Integrating service delivery in a regional homelessness service system.

Neale, Kate ; Buultjens, Jeremy ; Evans, Trish 等

Introduction

Current government policy approaches to homelessness place great emphasis on the assumption that integrating services would make them more effective in responding to homelessness. For example, the Australian Government's White Paper on homelessness entitled The Road Home: A National Approach to Reducing Homelessness (Australian Government 2009) identified service integration as one of three key interventions needed to reduce homelessness successfully. The White Paper also suggested that, although regional and rural areas often have a strong history of integrated services, there was still scope for improved service integration within these areas (Australian Government 2009). State and Territory governments also appear to be focused on improving service delivery through implicit service integration objectives (Council of Australian Governments 2009).

This focus on service integration finds some support in existing research on homelessness (see, for example, Rosenheck at al. 2001). However, there is little research on homelessness outside metropolitan areas (see Fine et al. 2000; Beer et al. 2005). Existing service integration strategies in Australia tend to be conceptualised within an urban setting, characterised by a 'one-size-fits all' approach that requires significant resources and a concentration of services within a small geographical area--characteristics which do not always exist in a regional setting (Beer et al. 2005).

Ensuring appropriate integration strategies for non-metropolitan areas of Australia is important since many have a substantial homelessness problem. Developing appropriate non-metropolitan strategies requires an understanding of opportunities, limitations and capacity for service integration within each regional setting. Factors such as the impacts of geographical dispersion, the existence of fewer agencies, more use of part-time agency staff and a lack of funded co-ordination positions all need to be considered (Beer et al. 2005).

This paper provides new information on service integration to address homelessness in a regional setting through a study set up to determine appropriate regional strategies in the Northern Rivers region of New South Wales, Australia. The paper describes the process through which the strategies were identified in order to provide a possible model for other regions. The paper begins with a brief description of the literature on homelessness and service integration, followed by an outline of the research methods used and a description of the case study area.

Defining homelessness

There is a general consensus within the literature that being homeless is more than just being without a home (Chamberlain & MacKenzie 2008). As described by Flatau and colleagues (2010), homelessness can be characterised by marginalisation, social exclusion and a lack of opportunity for meaningful activity. Chamberlain and MacKenzie (2008) developed a 'cultural' definition for homelessness which recognises the concept of a lack of access to adequate housing as well as the different tiers or degrees of homelessness, including: rough sleeping (primary homelessness); living in temporary or transitional accommodation with uncertain tenancy (secondary homelessness); and marginal housing with poor amenities or over-crowding (tertiary homelessness). The cultural definition of homelessness (whilst not official government policy) is widely accepted both in the literature and amongst services, and is therefore adopted as the basis for this study.

Service integration

Interest in researching and theorising concepts of service integration is long-standing (Bolland & Wilson 1994). Despite the development of a body of theory about service integration within the business and human services literature, there is no single, accepted definition. There are, instead, a multitude of definitions, utilising widely differing terminology, each with many possible interpretations. The definition of service integration continues to evolve. For example, Konrad (1996: 6) suggested simply that service integration 'is a process by which two or more entities establish linkages for the purpose of improving outcomes for needy people', whereas Browne and colleagues (2007: 2) go into more detail describing an integrated human service network as:

   a coalition or strategic alliance between appropriate agencies
   from multiple sectors (social, health, education) or funding
   sources (public, not for profit, private) that together collaborate
   and function to provide a continuum and spectrum of
   comprehensive services and opportunities for people of various
   ages with complex needs.

A number of authors have proposed that service integration can be measured according to a continuum or scale of intensity (Randolph et al. 1997; Brown & Keast 2005; Glasby 2005; Leutz 2005; Browne et al. 2007; Keast et al. 2008). For example, Brown and Keast (2005) and Keast and colleagues (2007, 2008) distinguish between three levels of integration, spanning what they label an 'Integration Continuum', that range from loosely to fully integrated. They stress, however, that it is not a progressive continuum, such that the end goal needs to be full integration. Instead each level describes an integration mechanism which should be chosen according to the goal of the particular network involved.

Indeed, there is a common view in the literature that 'full' integration (however one conceives it) may not be necessary in all situations and that the level of integration, or the types of strategies employed to achieve it, need to be determined according to the fundamental objectives of services being integrated (Mandell et al. 2009).

In addition to the intensity of service integration, the literature also considers the kinds of agencies or organisations should be included in order to claim integration. Bolland and Wilson (1994) assert that if the membership of the network is limited, so too is the potential outcome. From a more client-centric perspective, Randolph and colleagues (1997) point out the benefit of a broad membership, in that clients who experience homelessness often have complex needs and thus require a broad array of service support. In particular, in proposing a need for a cross-sectoral approach, Bryson and colleagues (2006) emphasise the importance of connections between mainstream services and those services which cater for specific disadvantaged groups. Lake (2005) also argues that the homelessness and housing system requires membership of other services outside the immediate sector itself, because clients overcoming homelessness or housing issues often also need assistance from health, education, employment and legal services, amongst others. It is therefore critical that they too are involved in the overall management of a client's support.

