of system change efforts designed to help existing and potential new providers develop the skills needed to operate new kinds of programs. A community-wide system change effort will need to account for all of these factors as it moves forward and find ways to accommodate each stakeholders current position while working toward a more effective overall system.

Using Data to Support Change

Accurate information is always a powerful weapon in the quest for change, but it is not always easy to find. Most critically, accurate information about ones own community is the most convincing to local decision makers, but often it does not exist. We began this paper with a fairly dramatic example of the use of data to effect change  how the National Alliance to End Homelessness along with other advocates used information about population size, the success of permanent supportive housing, and public costs avoided to promote a federal commitment to ending chronic homelessness. Local communities also generate and use data in a variety of ways to stimulate change and then sustain and expand the investment. Later in this paper we discuss ways that communities use information to manage change efforts, but here we want to offer some examples of how communities have made sure that they developed the local information they needed to prompt commitments to system change.

• In 1986, Columbus, Ohio, was one of the first communities in the country to develop and install a simple homeless management information system (HMIS) to track shelter use and get an accurate count of homeless people in the system (Burt et al., 2004). The Community Shelter Board director at the time, who insisted on installing the system, later said, Once we had data, we stopped arguing about whether we had a problem and started working on how to solve it.
• Hennepin County, Minnesota, also developed its own data system for its homelessness prevention and rapid shelter exit programs for families. The county staff use the system daily to manage the programs and to assess system outcomes  specifically, whether families who receive program services become homeless or return to homelessness (Burt, Pearson, & Montgomery, 2005). Being able to show the programs excellent track record with homeless families entering shelter (reducing the number of families in shelter by half, halving lengths of stay, and keeping further loss of housing and return to shelter within 12 months down to only 12 percent) has been instrumental in keeping the programs state funding flowing.
• Portland, Oregons Bureau of Housing and Community Development collected impact data on a pilot project that it hoped would become a model for future programming to end street homelessness. Transitions to Housing offers providers whatever it takes flexible funding to house and support the hardest-to-serve single homeless adults. Politicians were skeptical but willing to back a pilot. When the evaluation data showed clear success, it was the starting point for expanding the program and moving forward with more system-wide changes.
• In Seattle, a study by the county health department noted many deaths among single adult homeless people in King County. This study had a very powerful effect in generating political will because it got a lot of press in local newspapers. Public attitudes really drive the agenda at the state capital, and the study created strong public interest in reducing the vulnerability of homeless street people, for which permanent supportive housing was a clear solution. The study came at a time when the Taking Health Care Home (THCH) project was working to develop more resources for PSH and had created a Funders Group to think through how this might be done. The THCH coordinator was asked to make a presentation to the state legislature using THCH data and the results of the Funders Group deliberations. This testimony provided valuable information to state legislators who were attempting to address homelessness across the state and ultimately helped promote major new appropriations and other legislative initiatives.
• Californias very successful AB 2034 program (Burt & Anderson 2005; Mayberg, 2003) grew out of data from three initial pilot sites that showed significant reductions in time homeless and days hospitalized or incarcerated among homeless people with serious mental illness who participated in the programs whatever it takes funding approach. The evidence of success continued as the program expanded to 34 counties and was one of the key factors that prompted voter approval of Proposition 63 in 2005 and the consequent Mental Health Services Act in 2006, which is pouring major new service dollars into California communities.
• The San Diego Police Department gathered data to show the cost of one arrest and booking of a chronic alcoholic homeless street person, in an effort to develop support for a new approach. When decision makers learned that the cost of just one arrest (about $1,100) was more than one-third higher than the cost of one month of outpatient treatment and housing, San Diegos Serial Inebriate Program was born (Burt et al., 2004). Its success was one of the factors leading to HUDs new Housing for People Who Are Homeless and Addicted to Alcohol program, now funding 11 grants to 10 communities.

Economic and Social Climate

All of the factors just discussed are affected by the economic and social climate of the local community or state environment. The economic environment will affect the overall prevalence of poverty within a community and may affect a localitys revenue base available to address poverty issues. A poor job climate will make it even more challenging for homeless households to obtain living wage employment. An expensive housing market will expand the gap between market rents and incomes of households that are trying to avoid homelessness or re-enter the housing market, while in a depressed housing market, landlords are more willing to negotiate rents and payment plans to repay arrearages. A positive economic environment can present positive opportunities for system change to end homelessness, such as funds to support innovative service models or the development of subsidized and/or supportive housing.

The social environment can have an equally powerful effect. If there is significant social awareness and public support for social causes in general and ending homelessness in particular, community leaders may be very receptive to pursuing an agenda for change. If the community is negatively inclined toward social issues, the political leaders may be completely opposed to funding or even supporting change. Similarly, if the community is mobilized around different community issues, it may be difficult to secure public support for system change to end homelessness.

Interest and Commitment of Key Stakeholders

Various sets of public and private, homeless and mainstream system actors need to make commitments and play their parts for systems to change. A community is more poised for successful system change if all of the stakeholders share the goal of ending homelessness, are committed to bringing the goal to fruition, and are open to changing their own systems to make it happen. However, even if only a few agencies are on board, the agencies may act as champions of the process to engage other stakeholders. Ideally the founding partners will be agencies that are pivotal to change, but the specific agencies involved will vary from community to community depending on the population being targeted and the structure of the community. For instance, if a community is targeting family homelessness, critical agencies may be the child welfare, TANF, and workforce development agencies; the public housing authority; and key homeless system leadership. If chronic street homelessness is the issue, law enforcement; the courts; and mental health, substance abuse, corrections, and public benefits agencies will likely be involved in addition to homeless assistance agencies and, sometimes, the business community.

It is likely that several of the key stakeholders will not be at the table at the beginning of the process, and they will need to be convinced to participate. All stakeholders do not have to be involved from the beginning, nor do all stakeholders need to be involved in all aspects of system change. Different communities have had success using different models. Some work on system change within the homeless system, slowly engaging one mainstream agency at a time; some work with several mainstream agencies to develop one component of a community system such as PSH; and some start with mainstream agencies and work on changing the homeless system in later stages. The local context and motivation for change will determine which strategy is likely to work best.

Need to Involve the Agencies with Resources and Decision-Making Authority

Local communities seldom control key resources or are in a position to make policy decisions essential to ending homelessness. A city will be dependent on cooperation from county agencies that control key resources such as public benefits and health and mental health services. Cities and counties will be dependent on state agencies and their policies, especially policies affecting resources essential to addressing homelessness, including housing, health care, mental health care, and substance abuse treatment. As homelessness is, at base, lack of housing and the ability to afford housing, a local effort to end homelessness will have a much better chance for success if the agencies that can offer housing or that control housing policy are at the table. These include public housing authorities, state housing finance agencies, and community and economic development agencies. Many of the critical housing agencies have an autonomous or semi-autonomous status, being neither city, county, nor state agencies in the usual sense, adding another layer of who controls what to the mix of agencies needed for success.

For the past three years, the first author has been involved in evaluating the multisite Taking Health Care Home (THCH) initiative of the Corporation for Supportive Housing (CSH).^[2]  This project is designed to move systems in a direction that will promote the development of permanent supportive housing, using a grant as its primary lever for moving systems. All THCH sites invested a portion of grant resources in a coordinator. A recent report (Burt & Anderson, 2006) examined the changes in the study communities at the two-year mark.

Three THCH communities (Portland/Multnomah County, Seattle/King County, and Maine) were the most ready for change, in that at least one public agency had already realized the importance of PSH and had taken its own steps to move more of its resources toward PSH development.^[3]  The most involved agency in each community took the lead in applying for THCH funds, usually on behalf of a large collaborative body that was already in existence or with the explicit commitment of at least one other agency to work toward system change. The early buy-in of these agencies laid the foundation for relatively rapid and successful system change once the THCH funding was received and a coordinator was assigned to manage the change process.

These three communities differed substantially in the degree to which state agencies were involved in their system change efforts. Portland/Multnomah County proceeded largely without state-level involvement, as the climate for such involvement was unfavorable to investment in homeless issues and no other communities in the state could be counted on to apply pressure to turn that resistance around. Seattle/King County might have found itself in the same situation, were it not for the THCH investment strategy that put resources into both ends of the state  in Seattle/King County and in Spokane. From the beginning, an element of the Washington strategy was to develop new state-level funding streams; generating pressure from areas of the state other than the largest population center of Seattle/King County would be critical for success. Another part of the strategy was to include state agency representatives in the Seattle/King County funders group. These representatives became very familiar with the arguments for ending chronic homelessness through permanent supportive housing, and were later instrumental in helping to design a strategy for new state legislation and getting that legislation passed. By the end of two years of organizing, the Washington legislature had approved new legislation that is now providing resources to combat homelessness in every county based on real estate transactions, plus resources to combat family homelessness, and new resources for substance abuse treatment. In Maine, the THCH project was located in a state agency, the housing finance agency, and the primary work of the project involved organizing agencies at the state level. Action at the state level to provide capital resources, facilitate operating resources, and match clients to supportive services through Medicaid and other mechanisms has supported the work of regional councils that do the bulk of local planning, and the work of local providers that develop and run the programs that deliver services to actual clients.

There are other examples of states that have developed state-level strategies and resources for combating or ending homelessness that facilitate local planning and implementation and make it easier for providers in local communities to meet the needs of individual clients.

• In 2004, Minnesota promulgated a state 10-year plan for ending long-term homelessness (Minnesota Departments of Human Services and Corrections, and Housing Finance Agency, 2004). The plan calls for the development of 4,000 new units of permanent supportive housing within seven years at an estimated cost of $540 million. State sources were projected to supply two-thirds of this amount, including capital resources ($90 million in general obligation bonds, $90 million in housing finance agency resources, and $60 million in tax credit financing) and supportive service resources ($120 million through the Department of Human Services, including state appropriations and various public benefits). The working group that developed the plan issues regular progress reports. As of fall 2006:
  • 1,091 of the 4,000 promised new units had received funding commitments and were underway.
  • The legislature appropriated and the Department of Human Services awarded the first $10 million in state funding for supportive services to seven multi-county consortia.
• California has recently passed several important pieces of legislation or voter initiatives that make new resources available for addressing homelessness and the disabilities that often keep people homeless for long periods of time. These include: 
  • 2006  a new state housing bond issue for $2.85 billion to create affordable housing throughout the state, with a component aimed directly at developing permanent supportive housing for chronically homeless people.
  • 2005  Proposition 63, which became the Mental Health Services Act in 2006, provides close to $200 million a year statewide. Allocations are up to each county, but it is expected that a significant portion will be used to provide the supportive services that help keep people with severe and persistent mental illness in housing.
• In November 2005, New York announced a new wave of the New York/New York Initiative, known as New York/New York III, to create 9,000 new units of permanent supportive housing by the end of 2015. As did earlier waves, the first of which was signed in 1990, this third wave of combined state and city funding will focus on ending or preventing homelessness among single adults with severe and persistent mental illness. It will have a broader focus as well, serving single adults with substance abuse disorders or HIV/AIDS, families with a disabled head of household, and youth aging out of foster care.
• Investments in Connecticut and Massachusetts are described in more detail later in this paper.