The question of network membership also includes decisions about the levels within an organisation that should be involved in service integration. These might range from front-line service delivery to management or policy-making levels (Isett & Ellis 2007; Seuter et al. 2007). There is a strong consensus within the literature that front-line service delivery partnerships are most easily achieved (Bolland & Wilson 1994), easier to sustain (Bryson et al. 2006) and more innovative than those reported to occur at the system level (Keast et al. 2008). System-level integration generally involves integration strategies implemented through policy and formal procedures, including formal funding arrangements, joint use of resources and formalised protocols. Although both types of strategies carry obvious benefits, it is generally agreed that for integration to be sustainable it needs both the flexibility and 'buy-in' that comes with front-line approaches, as well as the support and resource allocation that can be achieved by integration at management and policy levels (Burnes 2004).

Having considered these definitions of service integration and its scope and dimensions, for the purposes of this research we adopted the definition put forward by Jones and colleagues (2007: 9) as that which most usefully and comprehensively summed up the key objectives of service integration as it applies to the homelessness sector. They describe service integration as:

   structures and processes which attempt to bring together
   the participants in human services systems with the aim of
   achieving goals that cannot be achieved by those participants
   acting autonomously and separately. These goals include greater
   coherence and cohesion, efficiency, effectiveness, and consumer
   accessibility. These structures and processes may occur at the
   policy or service delivery levels, or both, and can involve several
   different modes and instruments of integration.

Despite what appears to be a lack of consensus on what service integration is and how it can best be achieved, there is an overwhelming agreement among researchers that service integration is an important aspect in tackling homelessness (Fine et al. 2000; Keast et al. 2008).

A focus on service integration in this field arose when the responsibility for tackling homelessness shifted in the 1970s. This shift saw a move from a system where the clients were responsible for navigating their way between services to one where the services aimed to provide a pathway out of homelessness for the client (Lake 2005). Taking the 'needs of the client as a point of departure' was seen as an important step toward achieving better integration (Wihlman et al. 2008: 10). The benefits of this approach (and therefore the rationale behind service integration) include improved access and responsiveness and reduced costs (Provan & Milward 2001), greater client access to services (Isett & Ellis 2007), reduced duplication (Randolph et al. 1997), enhanced accountability, equity and consistency (Jones et al. 2007) and greater opportunity for early intervention and prevention (Konrad 1996). In addition, integration develops collective knowledge between individuals, their organisations and the networks as a whole. Increased collective knowledge also has the potential of providing organisations with improved ways of addressing homelessness and opportunities for further improvement of service integration itself (Mandeli et al. 2009).

A number of service integration strategies are currently employed by social services within Australia. These include: shared service or multi-purpose service hubs; integrated case management strategies; service networks; merging of government agencies; and school-linked services amongst others. However, it is not clear which of the many available strategies are the most effective and for whom they are most appropriate (Locke et al. 2007).

It is necessary to also acknowledge that no single strategy of service integration will be appropriate to all situations and organisations (Sueter et al. 2007). Service integration strategies require investment before costs are reduced and benefits accrued (Keast et al. 2008). The establishment and maintenance of service integration also requires resources (Kenis & Provan 2009) and cannot be viewed as an absolute solution to homelessness, particularly where an overriding lack of housing stock is apparent (Jones et al. 2007).

Another question for service integration involves determining which existing strategies are suitable for use within different geographical areas. Major considerations in assessing appropriate strategies for regional contexts include costs and resources, as regional areas often suffer from a scarcity of existing resources; the geographical spread of both populations and services; and prohibitive costs associated with establishing certain integration strategies in regional settings (Fine et al. 2000; Paterson 2000; Seelig et al. 2008).

Barriers to service integration

Barriers to integration reported in previous research include difficulties in acquiring and sharing knowledge between services (Lake 2005); a lack of opportunity to network across systems (Lake 2005); difficulties in achieving an appropriate skills mix within networks (Seelig et al. 2008); a lack of resources or funding constraints (Provan & Milward 2001); philosophical differences between organisations (Baulderstone 2008); and, different funding eligibility requirements and administrative policies (Randolph et al. 1997; Phillips et al. 2009).

Another reported barrier is the compulsory competitive tendering process that has been increasingly adopted by governments. Competitive tendering has been seen as being at odds with the philosophy of trust required for effective integration (Fine et al. 2000; Munn 2003). Bryson and colleagues (2006) have also argued that funding differences and power imbalances between service providers (especially government and non-governmental agencies) may reduce trust.

Randolph and colleagues (1997), Baulderstone (2008) and Keast and colleagues (2008) also suggest there is a need for key government agencies to foster the integration that their policies are often advocating. This includes actively participating in and supporting integration strategies, or considering funding strategies which would allow for the continual support of network or integration arrangements.

Study and methods

The purpose of this research project was to identify and develop effective service integration strategies suited to a non-metropolitan setting. The research study involved both quantitative and qualitative data collection from representatives of a range of service providers. The study was conducted between July 2010 and September 2011.The research was undertaken in the Northern Rivers region which is situated on the North Coast of New South Wales.

Background: The Northern Rivers region of NSW

The Northern Rivers region is comprised of three sub-regions commonly referred to by people within the region's social service sector and the wider community as the Tweed, Richmond and Clarence river regions. The names are derived from the local river systems and encompass the seven Local Government Areas of Tweed, Kyogle, Byron, Ballina, Lismore, Richmond Valley and Clarence Valley. The region is characterised by a fast growing population, high housing costs, low housing supply, low levels of social housing and services for homeless people, and relatively low socioeconomic conditions. Census data from 2006 (1) indicates that 15 per cent of primary homelessness in NSW was recorded in this region (NRSDC 2011), yet it has less than four per cent of the State's overall population. Rates of homelessness in the region stood at 124 per 10,000 residents--almost triple the NSW rate of 42 per 10,000 (Chamberlain & MacKenzie 2009). Conditions that contribute to homelessness in the region include median income levels across the seven local government areas that are approximately two-thirds of Sydney median income (ABS 2006), whilst over the past decade unemployment in the region has been consistently higher than State averages (DEEWR 2009).

Sampling and data collection

A participatory action research (PAR) framework (Hall 1981) was utilised to gather data about the homelessness and housing sector, its profile and existing and potential service integration strategies.

PAR has been defined as a collaborative and systematic investigation to bring about social change or to educate the people affected by the issue being studied (Hail 1981). In principle, PAR extends beyond simply investigating a problem, towards developing and implementing practical outcomes for those involved (Taylor et al. 2004), PAR requires the active involvement of those being studied throughout the research process, including joint consultations between researchers and participants as to how the project should progress (Park 1999). This philosophy of empowerment acknowledges the wealth of assets the participants of the study can bring to building knowledge and their capacity to effect change (Minkler 2000), By utilising PAR, the researchers aimed to ensure the research itself was relevant, grounded in demand within the sector and acted as a means, in itself, for improving service integration (as outlined below in the results).

The data collection was undertaken in three phases. These involved an online self-administered questionnaire; three sub-regional focus groups; and the development and implementation of two service integration strategies. Each phase was deliberately designed to capture more detailed information on possible strategies that could be implemented within the final phase of the research.

In line with PAR principles, it was critical to the research that the participants took an active role in determining and refining the research process and assessing the usefulness of the research throughout. It was important for both researchers and participants that the research was built on equal contribution and co-learning between parties. The researchers fed the results of each phase back to both participants and the sector more broadly. Plans for the research were presented to local organisations through email updates to local community sector mailing lists and presentations to local service committee. This provided a platform for informing the next phase of research, along with an opportunity to evaluate the accuracy of the findings so that more generalisable claims could be made.

For the purposes of this study, the whole range of service organisations that currently assist people who are homeless or at-risk of homelessness (as according to the cultural definition established by Chamberlain and MacKenzie (2008)) were considered relevant participants in the research. In acknowledgement of the wide range of issues that may impact upon an individual's housing arrangements, it was important that the study included personnel from service organisations from a range of sectors and services, both generalist and specialist. Participants included personnel from government agencies, non-profit organisations and private sector organisations. Services considered peripheral to the sector such as health, education and training, alcohol and drug treatment services, and legal services were also considered relevant participants for inclusion in the study.

An invitation to participate in the study was emailed out to a list of 181 service provider personnel within the study region meeting the above criteria. The mailing list was provided by the partner organisation (a community-based, not-for-profit organisation) involved in the research project. In order to reach as many potential participants as possible, email recipients were asked to pass on the project information to others within the sector that may be interested in participating.

The questionnaire was designed to collect descriptive information about the sector, including the profile of organisations who responded, the position within the organisation held by the participant, the average number of referrals an organisation made for each client, and participant perceptions of the usefulness of service integration strategies in helping clients, as well as in achieving organisational goals. The questionnaire also included questions about existing integration within the region, how participants thought greater integration could be achieved and perceived barriers to integration. The questionnaire incorporated service integration scales (based on Konrad 1996) to measure the level of service integration currently existing within the sector in the region. The questionnaire was completed by 71 participants who identified themselves as service providers working with those currently experiencing or at risk of homelessness within the Northern Rivers.

Once the findings from the questionnaire were finalised, the second phase of the research began. This involved three sub-regional focus groups. Invitations to participate in the focus groups were sent out to all questionnaire participants who provided contact details for further correspondence regarding the project. An invitation to participate in the focus groups was sent out using the same mailing list used to recruit participants for the questionnaire. Again email recipients were asked to forward on the details of the project and invitation to participate to others in the sector who might be interested in participating. Given the geographical spread of the region, sub-regional focus groups (held in Tweed, Lismore and Grafton) allowed for increased representation across the whole region. A total of 41 participants attended the three focus groups. It is impossible to ascertain precisely how many of the focus group participants completed the survey as there was an option to complete the survey anonymously but then still participate in the focus group. However, it is known that 18 focus group participants did identify themselves on the questionnaire.