These examples illustrate the importance of and potential results associated with involving agencies with resources and decision-making authority, particularly at the state level. Without their intimate involvement, it will be significantly harder to effect changes in power or money. With their involvement and support, these agencies may identify problems within their own systems and suggest solutions to address them. It is also important to note that some of these state-level changes focus on assisting the larger categories of extremely poor people or people with a certain disability rather than just people who have already become homeless. By implication, they also involve stakeholders who may not be directly involved in ending homelessness but who can be significant allies in securing policies that should reduce homelessness by reducing the likelihood that people in these categories will become homeless in the first place. Increasing the availability of affordable housing, whether through rent subsidies to low-income households or public investment to reduce capital costs, is probably the single biggest public policy that could affect levels of homelessness (Quigley, Raphael, & Smolensky, 2001; Dasinger & Spieglman, 2006). Assuring housing with supportive services to populations whose disabilities, coupled with extremely low incomes, are known to increase their vulnerability to becoming homeless is another non-homeless-specific strategy that could have a substantial impact in reducing the flow into homelessness.

Beyond The Usual Suspects

In the discussion above, we have talked mostly about the roles of the usual suspects  homeless assistance providers and government agencies whose missions connect to homelessness through funding, direct service, or both. Communities that have succeeded in involving a wider variety of stakeholders have found their presence to be useful in many ways. The participation of state and local elected officials can be critical to securing the funding needed to carry out the new plans, and also to helping interpret and champion the new plans to the general public. Business associations and business improvement districts have participated in developing and implementing plans to end homelessness, and have also contributed significant resources and developed service structures of their own in Philadelphia; Denver; Washington, D.C.; Columbus, Ohio; and many other communities. Community leaders were the main participants in Reaching Home, Connecticuts public education campaign that sought to win public support for state investments needed to end chronic homelessness. Foundations, such as the Melville Charitable Trust in Connecticut, have played major leadership and funding roles in some communities. The Conrad F. Hilton Foundation is another example  the foundation recently invested $8 million in efforts in Los Angeles to reduce chronic homelessness among people with serious mental illness, and pursues a number of initiatives to stimulate the city and county to develop and implement approaches to ending homelessness that have a known track record of success.

Scope of Desired System Change

The extent to which there is a shared vision for ending homelessness is likely to affect the success of system change. For some communities, it is more strategic and feasible to focus on solutions for chronic homelessness; for others, it is important to establish a broader vision to bring critical partners to the table. Communities may need to consider the implications of the scope of their goal, and whether system change needs to be organized separately for different subpopulations or aspects of the goal.

Stakeholders in Jacksonville, Florida, mobilized to establish the Home Safe project to permanently house individuals with chronic alcohol addictions who had been living on the streets or in emergency shelter for extended periods.^[4]  The opportunity to apply for federal funds to address the issue provided impetus for the collaboration to form. The focus on alcohol addiction brought new partners to the table, many of whom had not previously been involved in addressing homelessness. The collaboration involves the homeless coalition, local sheriff, two state-funded substance abuse treatment providers, a key homeless assistance provider, and the mental health center  all of which are working together to address a shared problem. The project has resulted in shared funding, joint decision-making, and regular service planning across all of these partners. Although the collaborative is currently limited to this single project, it has provided a positive experience that can be leveraged for future system change efforts.

Conversely, Indianapolis, Indiana, chose to adopt a Blueprint to End Homelessness that defined strategies for preventing and ending all homelessness, including family and short-term homelessness as well as chronic homelessness for individuals (Indianapolis Housing Task Force, 2002). The leaders of this effort determined that establishing a goal of sweeping change affecting a broad constituency was a more appropriate strategy for engaging the wide range of stakeholders they thought would be needed to achieve system change. Ambitious goals multiply the amount of work needed to create change, but they also expand the pool of willing funders, advocates, and allies. There was concern that a narrower goal might alienate potential allies. System change is still underway, and the Blueprint has continued to maintain widespread support. The community has achieved several critical implementation milestones, including creating new permanent supportive housing units, establishing the Marion County Housing Trust Fund (www.ahomewithinreach.org) and a new affordable housing placement clearinghouse (www.IndianaHousingNow.org) to expand access to permanent housing, successfully piloting two new cross-disciplinary housing and services initiatives (use of HOME tenant-based rent assistance to move families out of shelter, use of a mental health system of care model to provide resource coordination to persons who are chronically homeless), and sponsoring cost studies to measure the primary and behavioral health cost savings associated with the system-of-care pilot. Stakeholders are currently working on a detailed implementation plan that incorporates major funding shifts, investment of new resources from a broad range of mainstream housing and service agencies, and a carefully planned conversion of the homeless shelter system.

Governance and Management Structure for System Change

To achieve system change, communities will need to make decisions about the level of inclusion they need or intend in their processes, how the group will make decisions throughout the process of change, and how the group will manage the process of change. The structure will need to be closely related to the previously described factors of community context, commitment of key partners, and the goal that is established. Inevitably, the community can be more successful if its efforts are intentional and it establishes a leadership, decision-making, and management structure that fits its anticipated goal and process.

Leadership and decision-making structures can form from the top down or the bottom up, or can be a hybrid. Research indicates successes with all models, so the clear lesson about structure is that it should be what fits a given community. There is no one size fits all; attempts to impose one communitys structure on another community will usually waste time and possibly delay or derail the process.

Communities with a strong funders network or other powerful actors may organize themselves to streamline access to resources through a central organizing body such as the Community Shelter Board in Columbus, Ohio. The resulting top-down structure uses its central control of resources to drive change to address homelessness, once the direction of change has been established through a process of community-wide input. Communities such as Indianapolis, on the other hand, which rely on privately funded faith-based providers to run all of the emergency shelters and the majority of transitional housing programs, may need to employ a more bottom-up engagement model.

In a process in which providers are driving change from the bottom up, it may be hard to get mainstream agencies to the table. For instance, providers in Kansas City formed the Mid-America Assistance Coalition (MAAC), a collaborative of providers, to manage efficient distribution of limited Emergency Shelter Grant and Emergency Food and Shelter Program funds. This collaboration has been an effective solution for the original problem; however, it has not been able to get mainstream agencies to the table or to leverage additional resources to achieve the level of system change needed to truly impact homelessness (Burt, Pearson, & Montgomery, 2005).

On the other hand, if the process is driven from the top down, whether by government or private entities, providers and even local public agencies may distrust the process and resist change. The Annie E. Casey Foundations experience with its Building New Futures initiative, which gave $10 million, five-year grants to states and localities to promote extensive change in systems responsible for addressing the needs of high-risk youth, is an example in which the top is a private foundation with its own preferred vision of a changed system (Nelson, 1995). Federal government efforts to stimulate system change often face similar experiences. In the homelessness arena, for instance, local government agency partners in some HUD/HHS/VA and HUD/DOL Chronic Homeless Initiative projects faced the situation of having to comply with federal guidelines they had no hand in shaping. In some of these communities, a proposal was written with some official signoff from the participating public agencies but without the knowledge of the line staff who would have to be the collaborators. It took some time for the relationships to work out, especially since the federal grant conditions and specifications sometimes conflicted with established procedures of both public agencies and private providers.

Most communities appear to have a collaborative approach to managing system change that works from both directions. Regardless of structure, communities poised for system change must recognize that change is difficult and there will be times when stakeholders will disagree. When this happens, will the group rely on a consensus model or one in which a majority rules? Does every stakeholder get a vote in decision-making or only those that control funding or regulations? It is essential to define and document a process for making decisions related to system change from the beginning, preferably as part of a memorandum of understanding, before becoming embroiled in the many difficult issues that are inevitable when a community truly intends to change systems.

Identified Process of System Change

The process by which a community implements its shared vision will vary depending on all of the previously mentioned factors, but perhaps will be most significantly affected by the beginning state of the system (proportion of elements operating in isolation, communication, coordination, or collaboration) and the scope of the communitys vision. The process should focus on actions that will change power, money, habits, technology or skills, and ideas or values by concentrating on moving system elements from isolation to communication, from communication to coordination, and so on. These actions should be strategic and intentionally planned, though flexible enough to afford regular opportunities to revisit the course of action and redirect resources as needed.

A recent analysis by staff at the Corporation for Supportive Housing (Grieff, Proscio, & Wilkins, 2003) integrates the experiences of many communities to identify lessons learned about promoting policy reforms and developing coordinated systems of housing for long-term homeless adults with disabilities. The lessons are pertinent to all efforts at system change; they are presented linearly below, but they may occur in any order or simultaneously, and they work best if they are applied in continuing cycles of assessment and action. The Center for Mental Health Services (2003) incorporates many of the same steps in its guidance to communities on ending chronic homelessness for persons with serious mental illness. The steps are:

• fostering collaborative planning and consensus building;
• investing and leveraging resources;
• coordinating, streamlining, and integrating funding;
• building provider capacity;
• establishing and monitoring performance, quality assurance;
• building the case for system change through research and data;
• communicating and advocating: finding ways to make the need for system change compelling;
• cultivating leaders, champions, and advocates;
• capitalizing on trigger events that compel action; and
• designating an intermediary in the role of neutral catalyst, or coordinator.

The rest of this paper synthesizes the approaches and practices that research and our own experience working with communities indicate are promising ways to change systems for the purpose of preventing and ending homelessness.

Mechanisms That Facilitate Implementation of Change Goals

As hard as it is for communities to come together at the conceptual level to agree on new goals and new responsibilities, it is considerably harder to bring the new vision into being. Over the years communities that adopted the goal of ending chronic homelessness have developed a variety of mechanisms for implementing change. We focus here on four types:

• mechanisms that stimulate providers to bring their programs into line with the new goals;
• mechanisms that match homeless people to the most appropriate services and programs;
• funding mechanisms that help bring together the array of resources needed to develop and support homeless assistance programs and homeless people and support integration of mainstream and homeless systems; and
• the role of a coordinator to bring it all together and make these and many other things happen.