The aim of the focus groups was to propose service integration strategies that were appropriate for their region. At the beginning of the discussions, participants were provided with an overview of the survey findings. This provided a common starting point for the subsequent discussion. The focus groups were considered an important instrument that would foster co-learning and collaboration between the researchers and participants, as embedded in the principles of PAR.

In addition to the survey findings, participants were also given examples of current service integration strategies derived from the literature to discuss. These included a mix of service delivery level strategies such as shared information systems, staff secondments and local resource registers as well as organisation and system level strategies including joint/pooled funding, Memoranda of Understanding (MOUs) and pilots of integration strategies.

With this background information, participants were asked to work in small groups to identify actual or potential service integration strategies they believed were appropriate for meeting the specific needs of their region. They were also asked to identify possible barriers to the successful implementation of the service integration strategies they identified, as well as any existing resources that could assist in the implementation of identified integration initiatives. Once these activities had been completed, each small group presented all the strategies they had identified to the other focus group participants. Once all groups had presented, individual participants were then asked to vote for three strategies they wanted to see further developed by the researchers as part of Phase Three of the project. These strategies were then further workshopped in smaller groups to establish how the strategies could be developed and implemented. Participants were asked to consider questions such as: Who needs to be involved? What barriers need to be addressed? What existing capacity for fostering the strategy exists? And what additional attainable resources would be required?

Once the strategies had been identified, the researchers needed to determine which ones would be implemented in Phase Three of the study. Factors such as project timelines, budget limitations and the non-participation of some members of the service system in the study influenced the choice of strategies. The final selection was also influenced by consultations with service providers in the relevant sub-regions to ensure the chosen strategies were likely to be of local benefit and able to be implemented.

Results

Survey results

In total, 71 responses were received. Questionnaire participants identified themselves as frontline service workers (57 per cent), administration (eight per cent) or managers (35 per cent). Of the respondents, 20 per cent identified themselves as working with government agencies while the remaining 80 per cent worked with non-government organisations.

Approximately one-third of respondents (33 per cent) reported that their organisation offered multiple services. For some organisations this was due to a lack of specialist services in non-metropolitan regions, requiring available services to be multi-focused in order to cater for the range of client needs. The participation of services such as youth (16 per cent) and health services (four per cent) which are not directly linked to homelessness but that nevertheless identify as dealing with homeless or at-risk-of-homelessness clients, also confirms the need for a broad array of services to be involved in the sector. The breadth and complexity of client needs was also evident, indicated by the high rate of multiple referrals on behalf of their clients. Only 13 per cent of respondents indicated that less than one-third of their clients required referrals to other services, with 31 per cent of respondents indicating at least 90 per cent of their clients require further referrals to other services. On average, respondents indicated it was most common to make either two (31 per cent of respondents) or at least five referrals (33 per cent of respondents) for a client, as shown in Figure 1. These findings reinforce the need for services to work together to meet the diverse needs of people who are homeless or at-risk-of-homelessness.

[FIGURE 1 OMITTED]

Previous research has found that service integration is imperative for improving efficiency in the homelessness and housing sector. Participants were asked to rate the importance of integration in achieving the following outcomes or goals in: reaching organisational goals; improving outcomes for clients; helping the service system as a whole; and enabling them to adequately do their job. Most respondents (see Figure 2) felt it was either 'important' or 'very important' to achieving all of the proposed outcomes.

[FIGURE 2 OMITTED]

Barriers to integration

When questioned about barriers to service integration, participants were provided with a list of commonly cited barriers found within the literature and asked to rate their significance from 'not at all significant' to 'extremely significant'. The results (see Table 1) were similar to those in other studies. A lack of housing stock was found to be the most overwhelming inhibitor of integration, with 93 per cent of respondents classifying it as either 'somewhat' or 'extremely significant'. Without housing stock, service providers are limited in the opportunities to work with other services to provide solutions for their clients. More frequently Occurring responses included a lack of access to information, costs/resourcing problems, geographical spread of services and time constraints or high workload.

In addition to the list of barriers provided, participants were also able to add any other barriers not listed that they felt were either extremely or somewhat significant, Respondents identified barriers not commonly reported within existing literature, including inflated housing costs within the region; a lack of specialist clinical services and a lack of willingness to compromise amongst services when attempting to collaborate on joint funding submissions or joint projects. A lack of reliable and affordable public transportation and a lack of services in outlying areas were also listed as barriers to integration, along with unstable relationships with local real estate agents.

Critical factors fostering integration

When participants were asked in an open-ended question what they believed were the critical factors in fostering integration, the findings were also similar to those from previous studies. Open-ended responses were coded thematically and themes were entered into an Excel spreadsheet. The frequencies of themes within the responses were then collated. Trust and personal relationships were considered critical in fostering integration. There was also a call for more formalised joint planning and procedures involving commitment from more senior level staff. Although acknowledging that many agencies and individuals within the sector knew a lot about each other, respondents still emphasised that a more uniform and systematic way to share information about services and referral processes was necessary. The use of a centralised database or website was one suggestion provided. Integrated case management and sharing of client information was also suggested as a possible approach. Suggestions were made to overcome competition for funding through co-funding or co-resourcing of programs. It was felt that this could strengthen both the levels of integration and also the likelihood of additional funding for the region.