Re-Orienting the Continuum of Care

If communities are really going to end homelessness in 10 years, everyone who now provides homeless assistance will have to change to varying degrees, and new participants will also have to join in the effort. One Portland informant described the process of reorienting their whole community toward ending homelessness as turning the ocean liner; another described the reality of how many small steps this takes:

First, all of us working on the 10-year plan had to decide what the right thing to do was. After weeks of discussion, our decision was to develop PSH that prioritizes the hardest-to-serve people. Then we had to convince providers that they should adopt these priorities as their own. Even after they were convinced in theory, it soon became clear that providers did not really know what the change would mean in practice. That is, their habits had not changed. In their program structures and client recruitment practices they were violating the principles they had agreed to without even knowing it. It has required constant working on it, explaining it, and training for it, even with convinced providers. In addition, we still had to help providers move forward with implementation in the form of getting a proposal together, finding the various pots of money, developing a project plan, etc. This included helping them understand how to use the various new funding sources and mechanisms that were being put in place.

The signs of system change in Portland/Multnomah County involve changes in ideas (coming to agreement on the right thing to do), changes in how money is used (the three-point funding structure with assistance to access each funding element), changes in power (new commitments of local elected officials and public agency heads), and changes in habits (new approaches to getting the right clients into the newly opened units).

Chicagos Conversion Process

The process of implementing a 10-year plan to end homelessness in Chicago also relied on major program-level change efforts. The Chicago Continuum of Care Governing Board adopted the plan in 2002, and the mayor endorsed it in early 2003 (Chicago Continuum of Care, n.d.). The plan required a complete paradigm shift in the ways that homeless programs operated and worked in relation to each other. The CoC developed detailed descriptions of the new program models that articulated expectations for program outcomes. Many homeless assistance agencies were active champions of the plan and embraced the concepts of change; however, they still needed significant technical assistance to shift from their current practices to new ways of delivering services. To help, the CoC developed a self-assessment tool that agencies could use to assess whether their programs were consistent with the plan. The tool could also be used to help agencies decide how they wanted to change and to develop a plan for implementing change at the board, staff, and client levels. The CoC also hosted many training workshops for each program type, which were intended to help staff acquire new skills and develop peer support networks to jointly navigate the process of change. Simultaneously, the city and CoC began a multi-year process of using the city and CoC-controlled grant resources to phase in change, starting with incentives and culminating in mandating compliance with the plan in order to access funding. For instance, within the first couple of years of plan implementation, the CoC reduced the number of shelter and transitional beds funded in order to support greater investment in prevention, permanent housing with short-term supports, and permanent supportive housing for people who are chronically disabled.

System change in Chicago is beginning to be recognized for changes in how money is spent (reallocation of city and Continuum resources to support the Plan), changes in ideas (a paradigm shift about the communitys ability to end homelessness), and changes in skills (retraining agency leadership and staff). The annual State of the Plan reports also document progress in building new permanent supportive housing units, among other process milestones.^[5]  Over time the expectation is that habits will also change (realization of cross-system service delivery) and that the cumulative impact of these changes will be realized in reduced numbers of people who experience homelessness, shorter durations of homelessness, and improved individual housing and behavioral outcomes.

Southern New England Training for Developer/Service Provider Teams

Recognizing that production goals for new permanent supportive housing would never be met without expanding the pool of housing developers and service providers who could create and run PSH, the Corporation for Supportive Housings Southern New England office used THCH funding to create a training program to bring together potential partners and help them structure new projects. The training sought to change knowledge, skills, and ideas of appropriate ways to work together. The One Step Beyond Training Institute (OSB) began in 2004. It gets to the nitty-gritty of what it takes to develop PSH by training the agencies and people who will actually have to produce and operate it. Inspiration is also a part of this mix, as new players must be convinced to participate in PSH production if the goal of expanded PSH capacity is to be reached. OSB is designed to foster partnerships among housing developers and service providers, so that more organizations will get into the PSH business and those already in it will expand their capacity to develop and operate PSH. Each plan being developed involves collaborations among several agencies. The goal was for teams to have project plans and sites identified by the end of the training.

During OSBs second year, teams from Rhode Island included nonprofit housing developers for the first time. Their presence was a testament to system change in two senses. Getting these new players involved in PSH had been a major goal of THCH in Rhode Island. But it probably would not have happened, even with urging from THCH, if another change had not come first. Rhode Island Housing, the agency that controls HOME dollars, established new priorities that for the first time gave precedence to PSH development. If the nonprofit housing developers wanted HOME dollars, they were going to have to get involved in producing PSH. A power center, Rhode Island Housing, had changed the financial incentives, and changed behavior followed. Many of the teams that participated in OSB have since submitted funding proposals to state agencies, with considerable success.

The curriculum developed for OSB is enjoying continued life, as various CSH local offices are using it to stimulate new partnerships around the country. In Los Angeles, for instance, a training series using a curriculum based on OSB but modified for local conditions, called Opening New Doors, is about to begin a second year with a new set of partner teams. Teams from the first year are already writing applications to fund the projects they developed during the training.

Matching Homeless Clients with Appropriate Housing and Services

When a community is sufficiently advanced in creating appropriate housing and service models to end homelessness, it may encounter the problem of assuring that clients with multiple barriers or disabilities, who are most likely to fall through the cracks, get into the available new slots. Some communities hire staff to place clients into programs efficiently and appropriately; others use cross-agency communication protocols or direct service staffing meetings to identify and place clients; others have developed technology to support client referrals and manage waitlists; still others employ a single point of entry to triage clients. Research has not been conducted to assess whether one model is more or less effective than another. In the meantime, we present several examples that appear to be effective to illustrate how communities working on system change have addressed this issue and changed their habitual ways of doing things into new and more effective habits. In a broader sense, this section also illustrates how system change efforts cannot be limited to big picture policymaking, but must also consider and resolve even the smallest details if they are to realize positive benefits for clients.

We start with Portland, because staff there were especially articulate about the work still needing to be done even after everyone officially accepted the goal of serving the longest-term homeless people. We follow this discussion with descriptions of targeting mechanisms in several other communities as well as discharge planning efforts that are working to prevent homelessness.

Portlands Housing-Client Match Facilitator

In 2005, Portland/Multnomah County was sufficiently far along in creating PSH to have come up against a level of system change that does not become obvious until PSH units become available. The housing units are available, with operating and service supports in place, and there are people who need this housing. But the agencies with the people are not the agencies with the housing, so there is still the issue of getting the people with the most complex and challenging conditions into the available units. The issue was recognized and well defined, which local informants perceived to be a good part of the battle. A position was created within the Department of Community Health Services (DCHS) to coordinate this client-level matchmaking and smooth the way with providers  a position that would not have been needed, possible, or realized without the explicit system change work that had been going on in Portland for the previous two years. The time was right for this development. THCH staff had done the matchmaking at the provider level, getting development, operations, and services providers together to create PSH units. But the last steps had yet to be taken. The agencies that know the clients often are not on pick up the phone when you need to and just call terms with the agencies that have the housing. That is where the new DCHS coordinator forges the necessary linkages. As further support for providers, Portland has changed recruitment and referral patterns, found new sources of support for landlords, and generated the trust of landlords by delivering on promised tenant supports. These strategies all work to ensure the hardest-to-serve people get the housing they need.

Philadelphias Placement Approach for Supportive Housing

Some years ago, Philadelphia faced a situation in which permanent housing providers were reluctant to take some of the hardest-to-serve homeless people who needed housing, and there was no central or coordinated way to match people with housing. One result was that long-term homeless people did not get housed as quickly as possible, and providers also had relatively high vacancy rates, approaching about 10 percent of existing beds. The citys Office of Adult Services, which has responsibility for homeless programs, responded by taking over placements. It began sending specific people to a provider and asking the provider to take them. The result has been more of the hardest-to-serve homeless people receiving housing and services, and more efficient use of available resources (vacancy rates are now around 1 or 2 percent, just enough to leave some placement opportunities when new clients need housing).

Approaches for Reducing Family Homelessness

The emergency shelter system for homeless families in Washington, D.C., has been revamped over the past couple of years to reflect a triage or targeted approach to matching families with appropriate housing and services. In the past, all families were treated similarly regardless of their needs. As a result the system was overcrowded and even the crisis shelter frequently had a waiting list. Today, all families experiencing a housing crisis are directed to the central intake facility, where they undergo an assessment. Based primarily on the nature of each familys housing crisis, intake workers have three primary ways to assist the family. If the family needs a place to stay immediately, it is referred to a central crisis shelter until space opens in a more service-intensive apartment-style emergency shelter program that can help the family find permanent housing and link it with appropriate services. If the family is able to remain in its current housing for a few days and is fairly high functioning, the family is referred to the Community Care Grant program, which provides flexible housing assistance and case management to quickly rehouse families or support them in their current housing. If the family can remain in its housing for up to 30 days, workers attempt to avert homelessness by providing ongoing mediation to resolve family disputes and housing search assistance. Homeless prevention funds are also available through a community-based program in each ward of the city. Changes in D.C.s homeless system are evident, reflecting a change in ideas (adopting the notion that family homelessness can be prevented), habits (old ways have been revamped through structured service delivery improvements), skills (staff are newly equipped to respond to families in different ways), and in the way money is spent (resources were reallocated to support a rapid rehousing approach).

Columbus, Ohio, uses a single point of entry coupled with careful screening and consideration of available prevention/diversion resources to determine which families can be helped to avoid homelessness and which need to enter a shelter. The system succeeds in helping about half the families who call to avoid shelter entry. Hennepin County, Minnesota, has a similar screening mechanism for controlling shelter entry and diverting families with relatively simple housing problems to a network of prevention agencies.

Other efforts currently being planned are even broader. For example, Massachusettss Department of Transitional Assistance (the state TANF agency), sponsored pilot projects several years ago to see whether a shallow rent subsidy offered to families facing housing crises would keep them from becoming homeless. The results (Friedman, 2006) were encouraging enough that the department strategized a statewide implementation; its future, however, depends on a new gubernatorial administration, epitomizing the fragility of even the most well-justified change efforts.

New York/New York III and Client Targeting

In the first two rounds of the New York/New York Initiative, which provides housing and supportive services to people with serious and persistent mental illness, providers had a lot of flexibility in choosing the people they would serve. New York/New York III, which began in 2006, sets specific population targets, including several groups of homeless people that providers have been somewhat reluctant to serve in the past. For the first time, New York Citys Department of Homeless Services is expecting to take control of the placement process, including developing lists of the neediest homeless people in each target group and offering only these people to service providers. It remains to be seen how successful this new approach will be. But as the legislation governing New York/New York III is very explicit about who must be served, and as the Department of Homeless Services will be the entity paying providers to serve the targeted clients, some accommodation that meets the needs of all parties is likely to be reached.

Approaches to Preventing Homelessness at Institutional Discharge

A California state program to alleviate or prevent homelessness among people with serious mental illness, known as AB 2034 after the Assembly Bill that sponsored it, is being used in Los Angeles to assure that people with mental illness leaving the county jail do not end up homeless (Burt et al., 2004; Burt and Anderson, 2005). Eighteen nonprofit community mental health agencies receive the funding and work with the county jail to identify at-risk prisoners shortly before their release. The AB 2034 money allows providers to do what it takes to keep clients from being homeless; the resources have mostly been used for supportive services, with the programs becoming skilled at finding housing resources through partnerships with other providers in the community and Shelter Plus Care vouchers designated for Department of Mental Health clients.

In Massachusetts, the Department of Mental Health has spent years promoting the attitude that housing is a clinical issue  a significant change in ideas from previous ways of thinking. It has developed a way to identify clients who were homeless when they entered institutional care and who are at risk of homelessness at exit, which it couples with an elaborate discharge planning mechanism. Recognizing that discharge planning will only succeed in averting homelessness if housing is available, the department established housing coordinators in each service area and in its central office to help develop suitable independent and semi-independent housing in the community (Burt, Pearson, & Montgomery, 2005).

Funding Mechanisms

Most programs that serve homeless people are funded by a complex array of sources, forcing service provider executive and development directors to spend far too much time pursuing each piece of the ever-changing funding puzzle. One of the most important signs of real system change is the easing of this patchwork funding burden. A few communities have simplified funding for all or most parts of their continuum of care, assembling all funding resources in one place and requiring providers to submit a single application that covers what they need by way of operating and services dollars (and capital dollars if relevant). Several other communities have accomplished a similar simplification for one component of their CoC  typically PSH  usually on an ongoing basis but sometimes as only a one-time effort. Exhibit 2 summarizes these arrangements in eight communities (based on research reported in Burt et al., 2004, and Burt & Anderson, 2006).

Funnel Mechanisms That Combine All Needed Funding Types in One Application

In 1986, public and private agencies and organizations in Columbus, Ohio, that were routinely approached to fund local homeless services were looking for a coherent way to structure their funding activities. They came together and created the nonprofit Community Shelter Board (CSB) to serve as the central planning, funding, and monitoring entity for homeless assistance programs in Columbus/Franklin County, and funneled all of their homeless-related funding through CSB. For about 10 years CSB presided over a system that gave homeless service providers the luxury of preparing only one application for all or most of their funding, but that did not seriously challenge the array of services the system was providing. In the late 1990s, downtown development plans sparked a concern about what would happen to homeless people and provided the impetus for self-study and ultimately for a paradigm shift in goals, from managing to ending homelessness. After due deliberation, the community launched the Rebuilding Lives initiative in 1998 to develop up to 800 units of permanent supportive housing for chronically homeless people (Burt et al., 2004). To identify and secure the resources needed for Rebuilding Lives, a Funders Collaborative was established, whose membership includes all the major public and private funders and potential funders in the area. Through the Collaborative, individual agencies pool their resources, establish common expectations about what outcomes are to be achieved, and specify what reporting requirements are needed to document progress. Armed with these resources, CSB funds individual projects that meet the goals and standards of the Collaborative. Providers apply for capital, operating, and services funding using one application, receive one grant, and write one report. This centralized funding mechanism is a powerful tool for enacting system change, since programs that do not conform to the new standards and way of doing business are not funded.

In Philadelphia, the Office of Adult Services orchestrates all homeless-related activities, coordinating with other key agencies in the process. The budget for emergency shelter is part of Adult Services, and a variety of public agencies (e.g., housing and community development, child welfare, and some mental health and substance abuse services) transfer funds to Adult Services to improve the integration of funding mechanisms and ease the proposal burden on providers. Adult Services also coordinates with mental health and substance abuse agencies that operate an array of community-based supportive housing as well as provide supportive services for homeless people in Philadelphia. The city also used the resources under its control to shift the emphasis of its investments from shelter to permanent supportive housing and outreach, in essence changing the allocation of money to follow the change in ideas on how best to end homelessness.

Starting in 1992, the State of Connecticut and the Corporation for Supportive Housing joined forces to promote the Connecticut Supportive Housing Demonstration Program, which ultimately produced 281 units of PSH in nine projects located in six mid-sized Connecticut cities. From the start the funding package combined capital, operating, and service dollars contributed by several state agencies and distributed the funds through a consolidated request for proposals. Recognizing the low probability of getting any more money until they could demonstrate to the legislature and state agencies that the first investment had paid off, CSH also raised money for an evaluation (Andersen et al., 2000). The evaluation showed that homeless people and people at very high risk of homelessness accepted this housing and remained stably housed for significant periods of time. Results of a public cost avoidance component of the study showed that tenants used fewer expensive crisis health services (mostly emergency room and medical inpatient services) and used more routine and appropriate health care such as home health and outpatient substance abuse treatment services. This switch from crisis health services to more preventive and routine care in clinic and office settings is one of the common goals of permanent supportive housing. Case managers help clients to attend to health problems earlier, before they become emergencies, which means that clients are able to use the more appropriate and less expensive clinic settings for health care. Because they were getting more routine and preventive care, tenants were also better able to avoid hospitalization. These results, which show both improved health outcomes and lower outlays for health care, have been parlayed into two additional rounds of state funding for PSH, now approaching about 1,000 units. Funding for each wave is ongoing, not one-time, as the resources to support projects are line items in state agency budgets. The Department of Mental Health and Addiction Services issues the request for proposals and funds operations and services from its own budget, which includes Shelter Plus Care resources. State housing finance and housing and economic development agencies provide capital resources that providers access through the single application process.

In summer 2006, the Seattle/King County Funders Group issued its first request for proposals to create supportive housing that combined capital, operating, and services funding. As the RFP says, This is the first countywide public funding effort in King County to coordinate the application and allocation process for capital, operating and services funding for proposals that meet the goals of the 10 Year Plan to End Homelessness. The Funders Group was a structure deliberately created to promote system change under the Taking Health Care Home initiative (Burt & Anderson, 2006).

What these four communities do on an ongoing basis, San Diego did once, in 2003. Several agencies, including the redevelopment authority, which supplied funding for capital and operating expenses and administered the grant-making process, pooled their resources through memoranda of understanding and issued a joint request for proposals for new permanent supportive housing projects.

Funding Mechanisms Involving Facilitated Access to Resources from Several Agencies

Several communities involved in system change studies have not gone as far as those described above in integrating their funding streams for the purpose of simplifying provider applications and assuring adequate levels of operating and services resources. They have, however, gained a commitment to fund from the agencies controlling the resources that are most essential for supportive housing and have created mechanisms to help providers navigate their way through these agencies funding processes. Portland, Oregon staff supported by THCH funding helped housing developers and service providers form viable projects, obtain capital resources from the housing and community development department and state resources (e.g., Low Income Housing Credit), operating resources (housing subsidies) from the public housing authority, and services funding from the mental health and substance abuse agency. In Maine, THCH staff facilitate meetings of a funders/coordinating group that has as one of its primary tasks finding the service match money for tenants of supportive housing projects that receive capital and operating resources from the state housing finance agency. And in Massachusetts, the Department of Mental Health routinely brokers resources for housing projects to support its homeless and at-risk clients, offering its own service resources to leverage housing dollars from a wide variety of sources including HUD, the state housing finance agency, numerous local public housing authorities, and the Massachusetts Department of Housing and Community Development (Burt, Pearson, & Montgomery, 2005).

Mechanisms That Integrate Funding for Clients

Resource management innovations can do for clients what funnel mechanisms do for providers  enable them to get the care they need with someone else worrying about how to match dollars to services. In their simplest form, resource management systems are being used to match available resources with clients who need them. The systems are used to track resources at the client level to ensure that clients needs are being met holistically and to ensure that the resources are managed efficiently and appropriately. One concept widely used in the childrens mental health field, system of care, (http://systemsofcare.samhsa.gov) is being adopted as part of the Indianapolis Blueprint to End Homelessness. A system of care assembles the resources to do what it takes from whatever system has relevant resources to meet client needs. This model involves two important paradigm shifts. The first is a recognition that agency silos do not meet client needs, as clients frequently fall through the cracks as they try to negotiate the mental health system to get mental health services, housing providers to get housing assistance, and so on. Instead, resources from each of these systems are pooled and managed by a resource coordinator to achieve the clients goals. The second important change is that, in contrast to the funding practices of most mainstream systems, funds are available up front rather than having to be claimed and justified after service delivery through a cost-reimbursement process. The community or collaborative of funders identifies the approximate annual or one-time level of resources that different subpopulations are likely to need, and the resource coordinator uses this pool of funding in discretionary ways to purchase services, pay for housing, and support client-identified activities.

This resource coordination model is consistent with the literature cited earlier regarding the ability of certain interventions to help mainstream and homeless assistance systems avoid unnecessary costs. The key to its success is a communitys ability to convince funders of its merits and to secure their commitment to participate in a system of care funding approach. Implementing this model would be a significant indication of system change, as it involves a change in power (change in control of expending resources), money (the act of pooling resources), habits (new ways of delivering services), technology/skills (new skills in working with clients to achieve goals), and ideas (breaking down the silos to deliver client-centered services). Resource management systems can also support dual purposes  direct service coordination and resource use documentation. A community could see the level of resources being used per client, how those resources vary or need to vary based on client characteristics and service requirements, and how the intervention (or involvement in the resource coordination model) changes the use of services. These integrative service delivery and funding systems can help a community understand and set resource allocation levels and measure whether application of funds in this way results in cost savings to other parts of the system.

Reallocation of Funds

The descriptions of Chicago, Columbus, and Philadelphia discuss how communities are using their resources to influence and leverage system change. Beginning with the 2005 CoC application, HUD provided a new tool, the Hold Harmless strategy, within the annual CoC application, to assist in this process. Communities can use the Hold Harmless provision to reallocate funds from poorly performing or lower priority projects to new permanent housing projects that target people who are chronically homelessness. This approach to system change is likely to increase in practice as other CoCs gain greater understanding of how to use this new tool.