Respondents involved in direct service delivery overwhelmingly emphasised the importance of good personal relationships, access to information and informal opportunities to network in fostering integration. A number of direct service delivery respondents also, however, saw benefit in more structured joint planning, Memoranda of Understanding (MOUs) and joint funding submissions.

Senior management participants appeared more likely to respond that formal joint planning and top-down integration strategies are important in fostering integration, although in one case a senior manager echoed the responses of front-line workers:

   Informal meetings; Taking the time to build relationships;
   Senior management building on the ground work that our
   community services workers do every day in working together;
   Dissemination of information by representatives on peak and
   regional bodies so that everyone understands what's happening
   in the region; Trusting each other, being transparent, ethical and
   supporting each other.

The final section of the questionnaire asked participants to indicate on a scale of integration (adapted from Konrad 1996) the existing level of integration between themselves and other organisations within the region's sector. Definitions for levels of integration as found within the literature (for example, Konrad 1996; Flatau 2010) were provided to participants. These relationships were then mapped using network analysis software (UCINET). The results provide a snapshot of regional integration which shows differing levels of integration dependent on the relationship and usual contact between the organisations (see Evans et al. 2011). Organisations within sub-regions tended to show higher engagement with each other, as opposed to an even spread of integration across the entire region. Also evident was the role of a few organisations in connecting the sub-regions together. This could be seen when mapping integration across all three sub-regions. These organisations acted as conduits for information sharing as shown through the levels of integration reported. These organisations also tended to be larger in size, located in more than one location (offices within multiple sub-regions) or funded to work across the entire region as opposed to within a sub-region. However, because not every service provider within the region participated in the survey, it is acknowledged the results do not capture all possible relationships within the sector.

Focus groups

As stated previously, all 41 participants involved in Phase Two of the research were provided with a summary of the survey results at the commencement of the focus group. This enabled the participants to use the findings as a platform to provide further data that could be collected or verified, and to enable the establishment of potential service integration strategies appropriate to the region.

There was general agreement amongst the focus group participants about the findings from the questionnaire. In particular, they agreed that the strength of existing relationships was critical in fostering cooperation between agencies. Participants felt that strong relationships and a high level of regional service system knowledge were preserved as a result of personnel remaining within the sector, even if their actual positions or employer changed. However, it was also noted that unless the region's sector continued to develop and attract new personnel, it may become static and fail to take up new opportunities or approaches.

Across the three sub-regional focus groups, 29 strategies for service integration were initially identified. Sixteen of these suggested strategies that were not currently in operation within the study region. Of the 13 local strategies given, eight relied on local capacity rather than deliberate policy or system-level integration interventions. Participants who put forward system-based integrated strategies were identified as senior managers within the organisations where they worked or were people who had worked in metropolitan service systems.

The local examples of integration put forward by participants included specific homelessness prevention and intervention responses by public sector agencies targeting households at risk of losing their tenancies in the social and private rental housing sectors. Partnership agreements between housing providers and support agencies, privately funded short-term accommodation initiatives, and service-level communication and networking mechanisms also featured heavily. A 'service hub' currently operating in one of the sub-regions, along with a facilitated interagency network operating in another region, were identified by all three focus group participants as effective and replicable service integration strategies for regional areas.

The participants' vote identified five achievable priority service integration strategies. These were Integrated Case Management, Establishment of Service Hubs, Development of Networks, Establishment of Project-based Working Groups and the Integration of Aboriginal Services with Mainstream Services.

Development and implementation of service integration strategies

In deciding which of the five integration strategies should be developed and implemented as a part of Phase Three of research project, consultation was undertaken with key stakeholders needed in the development and implementation of each. These stakeholders were either participants in previous phases of the research project, had been identified as key to the development of a strategy, or had previous experience in the development or facilitation of proposed strategies. Time and budgetary constraints were also a necessary consideration for the research team, so it was imperative that strategies could be developed and implemented within the timeframe of the project and to a satisfactory level. As a result, two strategies were chosen, The first was the development of a Housing and Homelessness Network in the Tweed sub-region and the second was a strategy to facilitate better integration through an Integrated Case Management 'Do Tank' involving participants across the entire study region. The 'Do Tank' was an action-focused workshop that aimed to develop integrated case management strategies that could be implemented by participants within the group using existing resources or with the likely ability to access the resources needed.