Having Someone Whose Job Is System Change  The Coordinator Role

Through many studies and many site visits, the authors have identified the critical importance of having one or more people facilitating, coordinating, stimulating, reminding, organizing, assessing progress, bringing in new players, and keeping the many actors moving in the right general direction. As mentioned in the framework, creating this position represents a change in power to support system change efforts. The THCH project clearly demonstrates this finding (Burt & Anderson, 2006). THCH funds have supported these essential functions in every THCH site.^[6]  Key informants consistently stressed how vitally these functions have contributed to progress and the role and effects of coordination were obvious everywhere and at every level of system change observed. The basic phrase heard repeatedly was, it wouldnt have happened without [insert name of key THCH coordinator].^[7]  Without it, even a community with a dedicated council, committee, task force, or other mechanism that in theory could assume leadership runs up against the reality that committee members have other jobs to do. With the best will in the world, they cannot take on the coordinating function.

In all likelihood providing someone to mind the store is the key way that THCH has been able to have such a strong influence in many of its communities in such a short period of time  the grant pays the salary of someone whose job is to pay attention. The THCH evaluation also addressed the issue of where a coordinator should be located to be most effective. Some THCH site coordinators were employees of government agencies, while others operated from independent CSH offices, two of which were newly created for THCH. So the lever of change in some THCH communities was internal to government and thus subject to government changes in direction and policy, while in others it was external to government and had a primary and continuing mission to promote system change.

The decision to place the THCH project within or outside government was not random, which complicates analysis. Four sites had government agencies that were very ready for change and had also taken significant steps of their own toward investing in interventions to end long-term homelessness. These are the sites with coordinators internal to government, as there was an obvious governmental home eager to receive and support them. However, governments change, so it is especially telling to note what happened in the one or two THCH communities where the coordinator role was not as strongly realized, or not realized as quickly or at the highest levels. System change in these communities happened more slowly, or did not happen at all, because political or administrative changes (changes in power) occurred soon after THCH began, and hampered the coordinative function.

Two sites in which the THCH grant went to government agencies, Maine and Kentucky, began their grant period with their state housing finance agencies well positioned to involve other state agencies in expanded commitments to PSH development. A change of governors in Kentucky greatly reduced the potential coordinating function that THCH was able to play because agency priorities changed from ending homelessness to fostering recovery from substance abuse for people with housing. The Council on Homeless Policy, with its complement of state agency, provider, and advocate representatives, continued to meet, but operated mostly at the communication level, with some minimal coordinating activities. Each representative of a government agency operated in the context of his or her agency practices to facilitate PSH development. No one fulfilled a strong coordinator position urging new mechanisms, streamlined mechanisms, joint funding options, or changed policies and practices to stimulate even more PSH. Perhaps the time was not right in Kentucky for even the strongest coordinator or facilitator to pursue a PSH agenda with state agencies, and perhaps the results would have been the same whether someone was trying to fulfill this role from inside or outside of government. But the fact remains that without a strong coordinating influence the need for system change was not recognized or acted on.

The Maine State Housing Authority also had a commitment to PSH when THCH began, in a state that had already made significant commitments on paper to ending homelessness in the form of a statewide Action Plan. For various reasons unrelated to THCH, steps to endorse and then implement the Action Plan stalled. THCH stepped into these difficult circumstances; state housing finance agency staff proceeded to create an important multi-agency work group focused on PSH production. This group of mid-level government officials, working below the radar screen of agency heads but with their knowledge, made significant headway in moving projects toward realization through the commitment of new public resources (additional capital from the housing authority and Medicaid to pay for services from the Department of Human Services). When the state-level process began moving again and the new governor endorsed the Action Plan, THCH staff were in position to continue and expand their coordination activities. The governor also created a cabinet-level Director of Homeless Initiatives, making Maine the only state in the country to elevate the issue of homelessness to the cabinet level. This significant shift in power is leading to shifts in money and ideas.

The two remaining THCH sites with coordinators internal to government, Portland/Multnomah County and Seattle/King County, are prime examples of how far a person whose job is system change can move a system from a platform inside a government agency. Even when the system was ready to be moved, far less would have happened, in the opinion of key stakeholders, without the facilitation offered by the THCH coordinator. Having THCH money and someone in the coordinator position facilitated bringing everyone together, including politicians, agency heads, middle management, providers, and the clients in need of PSH units. With coordinators in place, these communities moved to establish one or more working groups. The groups had some common charges, including bringing more agencies to the table, finding more money for PSH, and smoothing the process of putting together PSH funding packages. In these communities, the agency responsible for mental health and substance abuse services was a primary target for inclusion, and both succeeded in bringing these very important agencies and their service-oriented resources on board. Law enforcement is also an important new partner in Portland.

Multi-agency groups in both communities have made great progress in identifying and committing public resources. These include completely new funding sources (e.g., Washingtons bill 2163), more funding and redirected funding from existing sources (e.g., use of state and local mental health dollars as service matches for PSH), and more streamlined funding mechanisms. They have also reduced bureaucratic entanglements that can slow the process of PSH development. Finally, these two THCH communities established new procedures for assuring that the hardest-to-serve long-term homeless adults were most likely to become tenants of new units.

The THCH sites in Los Angeles and Rhode Island were structured with coordinators in new CSH offices external to government because local government was not active in seeking solutions to homelessness. There was thus no obvious place to locate an internal coordinator. The external THCH coordinators initial goals in these communities were to educate relevant stakeholders about PSH and demonstrate to public agencies that PSH could help them fulfill their own agency objectives. Working from their nongovernmental platforms, THCH staff in both sites sought a foothold in the most relevant committees, councils, or task forces and proceeded from there. They were also able to capitalize on activities of their affiliated CSH offices (California and Southern New England) to help mobilize these new communities.

In Los Angeles THCH resources were used to staff the Special Needs Housing Alliance (a task force of county agencies charged with assessing and then augmenting available special needs housing, including housing for homeless people). This staffing provided coordination and technical assistance to help the Alliance articulate its agenda, complete a countywide inventory of special needs housing, develop a strategic plan, and see important components of that plan funded by the County Board of Supervisors, including a new position of housing and homelessness coordinator under the countys Chief Administrative Officer  several firsts for Los Angeles County that in turn are leading to new developments and partnerships. In Rhode Island, THCH intervention helped make PSH a recognizable concept to key stakeholders (change in ideas), leading to a new state agency with a housing and homelessness portfolio that gathers most of the states housing and homelessness-related funding streams into one coordinating office (change in power), staff to make it happen, a re-established interagency council, a partnership of philanthropy and government, and a first-ever public-private funding commitment for new units of PSH (change in money).

In the remaining THCH community, Connecticut, a good argument could be made that Connecticut had already achieved system change before THCH. But THCH staff in Connecticut see system change as an ongoing process and one that will always need some level of tending. Systems can always be improved, new agencies and populations brought in, service approaches expanded and made more effective, new provider teams created, prevention tackled, and real public understanding and commitment to ending homelessness secured. Connecticut used its THCH resources to many of these and other ends. It is the best example within THCH, so far, of what might be called a self-renewing system  one that regularly reflects on where it is and where it wants to be and keeps moving forward. As the nongovernmental entity whose eyes are always on the PSH prize, THCH and CSH in Connecticut still find significant roles in promoting the means to end homelessness for people with disabilities who are unlikely to be able to manage on their own.

The issue of the most effective location for a coordinator as change agent has no simple answer. Internally placed coordinators may be extremely effective in communities where at least some agencies and providers are ready for and interested in change. However, they are vulnerable to alterations in political support, and if support shifts substantially, their internal position may make it difficult for them to continue facilitating and advocating for system change. An externally placed coordinator may remain single-focused through all political changes, but has no official clout to wield in the process of gaining peoples attention and beginning to influence their choices. Further, an externally placed coordinator must have a home somewhere, so an external organization must create and sustain that home. To be most effective, the external organization should be seen as neutral or nonpartisan but politically savvy, able to contribute expert knowledge and technical assistance, respectful of all parties, and good at listening and facilitating.

Documenting the Impact of System Change

Throughout this paper we have noted the importance of regularly assessing the progress of system change and redirecting efforts, as needed, to fulfill the goals of the community. Forward-looking communities create mechanisms to measure their impact from the beginning, ideally building upon infrastructure that can also advance the change itself. This section documents some of the successful efforts that communities have used to establish regular processes for assessing progress on system change. The discussion will cover two primary areas: system infrastructure components and evaluation processes.

System Infrastructure Components

Many communities have established infrastructure to improve the delivery of services to clients. If well designed, this same infrastructure can be used to collect data for evaluation purposes. For instance, an information and referral (I&R) system is a valuable asset for sharing information on available services and criteria to access those resources with case managers and clients who may need them. Some I&R systems also automate the referral process to expedite client access to resources and to reduce service under-use or duplication. I&R systems can also inventory system assets to permit monitoring over time. If a community sets a goal to increase the number of prevention resources, mainstream supportive service linkages, or permanent supportive housing units, the I&R database becomes an objective way to measure progress toward that goal.

A homeless management information system (HMIS) that collects client-level data to enable coordinated case management also yields extremely valuable longitudinal information on the extent and nature of homelessness episodes, service use patterns, and short- and long-term client outcomes. The State of Arizonas Homeless Evaluation project exemplifies the value of HMIS for case management and evaluation purposes. Arizonas structure encompasses three continuums of care, all of which have functional HMIS implementations. The homeless providers within each continuum use the HMIS to support case management and internal agency record-keeping. Client information is aggregated and analyzed at the continuum level for each communitys planning purposes. The State of Arizona worked with the continuums to develop a Family Self-Sufficiency (FSS) matrix, which uses 13 domains to track a households change in self-sufficiency. The FSS matrix has been incorporated into each continuums HMIS, and case managers report on each of the 13 domains at program entry and exit, and sometimes more frequently. Case managers use the matrix during client assessment to develop a case plan for promoting greater family self-sufficiency. The Arizona Homeless Evaluation Project has begun to analyze the change in FSS results at the program and continuum levels and is using the initial findings to identify which programs are most successful with different client groups. Early results indicate the ability to predict client success in different program models from an initial client FSS assessment. Results are now being used to guide technical assistance, target appropriate client referrals, and develop baselines for program performance. Over time, the FSS measures will likely be integrated into an ongoing performance-based funding process.

In Philadelphia, the city has used HMIS data to understand client characteristics and patterns of shelter use (personal communication with Dennis Culhane and Rob Hess, Philadelphias homeless czar for many years). This information shaped policy decisions that fueled the dramatic strides in building permanent supportive housing and targeted interventions for individuals and families who are homeless. In addition, the HMIS system has become a day-to-day tool for improving services to homeless clients across disciplines (e.g., homeless programs, child welfare services, and behavioral health treatment). For instance, as interventions for chronically homeless people are developed, outreach staff can use the HMIS to identify specific individuals who have experienced long-term homelessness and would benefit most from permanent supportive housing. City staff also use daily statistics to monitor and immediately fill shelter vacancies, manage caseloads, and redeploy case managers to assessment centers with significant numbers of families waiting to be served, among other operational uses. In the aggregate, this information is also used to allocate annual city-controlled grants, benchmark progress on the citys 10-year plan to end homelessness, and inform homeless policy decisions.

As with I&R, the HMIS is fulfilling two important roles  one for direct service, another for evaluation  both important tools that support system change. HMIS presents opportunities for the future by building predictive models using longitudinal system data, which in turn can be used to triage clients the first time they present with a housing crisis and direct them to the programs and services most likely to be effective given their circumstances. Conversely, the data collected over time about these clients can be used to assess the effectiveness of and make improvements in the communitys interventions.

In most communities, only homeless providers and a handful of mainstream agencies participate in the HMIS. For HMIS to truly support and/or measure system change, the infrastructure will need to expand and achieve participation among providers from mainstream systems, or develop ways through data warehousing or other techniques to match and integrate data across systems.

Evaluation Processes

Communities have established a range of practices to measure progress and to influence further change, many of which rely on the infrastructure components described above. The primary methods include program-level evaluation, performance-monitoring and funding tied to performance, and benchmarking system progress. Many communities and providers note that having a process in place to measure their actions and results holds them more accountable, and therefore makes them work harder to be productive so they will be able to demonstrate results.

Program-Level Evaluation and Performance Monitoring and Accountability

Client records tracked in a longitudinal database, such as an HMIS, can be used systematically to understand program performance. The results can identify effective program practices or low-performing programs that need technical assistance to improve their performance. Of course, a communitys ability to use HMIS information in this way is only possible if the database contains fields for the relevant outcomes and if providers are diligent about collecting and entering the relevant information.

The results of performance monitoring can be used to direct clients to the programs that appear to be most successful for people with similar characteristics and issues. They can also be used to direct funding to successful programs and divert limited community resources from less successful efforts. Communities across the country are putting these and other strategies in place in as part of their efforts to reduce or end homelessness and to understand the processes of system change that can help them reach that ultimate goal. The earlier discussion of Philadelphias approach provides one example, in addition to those below.

Columbus has implemented advanced processes for analyzing program effectiveness with HMIS data and uses the results to influence program funding. To support its process, the Community Shelter Board (CSB) has developed data quality assurance standards for all funded agencies to ensure that the community has reliable, complete data on which to base decisions. CSB has also developed comprehensive program standards and performance expectations for each program type, and incorporates these expectations into each agencys contract. Performance measures pertain to the number of clients served, average length of stay, housing and income outcomes at exit, return to homelessness, client movement toward agreed-upon goals, direct client assistance utilization, occupancy rates, housing stability and retention, and efficient use of funding resources (average per client costs). CSB clearly communicated to agencies that these outcomes were a priority in the homeless system. It identified measures to support performance-based funding and put them in place throughout the system of homeless assistance services. Over the years, tracking performance has helped to fuel program-level change to support these goals.

The Michigan Measurement Project was established in 2006 to develop a sophisticated program outcomes measurement system to track intermediate and long-term client outcomes by program type. Once implemented through the statewide HMIS, aggregate performance data will be viewable through the HMIS at the program, agency, CoC, and statewide levels. Programs can compare their own results to those of other like programs throughout the state to assess their own effectiveness and to identify strategies for improvement. CoCs and state agencies can use the outcome measures to inform community planning and resource allocation processes. Because it is a state-level activity, the Michigan Measurement Project is more focused on identifying promising program practices and benchmarking program progress than local evaluation processes might be. The latter might focus more on performance monitoring and performance-based funding.

The model of measuring system change at the program level is also being carried to the federal level (Khadduri, 2005). HUD is presently reconfiguring its Annual Progress Report to improve the quality of information it receives from continuum-of-care grantees on the activities and outcomes of HUD-funded programs in relation to its national goal and other related objectives. To reflect the diverse goals of homeless assistance programs, particularly in light of local system change efforts, HUD is contemplating establishing a range of performance measures from which funded programs could select when they apply for HUD funding. HUD would then use the selected measures to monitor each grantees accomplishments. HUD can integrate the program-level results with two other primary sources to measure national progress toward the goal of ending chronic homelessness. First, HUD is analyzing HMIS data from a sample of communities across the country to produce Annual Homeless Assessment Reports on the extent and nature of homelessness and use of homeless services. Second, HUD asks CoCs to report numbers of sheltered and unsheltered persons in their annual CoC applications. Together, these sources help HUD understand better how homelessness is changing over time and suggest how programs funded through CoC Care grants are contributing to these changes.

System-Level Evaluation

Communities are also instituting processes to track their own effectiveness in achieving system change for the ultimate purpose of ending homelessness. Establishing an evaluation framework forces a community to set deliberate change goals and to identify strategies to accomplish them, from which it can easily document progress related to its multi-year action plan. As noted earlier, evaluating the process may actually promote progress itself. The change goals and strategies become the outcomes and indicators of the effects of changing systems that are tracked over time. A comprehensive evaluation framework is likely to include process outcomes (what indicators will help the community know if it has completed the strategies that it anticipates are needed to effect system change), program outcomes that can be used to guide a program-level evaluation and performance-based funding process (what client outcomes are needed at the program level in order to achieve the system outcomes), and system outcomes (the domains that the community hopes will be impacted as a result of the efforts). If all of these components are in place, a community will be able to observe whether its intended system changes have occurred and if they are making an impact. If the expected effects are not observed, the community will also have data to help indicate whether it failed to meet its outcomes because plans were inadequately carried out, or whether the strategies the community thought would help were insufficient. The information can be used to improve their process over time, and the community can share effective strategies and/or pitfalls with others trying to accomplish similar things.

The Chicago continuum of care developed a series of system measures to assess progress in ending homelessness, including indicators of the number of people (overall and chronically homeless) who present for homeless assistance each year, the number of days it takes to help someone presenting with a housing crisis return to permanent housing placement, rates of permanent housing retention, and rates of recidivism. Program outcomes have been defined for each program type to help set common expectations for what is expected from each part of the system and to guide resource allocation. The CoC also developed process and efficiency measures, such as annual projections of units and services slots by year (some program types will increase, some will decrease over time), the vacancy rates of residential programs, user satisfaction rates, and indicators related to increasing resources for appropriate interventions. Chicago has been publishing semi-annual State of the Plan reports (available at www.cityofchicgao.org) that share accomplishments and annual outputs related to unit conversion and development targets as well as performance on the system measures.

Various formal methods exist to evaluate changes in system connectedness and integration, including changes in how people and agencies relate to each other, changes in how clients are referred among agencies, and changes in how funding does or does not flow among agencies. In the homeless arena, some of these methods were used in evaluating the Program on Chronic Mental Illness that began in the mid-1980s with support from the Robert Wood Johnson Foundation and HUD, and also the ACCESS demonstrations of the early 1990s (Morrissey et al., 1994, 2002; Rosenheck et al., 1998, 2001, 2002, 2003b). Modifications of the same methods are now being used to assess changes in system integration associated with the Collaborative Initiative on Chronic Homelessness (Greenberg & Rosenheck, 2006). These methods are available for use by communities seeking formal quantitative documentation of system change.

A community or researcher could also apply one or both frameworks presented in the beginning of this paper to measure process aspects of system change: the five indicators of change from Laying a New Foundation, and the five stages of integration. Laying a New Foundation indicators of change in money, ideas, habits, power, and skills/technology can be measured as process indicators. For instance, a community that sets a goal to shift system resources from shelters to permanent housing could track the percentage of resources going to each component of its homeless assistance system over time. Alternatively, a community could track the number of units/service slots in each program area over time. Changes in skills and habits could be measured by tracking first whether staff have acquired the skills and tools they need to conduct business differently, and second whether the business practices have actually changed. For instance, are people being placed within 14 days or are staff continuing to work with clients using the old patterns of doing business? These measures need to be constructed locally depending on the ways in which the local systems need to change; however, the framework can provide useful categories to classify and set expectations for change.

The stages of integration and changes in money, ideas, habits, skills, and power are or should be interactive. That is, as a community progresses toward more collaboration and coordinated community response, one expects these changes in process to produce changes in money, power, ideas, etc. But it also works the other way, as changes in money, power, and so on can push systems to change more and cause more stakeholders to join the system change bandwagon. In addition to the usual approach involving qualitative methods to assess changes in stages of services and systems integration, the formal methods referenced above can be used to document changes in the flow of clients, ideas, and money. Some changes, such as changes in money or new housing units, are relatively easy to document (e.g., Burt & Anderson 2005, 2006).

Communities could commit themselves to assessing the extent of change in the level of integration across all the agencies in a whole system as well as between any two agencies or among any three or more agencies within a particular agency or system. Also relevant are the isolated or integrated activities of various divisions within large umbrella public agencies. For example, many Departments of Human Services include divisions responsible for income maintenance (TANF, food stamps), child care, child welfare, and sometimes mental health and homeless assistance. It is all too common that these divisions do not work together at all, despite all existing under a common roof.

One can use the framework to prescribe, as well as describe, changes in various systems that are expected to promote the goal of ending homelessness. A community can assess its initial or starting point of integration and set explicit goals as to where, along the continuum of integration stages, the systems should be in one year, two years, five years, and so on. In doing so, it is essential to realize that it would be very unusual for all the agencies and organizations in a community to be at the same place at the same time in this five-level framework. Rather, what usually happens is that some parts of a potential system begin moving toward increased integration early on and, if their progress is viewed as useful, begin to bring other elements on board and to expand the scope of activities. It is therefore most fruitful to use this framework to characterize movement and change rather than a steady state or a comprehensive overview. A community does not need all potentially relevant agencies on board at the start, but as ending homelessness will not happen without the substantial commitment of public resources, at least some major public agencies need to be committed at the outset for a community to be able to speak realistically about embarking on a campaign to end homelessness.

An illustrative example for a community deciding to work toward ending family homelessness is shown in Exhibit 3. The matrix in Exhibit 3 should really be three-dimensional, as the issue with integration is how one agency or system works with one or more other agencies or systems. This multi-agency nature of integration is reflected in the last column of Exhibit 3, where the coordinated community response cell is shown as including participation of six different systems.

Similar assessments of the current stage of integration can occur at subsequent time intervals, and the community can use the results to help assess whether the past actions were effective in meeting change goals. This type of evaluation does not replace the need to assess whether an increasingly integrated system is reducing homelessness or positively affecting the problem that the community is targeting. Both the formative (process) and summative (end result) measures are required for communities to understand which strategies are most effective in meeting identified needs.