The Tweed sub-regional network was to include a full range of stakeholders involved in homelessness and housing issues within the Tweed region. The network's objective would be to pursue goals identified and agreed upon through general consensus by its members. The development of this network drew on existing social capital in the region and a history of local service providers having banded together to achieve specific outcomes. Participants hoped the establishment of this network would act as a framework for generating and capturing the social capital, and allow them to work together more closely in future. In fostering the development of this network, the researchers called meetings for interested parties to agree collectively on the terms of reference for the network and the outcome-based objectives for the following year. At the time of writing, the network was meeting every eight weeks and was operating autonomously from the research project (although some of the researchers were still participating as members). The network had identified priorities for action for the coming year and was planning its official launch. Through the network they have been working together to secure premises for a drop-in centre. Its membership profile includes individuals from local, State and federal government offices; non-profit organisations; private enterprise; interested individuals from the community; and researchers from the local university.

The second strategy identified in Phase Three by the Richmond sub-regional focus group was to facilitate better integration through an Integrated Case Management 'Do Tank'. The 'Do Tank' was an action-focused workshop named in recognition of the fact that participants were encouraged to show a willingness to act upon strategies proposed during the workshop. It was hoped the 'Do Tank' would improve integrated case management across the region's sector in a manner which involved a care team and client-inclusive approach. In preparation for this, the researchers, in consultation with experienced case managers within the region, utilised the literature to develop strategies for effective case management.

The researchers, with the assistance of others within the sector, identified and then interviewed experienced facilitators who worked in the sector as case managers, McCashen's (2005) method was suggested by one interviewee as appropriate for facilitating a group approach to identify opportunities and strategies for increasing integrated case management across the region's sector. In line with the assumptions of participatory action research, this approach also assumes that the participant is the expert within the process and their knowledge should foster the development of strategies for change (McCashen 2005). The process would encourage participants to work towards an action plan for what needs to be done in order to achieve a preferred model of an integrated case management system. The desired outcome would be one where the participants owned the process and would commit to implementing action required to meet the end goal. The emphasis throughout this process would remain on what current strengths within the sector could be drawn upon in order to achieve the preferred outcome.

The researchers concluded that an opportunity to bring the sector together for a 'DO-TANK' would provide a valuable opportunity for stakeholders to work together, learn about each other and each other's services and the collective potential of the group. In itself, this exercise was expected to be valuable in building service system cohesion. The DO-TANK involved 21 case managers representing 17 organisations, along with the researchers as co-learners and participants interested in pursuing integrated case management strategies. The focus for this day was a participative and collaborative process (Burnes 2004) facilitating what Gronda (2009) calls a 'realist' perspective. Gronda (2009) suggests a 'realist' approach by participants ensures that strategies put forward were relevant to the local sector and achievable with the available resources.

Additional outcomes from the research project

In addition to identifying strategies that would improve service integration, participants during each phase of the research project expressed appreciation for the research project as a strategy for fostering service integration in itself. For many, the physical act of coming together to participate in the research was an opportunity to network, share information, and discuss client cases. Although not a primary consideration in the design of the methodology, the researchers quickly appreciated the project as a mechanism for fostering immediate service integration, as well as for identifying future strategies. This outcome supported the researchers' prediction that the participatory approach of the research project would stand alone as an effective tool for facilitating integration, increasing local service integration and service system knowledge.

Conclusion

Participants in this study strongly supported the notion that service integration is an imperative for improving efficiencies in the homelessness and housing sector. More than 90 per cent of participants felt integration was either 'important' or 'very important' to achieving organisational goals; improving outcomes for clients; helping the service system as a whole; and enabling their job to be done more efficiently.

The findings from this study support Roufeil and Battye's (2008) assertion that there is insufficient information about specific non-metropolitan problems, strengths and approaches, and that greater resourcing and trialling of a variety of service integration strategies outside metropolitan settings is warranted. There is a need to assess local capacity (both strengths and limitations), drawing from local knowledge and existing linkages, in order to tailor interventions for a particular service system. This study emphasises the importance of involving local service providers in determining suitable integration strategies for a region. It was apparent from the study that an effective way for this to occur was to ensure the involvement of the sector through a participatory process from the early stages of identifying and developing strategies.

The utilisation of a participatory process ensured that the initiatives instituted as part of the research project have continued since its completion. For example, the Tweed network of stakeholders involved in homelessness and housing issues continues to work together to secure a location for a drop-in centre.

The research has also influenced non-metropolitan policy development and implementation. For example, the study report (Evans et al. 2011) has been placed on the NSW Going Home Staying Home website--the New South Wales Government's homelessness service reform website. It is the first 'evidence' report on the website. By outlining the methods employed or suggested strategies which other regional; service sectors could adopt, the study also provides a starting place for other non-metropolitan areas to consider and develop service integration strategies relevant to their particular local area.

References

Australian Government (2009) The Road Home: A National Approach to Reducing Homelessness--The Australian Government's White Paper on Homelessness, Canberra, Australian Government, http://www.fahcsia.gov.au/sa/families/pubs/SFCSevaluation/Pages/evaluationpubssupport.aspx

Baulderstone, J. (2008)'I know that I am not alone', Report on the Integrated Service Delivery Project for Clients with Multiple Complex Needs, Southern Housing Round Table, Flinders Institute of Public Policy and Management, Adelaide.