Some communities may also find it useful to think of how they might modify the stages of change framework used in the substance abuse treatment field for use in documenting community change. The stages of change framework classifies a persons willingness and progress in addressing a substance addiction. Such an approach recognizes the importance of assessing where an individual is relative to his/her understanding of a need to change and his/her commitment to the change process and identifies a predictable sequence of stages through which an individual passes during the process: precontemplation, contemplation, determination, action, maintenance, and relapse prevention (Prochaska et al, 1994). Programs that use a stages of change approach assess and respect where an individual is in that process and encourage and motivate movement. The strategies and actions that a treatment professional might use to engage an individual in treatment or the change process will vary depending on a determination of the persons current stage of acknowledgement and engagement. The stages in the addictions scheme are used over time to benchmark progress and regression. If one is willing to think of a whole community in terms of its acknowledgement of the need to change and its willingness to take steps to change, this model might be applicable. It would certainly be applicable to measuring changes in individual homeless people as they move toward leaving homelessness.

Implications for Preventing and Ending Homelessness

Research on the impacts of system change itself is relatively rare, compared to research on the effectiveness of particular program models to serve particular populations (e.g., PSH to serve long-term homeless people with disabilities). Although we are unable to speak definitively about the impact of system change, many communities are able to report process measures and some impact information; therefore, we can say the following:

• Most works cited in this paper attest to the fact that explicit system change efforts can get previously uninvolved agencies to the table and involved in developing more effective approaches to serving homeless people and ending their homelessness.
• Process results, with relevant outcomes: System change efforts can succeed in increasing funding for and production of supportive housing. 
  • Most THCH sites stimulated significant new funding, and also brought new stakeholders to the table and strengthened and integrated the involvement of the original stakeholders (Burt & Anderson, 2006). Outcome: Added more PSH units to the pipeline in the first two years of their grants than they had expected to do in five years, and ended the homelessness of a corresponding number of chronically homeless people.
  • San Franciscos Direct Access to Housing (DAH) approach grew out of a pipeline group of relevant agencies. Outcome: Added thousands of units of PSH to the San Francisco portfolio of programs to end homelessness (Corporation for Supportive Housing, 2004), with corresponding reductions in street homelessness.
  • A confluence of events and history put Philadelphia in a position to develop a great deal of PSH rapidly (Wong et al., 2006) and couple it with a redesigned outreach system to help move street homeless people into the new housing. Outcome: Major reductions in street homelessness.
  • Columbus, Ohios Rebuilding Lives Initiative changed the local homeless assistance system toward one designed to end homelessness. Outcome: Created more than 200 short-term shelter beds and upwards of 600 new PSH units (so far) with a combination of new and redirected funding. Almost 800 single adults have secured housing and left homelessness (www.csb.org). Other parts of the system concentrated on prevention.
• Impact results: Communities that have invested in permanent supportive housing on a significant scale are beginning to see the effects in reduced counts of unsheltered homeless people; likewise, there is some evidence that communities that have instituted new, integrated ways to address family homelessness have seen reductions in family shelter use, because housing crises are being resolved before they progress to the stage in which a family becomes homeless.
  • Communities that have invested in permanent supportive housing are reporting reductions in street homelessness  San Francisco, down about 20 percent between 2004 and 2006; Portland, down 20 percent and 600 people moved into PSH in past two years; New York City down 13 percent from 2005 to 2006 (all as described in a federal Interagency Council on Homelessness electronic newsletter, available at www.ich.gov); Philadelphia down more than 75 percent over five or six years (as described in Burt et al., 2004). Outreach and other mechanisms deliberately focused on bringing street people into housing can help this process.
  • Integrated services that include housing can increase access to housing and successful housing outcomes for homeless people with serious mental illness (Mayberg, 2003; Rosenheck et al., 1998). Further, the effects last for some years (Burt & Anderson, 2005; Rothbard et al., 2004).
  • Communities such as Hennepin County, Minnesota; Washington, D.C.; and Columbus, Ohio, that have focused on strategies for shelter diversion can dramatically reduce the numbers of people entering shelter. Strategies in those communities to reduce lengths of stay in shelter have enabled them to reduce shelter beds and apply those resources to other housing and services.

The knowledge gained from years in the system change trenches is being applied widely, thanks to active promotion through advocacy and technical assistance (Center for Mental Health Services 2003; Corporation for Supportive Housing 2002, 2004, 2005; and many others). But systematic evaluation of these change efforts in the homelessness arena remains all too rare.

Recommendations for Future Research

Given the paucity of current documentation of system change and its effects, we cannot reliably identify gaps in knowledge; instead, we must point to a broad array of important questions for which we have mostly anecdotal answers. With more than 300 communities around the country developing 10-year plans to end homelessness (see Interagency Council on Homelessness Web site (www.ich.gov), and at least 90 of them promulgating those plans and taking some steps toward implementing them (Cunningham, et al., 2006), and independent efforts to change systems in many locations, there is great need to evaluate the impact of these efforts and the factors that were most important in shaping (or blocking) that impact. We know of no plans to do so, beyond the self-assessments that are likely to show up in annual applications for continuum-of-care funding in response to HUD requirements. It would indeed be a great shame if, at the third National Symposium on Homelessness Research 10 years from now, there is no more systematic research evidence for the impacts of system change than we have been able to report here.

This is an area where practitioners and advisors abound, but hard evidence is elusive. In this paper we have tried to lay out some basic answers to a preliminary question: How will we know that systems have changed? We must go on from there, to design and fund research that answers the following questions:

• Which actions were pivotal in achieving change?
• What level of effort, staffing, and political will is required to implement change?
• How will we know we are making progress toward ending homelessness, and will we be able to say with confidence that systems change contributed to any observed reductions?
• Are some systems more important to change than others in community efforts to end homelessness?
• How much of any change we observe will we be able to attribute to the introduction of more effective program approaches (e.g., adoption of best practices), compared to streamlining system processes?
• How much of any change we observe will we be able to attribute to the structure of the change effort? As one reviewer noted, the communities currently organizing for ending homelessness or chronic homelessness are using very different structures. The motive for seeking an answer to this question would be to help communities just starting on the process to select the most effective structure.^[8]  The suggested structures include:
  • A centralized, coordinated, state-to-local model (Utah),
  • A two-tier framework that aligns state with locals to implement state framework (Michigan, Missouri, North Carolina),
  • Integrated CoC and 10-year plan efforts (District of Columbia, Contra Costa County, California) that plan to implement across multiple populations all at once, and
  • An incubator approach (Montana).

Individual communities can and should implement formative and summative evaluations using the frameworks and methods described in the Evaluation Processes section of this paper. In addition, a systematic system change-oriented multisite, multi-year research project similar to the ACCESS evaluation, set up to measure a spectrum of change processes and their impacts over time, could be very fruitful.

References

Arthur Andersen, LLP, University of Pennsylvania Health System, Sherwood, K. E., & TWP Consulting. (2000.) Connecticut Supportive Housing Demonstration Program: Final program evaluation report. New Haven, CT: Corporation for Supportive Housing.

Borman, G. D., Hewes, G. M., Toverman, L., & Brown, S. (2003). Comprehensive school reform and achievement: A meta-analysis. Review of Educational Research, 73, 125230.

Burt, M. R., & Anderson, J. (2005). AB 2034 Program experiences in housing homeless people with serious mental illness. Oakland, CA: Corporation for Supportive Housing.

Burt, M. R., & Anderson, J. (2006). Taking health care home: Impact of system change efforts at the two-year mark. Oakland, CA: Corporation for Supportive Housing.

Burt, M. R., Aron, L. Y., & Lee, E. (2001). Helping Americas homeless: Emergency shelter or affordable housing? Washington, DC: The Urban Institute.

Burt, M. R., Hedderson, J., Zweig, J. M., Ortiz, M. J., Aron-Turnham, A. Y., & Johnson, S. (2004). Strategies for reducing chronic street homelessness. Washington, DC: U.S. Department of Housing and Urban Development.

Burt, M. R., Pearson, C., & Montgomery, A. E. (2005). Strategies for preventing homelessness. Washington, DC: U.S. Department of Housing and Urban Development.

Burt, M. R., Resnick, G., & Novick, E. R. (1998). Building supportive communities for at-risk adolescents: It takes more than services. Washington, DC: APA Books.

Burt, M. R., Pollack, D., Sosland, A., Mikelson, K. S., Drapa, E., Greenwalt, K., & Sharkey, P. (2002). Evaluation of continuums of care for homeless people. Washington, DC: Department of Housing and Urban Development.

Burt, M. R., Zweig, J. M., Schlichter, K., Kamya, S., Katz, B. L., Miller, N., et al. (2000). 2000 report: Evaluation of the STOP formula grants to combat violence against women under the Violence Against Women Act of 1994. Washington, DC: The Urban Institute.

Center for Mental Health Services. (2003). Blueprint for change: Ending chronic homelessness for persons with serious mental illnesses and/or co-occurring substance use disorders. Rockville, MD: Author.

Chicago Continuum of Care. (n.d.). Getting housed, staying housed: A ten year plan to end homelessness in Chicago. Retrieved February 6, 2007, from www.chicagocontinuum.org

Clark, S. J., Burt, M. R., Schulte, M. M., & Maguire, K. E. (1996). Coordinated community responses to domestic violence in six communities. Washington, DC: The Urban Institute.

Cocozza, J. J., Steadman, H. J., Dennis, D. L., Blasinsky, M., Randolph, F. L., Johnsen, M., & Goldman, H. H. (2000). Successful systems integration strategies: The ACCESS program for persons who are homeless and mentally ill. Administration and Policy in Mental Health, 27, 395407.

Corporation for Supportive Housing. (2005). How public leaders change multiple systems: Reducing costs and improving outcomes through supportive housing. New York: Author.

Corporation for Supportive Housing. (2004). Local financing of supportive housing in San Francisco. Oakland, CA: Author.

Corporation for Supportive Housing. (2004). Toolkit for ending long-term homelessness. New York: Author.

Corporation for Supportive Housing. (2002). Strategic framework for ending long-term homelessness. New York: Author.

Culhane, D. P., Metraux, S., & Hadley, T. (2002). Public service reductions associated with placement of homeless persons with severe mental illness in supportive housing. Housing Policy Debate, 13, 107163.

Cunningham, M., Lear, M., Schmitt, E., & Henry, M. (2006). A new vision: What is in community plans to end homelessness? Washington, DC: National Alliance to End Homelessness. Retrieved December 17, 2006, from www.endhomelessness.org/content/article/detail/1397

Dasinger, L. & Spieglman, R. (2006, October 2021). Homelessness prevention: The effect of a shallow rent subsidy program on housing outcomes for people with HIV and AIDS. Paper presented at the NAHC National Housing andHIV/AIDS Research Summit, Baltimore, MD.

Dennis, D. L., Cocozza, J. J., & Steadman, H. J. (1999). What do we know about systems integration and homelessness? In L. B. Fosberg & D. L. Dennis (Eds.), Practical lessons: The 1998 National Symposium on Homelessness Research. Washington, DC: U.S. Department of Housing and Urban Development and U.S. Department of Health and Human Services.

Friedman, D. H. (2006). Prevention at work: The Homelessness Prevention Initiative, interim evaluation report. Boston: University of Massachusetts Boston, The Boston Foundation and the Center for Social Policy at the McCormack School.

Greenberg, G. A., & Rosenheck, R. A. (2006). An evaluation of an initiative to improve coordination and service delivery of homeless services networks. Preliminary Report: Draft 2. West Haven, CT: NorthEast Program Evaluation Center.

Grieff, D., Proscio, T., & Wilkins, C. (2003). Laying a new foundation: Changing the systems that create and sustain supportive housing. Oakland, CA: Corporation for Supportive Housing.

Huz, S., Richardson, G. P., Andersen, D. F., & Boothroyd, R. A. (1997). A framework for evaluating systems thinking interventions: An experimental approach to mental health system change. System Dynamics Review, 13, 149169.

Indianapolis Housing Task Force. (2002). Blueprint to end homelessness. Retrieved February 6, 2007, from http://www.chipindy.org/pdf/ReaderFormatAll.pdf

Interagency Council on Homelessness. (n.d.) City and county 10-year plan update. Retrieved February 11, 2007, from http://www.ich.gov/slocal/plans/10-year-plan-communities.pdf

Khadduri, J. (2005). Measuring the performance of programs that serve homeless people. Cambridge, MA: Abt Associates Inc.

Koegel, P., Sullivan, G., Jackson, C., Morton, S. C., Jinnett, K. J., Miu, A., & Chien, S. (2004, November). Public sector service use among homeless adults: A cross-system perspective. Presented at the Annual Meeting of the American Public Health Association, Washington, DC.

Kuhn, R., & Culhane, D. P. (1998). Applying cluster analysis to test a typology of homelessness by pattern of shelter utilization: Results from the analysis of administrative data. American Journal of Community Psychology, 26, 207232.

Markowitz, F. (2006). Psychiatric hospital capacity, homelessness, and crime and arrest rates. Criminology, 44, 4572.

Martinson, K., & Holcomb, P. A. (2002). Reforming welfare: Institutional change and challenges. Assessing the new federalism (Occasional Paper No. 60). Washington, DC: The Urban Institute.

Mayberg, S. W. (2003). Effectiveness of integrated services for homeless adults with serious mental illness: A Report to the legislature as required by Division 5, Section 5814, of California Welfare and Institutions Code. Sacramento, CA: California Department of Mental Health.

Melaville, A. I., & Blank, M. J. (1991). What it takes: Structuring interagency partnerships to connect children and families with comprehensive services. Washington, DC: Education and Human Services Consortium.

Minnesota Department of Human Services, Minnesota Department of Corrections, Minnesota Housing Finance Agency. (2004). Ending long-term homelessness in Minnesota: Report and business plan of the Working Group on Long-Term Homelessness. Retrieved December 17, 2006, from www.mfha.state.mn.us/about/homelessbusinessplan.pdf

Morrissey J. P., Calloway, M. O., Thakur, N., Cocozza, J. J., Steadman, H. J., & Dennis, D. (2002). ACCESS National Evaluation Team. Integration of service systems for homeless persons with serious mental illness through the ACCESS program. Access to community care and effective services and supports. Psychiatric Services, 53, 94957.

Morrissey, J. P., Calloway, M., Bartko, W. T., Ridgely, M. S., Goldman, H. H., & Paulson, R. I. (1994). Local mental health authorities and service system change: Evidence from the Robert Wood Johnson Foundation Program on Chronic Mental Illness. Milbank Quarterly, 72, 4980.

National Alliance to End Homeless. (2000). A plan, not a dream: How to end homelessness in ten years. Washington, DC: Author.

Nelson, D. W. (1995). The path of most resistance. Baltimore, MD: The Annie E. Casey Foundation.

Pindus, N., Koralek, R., Martinson, K., & Trutko, J. (2000). Coordination and integration of welfare and workforce development systems. Washington, DC: The Urban Institute.

Prochaska, J. O., Velicer, W. F., Rossi J. S., Goldstein, M. G., Marcus, B. H., Rakowski, W., et al. (1994). Stages of change and decisional balance for 12 problem behaviors. Health Psychology, 13, 3946.

Proscio, T. (1997). Forming an effective supportive housing consortium. Oakland, CA: Corporation for Supportive Housing.

Provan, K. G., & Milward, B. (1995). A preliminary theory of interorganizational network effectiveness: A comparative study of four community mental health systems. Administrative Science Quarterly, 40, 133.

Quigley, J. M., Raphael, S., & Smolensky E. (2001). The links between income inequality, housing markets, and homelessness in California. (Research Brief). San Francisco: Public Policy Institute of California.

Randolph, F., Blasinsky, M., Morrissey, J. P., Rosenheck, R. A., Cocozza, J., Goldman, H. H., & the ACCESS National Evaluation Team. (2002). Overview of the ACCESS program. Psychiatric Services, 53, 945948.

Rosenheck, R., Kasprow, W., Frisman, L., & Mares, W-L. (2003a). Cost-effectiveness of supportive housing for homeless persons with mental illness. Archives of General Psychiatry, 60, 940951.

Rosenheck, R. A., Resnick, S. G., & Morrissey, J. P. (2003b). Closing service system gaps for homeless clients with a dual diagnosis: Integrated teams and interagency cooperation. Journal of Mental Health Policy and Economics, 6, 7787.

Rosenheck, R. A., Lam, J., Morrissey, J. P., Calloway, M. O., Stolar, M., Randolph, F., et al. (2002). Service systems integration and outcomes for mentally ill homeless persons in the ACCESS program. Access to Community Care and Effective Services and Supports. Evidence from the ACCESS Program. Psychiatric Services, 53, 958966

Rosenheck, R. A., Morrissey, J. P., Lam, J., Calloway, M., Stolar, M., Johnsen, M., et al. (2001). Service delivery and community: social capital, service systems integration, and outcomes among homeless persons with severe mental illness. Health Services Research, 36, 691709.

Rosenheck, R. A., Morrissey, J. P., Lam, J., Calloway, M., Johnsen, M., Goldman, H. H., et al. (1998). Service system integration, access to services, and housing outcomes in a program for homeless persons with severe mental illness. American Journal of Public Health, 88, 16101615.

Rothbard, A. B., Min, S-Y., Kuno, E., & Wong, Y-L. I. (2004). Long-term effectiveness of the ACCESS program in linking community mental health services to homeless persons with serious mental illness. Journal of Behavioral Health Services and Research, 31, 441449.

Shern, D. L., Felton, C. J., Hough, R. L., Lehman, A. F., Goldfinger, S. M., Valencia, E., et al. (1997). Housing outcomes for homeless adults with mental illness: Results from the second-round McKinney Program. Psychiatric Services, 48, 239241.

Tsemberis, S., & Eisenberg, R. F. (2000). Pathways to housing: Supported housing for street-dwelling homeless individuals with psychiatric disabilities. Psychiatric Services, 51, 487493.

Wong, Y-L. I., Hadley, T. R., Culhane, D. P., Poulin, S. R., Davis, M. R., Cirksey, B. A., & Brown, J. L. (2006). Predicting staying in or leaving permanent supportive housing that serves homeless people with mental illness. Washington, DC: U.S. Department of Housing and Urban Development.

Endnotes

[1]     To be included, a community had to be taking significant steps to end long-term street homeless and also had to have at least some data to prove that these efforts were making a difference. It was not easy to find communities that met both criteria.

[2]     The communities involved were four states (Maine, Connecticut, Rhode Island, and Kentucky), two city/county sites (Los Angeles and Portland/Multnomah County, Oregon), and one multi-jurisdiction site (Seattle/King County, Spokane City and County, and the state of Washington).

[3]     A fourth THCH community had these same characteristics initially, but a change in state leadership and direction reduced interest in PSH development and stiffened resistance to system change.

[4]     Jacksonville 2005 Grant Application to HUDs Housing for People Who Are Homeless and Addicted to Alcohol program.

[5]     State of the Plan reports and other relevant documents about the plan and its implementation can be downloaded from the Chicago Department of Housing Web site (www.cityofchicago.org. Click on Departments, then Housing, then "There's No Place Like a Home" section.)

[6]     We discuss THCH findings on the value of the coordinator role because THCH focuses on homeless issues. Dennis, Cocozza, & Steadman (1999) reviewed earlier evidence for the coordinators importance in creating integrated service systems, focusing mostly on ACCESS-related research. Further published research on ACCESS provides additional documentation (Rosenheck et al., 1998, 2001, 2003a and 2003b). Proscio (1997) discusses the coordinators importance in developing an effective supportive housing development alliance, as do Greiff, Proscio, & Wilkins (2003). In addition, the experience of many other systems confirms this importance. See, for example, Burt, Resnick, & Novick (1998) with respect to services integration for high-risk youth, Burt et al. (2000) and Clark et al. (1996) with respect to serving women victims of violence, and Center for Mental Health Services (2003) and Huz et al. (1997) with respect to children and family services.

[7]     Alternatively, guidance materials stress that participating agencies must give their staff time for the planning and coordinative work of system change, often advising that the key person be given at least half time to devote to system change activities (e.g., Proscio, 1997).

[8]     The problem with attempting to submit different structures to an impact evaluation is that each is a unique product of its place, time, and circumstances. No community would ever be able to adopt a particular structure in its entirety, let alone expect it to function as it did in the community from which it was copied. These structures develop in response to the particular attitudes, histories, and cultures of their respective communities, as well as reflecting who cared the most and the agencies or organizations in which those people were situated when the need for mobilization occurred. Another community facing different conditions could not simply decide to adopt a particular structure; it would have to select the one with the best fit, and then modify it until the fit meets community needs and capacities. It would be important to examine whether the structure itself affects what can be accomplished, or whether the strength or weakness of the structure reflects the willingness of local actors to submit themselves to its direction. It would also be important to examine how much power the structure is given and how it wields that power. Further, it would be important to examine the existence and scope of a coordinator role, and whether a clever and dedicated coordinator can make any structure work, albeit using different approaches depending on whether the structure is strong or weak. A good research design would include approaches for sorting out which factors are most important for system change and its ultimate effects, and what is different but does not make a difference.