Beer, A., Delfabbro, P., Natalier, K., Oakley, S. & Verity, E (2005) Developing models of good practice in meeting the needs of homeless young people in rural areas. Final Report. Australian Housing and Urban Research Institute, Southern Research Centre.

Bolland, J.M. & Wilson, J.V. (1994) 'Three faces of integrative coordination: a strategy of interorganizational relations in community-based health and human services', Health Services Research, 29 (3), 341-366.

Brown, K. & Keast, R. (2005) 'Social services policy and delivery in Australia: centre-periphery mixes', Policy and Politics, 33 (3), 505-518.

Browne, G., Kingston, D., Gardisa, V. & Markle-Reid, M. (2007) 'Conceptualization and measurement of integrated human service networks for evaluation', International Journal of Integrated Care, 7 (20), article e51.

Bryson, J.M., Crosby, B.C. & Stone, M.M. (2006) 'The design and implementation of cross-sector collaborations: propositions from the literature', Special Issue, Public Administration Review, 66 (Supplement), 44-55.

Burnes, B. (2004) 'Kurt Lewin and the planned approach to change: a re-appraisal', Journal of Management Studies, 41 (6), 977-1002.

Chamberlain, C. & MacKenzie, D. (2008) Counting the Homeless 2006, Australian Bureau of Statistics, Cat. No. 2050.0.

Chamberlain C & MacKenzie, D, (2009) Counting the Homeless 2006, New South Wales, Australian Institute of Health & Welfare, Canberra.

Council of Australian Governments (2009) National Partnerships Agreement on Homelessness, Australian Government, http://www.coag.gov.au/intergov_agreements/federal_financial_relations/docs/national_partnership/national_partnership_on_homelessness.pdf

DEEWR (Department of Education, Employment and Workplace Relations) (2009) North Coast Keep Australia Working Report, www.deewr.gov.au/Employment/Programs/PEA/.../NorthCoast.doc

Evans, T., Neale, K., Buultjens, J., Davies, T., (2012) Service integration in a regional homelessness service system. Report for Department of Families, Housing, Community Services and Indigenous Affairs, http://homelessnessclearinghouse.govspace.gov.au/files/2012/04/NHRA-Northern-Rivers-Report.pdf

Fine, M., Pancharatnam, K & Thomson, C. (2000) Coordinated and Integrated Human Service Delivery Models: Final Report, Sydney, Social Policy Research Centre, The University of New South Wales

Flatau, E, Conroy, E., Clear, A. & Burns, L. (2010) The integration of homelessness, mental health and drug and alcohol services in Australia, AHURI Positioning Paper No. 132, Perth, Western Australian Research Centre, AHURI, www.ahuri.edu.au/publications/download/80568_pp

Glasby, J. (2005) 'The integration dilemma: how deep and how broad to go?' Journal of Integrated Care, 13 (5), 27-30.

Gronda, H. (2009) What makes case management work for people experiencing homelessness?--evidence for practice, AHURI, Final Report No. 127, Perth, Western Australian Research Centre, AHURI, www.ahuri.edu.au/publications/download/90437_fr

Isett, K. R. & Ellis, A. (2007) Explaining new relationships: Sector, network, and organizational impacts on the growth of linkages in multi-sector service delivery networks, paper presented to The 12th Annual Conference of the International Research Society for Public Management, Brisbane 26-28 March.

Jones, A. Phillips, R. & Milligan, V. (2007) Integration and social housing in Australia: challenges and options, AHURI Positioning Paper No. 102, Brisbane, Queensland Research Centre, AHURI, www.ahuri.edu.au/publications/download/20336_pp

Keast, R.L., Brown, K. & Mandell, M. (2007) 'Getting the right mix: unpacking integration meanings and strategies', International Public Management Journal 10(1), 9-33.

Keast, R., Waterhouse, J. Brown, K. & Murphy, G. (2008) Closing gaps and opening doors: The function of an integrated homelessness service system, Final Report, Brisbane, Queensland Department of Communities.

Kenis, P. & Provan, K.G. (2009) 'Towards an exogenous theory of public network performance', Public Administration, 87 (3), 440-556.

Konrad, E. L. (1996) 'A multidimensional framework for conceptualizing human services integration initiatives', New Directions for Evaluation, 1996 (69), 5-19,

Lake, P (2005) 'Responding to homelessness--a 'Joined Up' government approach', paper presented to Building for Diversity, National Housing Conference, Perth 28-29 Oct.

Levine, I. & Fleming, M. (1986) Human Resource Development: Issues in Case Management. Rockville, MD: National Institute of Mental Health.

Leutz, W. (2005) 'Reflections on integrating medical and social care: five laws revisited', Journal of Integrated Care, 13(5):3-12.

Locke, G., Khadduri, J. & O'Hara, A. (2007) 'Housing strategies', Paper presented at the 2007 National Symposium on Homeless Research, US Department of Housing and Urban Development, 1-2 March, http//aspe.hhs.gov/hsp/homelessness/symposium07/locke/.

Mandell, M., Keast, R.L. & Brown, K.A. (2009) The importance of a new kind of learning in collaborative networks, proceedings of the European Group of Public Administration Conference: the Public Service: Service Delivery in the Information Age, Malta, 2-5 September.

McCashen, W. (2005) The Strengths Approach: A Strengths-Based Resource for Sharing Power and Creating Change. Bendigo: St Luke's Innovative Resources.

Minkler, M. (2000) 'Using participatory action research to build health community', Public Health Reports, 115 (2-3), 191-7.

Munn, P. (2003) 'Factors influencing service coordination in rural South Australia'. Australian Social Work, 56 (4), 305-317.

Northern Rivers Social Development Council (NRSDC) (2011) Summary: Housing information for the Northern Rivers Region, NSW. Lismore: Northern Rivers Social Development Council, http://www.nrsdc.org.au/housing-news

Oakley, D. A. & Dennis, D. L. (1996) 'Responding to the needs of homeless people with alcohol, drug, and/or mental disorders'. In J. Baumohl (Ed). Homelessness in America. Phoenix, AZ: Oryx Press.

Park, P. (1999) 'People, knowledge, and change in participatory research'. Management Learning, 30 (2) 141-157.

Phillips, R., Milligan, V. & Jones, A. (2009) Integration and social housing in Australia: theory and practice, Final Report No. 129. Brisbane, Queensland Research Centre, AHURI.

Provan, K.G. & Milward, H. B. (2001) 'Do networks really work? A framework for evaluating public-sector organizational networks', Public Administration Review, 61 (4):414-423.

Randolph, E, Blasinsky, M., Leginski, W., Parket, LB. & Goldman, H. (1997) 'Creating integrated service systems for homeless persons with mental illness: The ACCESS program', Psychiatric Services, 48 (3):369-373.

Rog, D., Andranovich, G. & Rosenblum, S. (1987) Intensive Case Management for Persons Who are Homeless and Mentally Ill: A Review of CSP and Human Resource Development Program Efforts, Vol. 1-3. Rockville, MD: National Institute of Mental Health.

Rosenheck, R., Morrissey J., Lam, J., Calloway, MS, Johnsen, M., Randolph, E, Blasinsky, M. & Goldman, H. (2001) 'Service delivery and community: Social capital, service systems integration, and outcomes among homeless persons with severe mental illness', Health Services Research, 36 (4):691-710.

Roufeil, L. & Battye, K. (2008) Effective regional, rural and remote family and relationships service delivery, Melbourne, Australian Institute of Family Studies.

Seelig, T., Thompson, A., Foster, M., Phillips, R. & Ramsden, D. (2008) Evaluation of the Queensland Health Homeless Initiative. Final Report, Brisbane, University of Queensland.

Sueter, E., Oelke, N.D., Adair, C.E., Waddell, C., Armitage, G.C. & Huebner, L.A. (2007) Health systems integration', Definitions, processes and impact: a research synthesis, http://www.calgaryhealthregion.ca/hswru/documents/reports/HealthSystemsIntegration_2007pdf.

Taylor, R.R., Braveman, B. & Hammel, J. (2004) 'Developing and evaluating community services through participatory action research: Two case examples.' American Journal of Occupational Therapy, 58 (1), 73-82.

Wihlman, U., Lundborg, C., Axelsson, R. & Holmstom, I. (2008) 'Barriers of inter-organisational integration in vocational rehabilitation', International Journal of Integrated Care, 19 (8), 1-12.

Endnotes

(1) The study draws on 2006 Census data surrounding homelessness figures in the Northern Rivers Region because the 2011 Census reports only provide metropolitan or all non-metropolitan areas figures for homelessness. As such we can no longer rely on Census data to provide specific figures for the Northern Rivers Region.

Table 1: Participants' assessments of barriers identified within
the literature, ranked by barrier considered extremely
significant, n=71

Barriers                        Not at all     Slightly
                                significant   significant   Neutral
                                     %             %           %

Lack of housing                      0             0           7
High workload                        1            13          10
Geographic spread of services        4             4          13
Lack of support services             3            12          16
Costs/ resourcing problems           1            16           4
Competition for resources            4            19          16
Time constraints                     4            12          19
Tight eligibility criteria           3            13          20
Access to information                4            25          12
on other services
Securing commitment from            12            22          25
senior levels
Lack of confidence/ trust           16            15          25
Maintaining links                    2            21          26
Competition for clients             28            25          22
Potential loss of autonomy          22            19          29
Philosophical differences            7            27          46
No barriers/ problems               17            14          59

Barriers                         Somewhat      Extremely
                                significant   significant
                                   off.            %

Lack of housing                     13            80
High workload                       32            43
Geographic spread of services       45            33
Lack of support services            39            30
Costs/ resourcing problems          49            29
Competition for resources           35            26
Time constraints                    41            25
Tight eligibility criteria          42            22
Access to information               41            18
on other services
Securing commitment from            27            15
senior levels
Lack of confidence/ trust           31            13
Maintaining links                   41            10
Competition for clients             13            10
Potential loss of autonomy          20            10
Philosophical differences           13             7
No barriers/ problems                5             5