Barriers of inter-organisational integration in vocational rehabilitation.

INTRODUCTION
A project of vocational rehabilitation was studied in Sweden between 1999 and 2002. The project included four public organisations: the social insurance office, the local health services, the municipal social service and the office of the state employment service. The aim of this paper was to analyse perceived barriers in the development of inter-organisational integration.


THEORY
Theories of inter-professional and inter-organisational integration, and theories on organisational change.


METHODS
In total, 51 semi-structured interviews and 14 focus group discussions were performed with actors within the project between 1999 and 2002. A thematic approach was used for the analysis of the data.


RESULTS
THREE DIFFERENT MAIN THEMES OF BARRIERS EMERGED FROM THE DATA: A Uncertainty, B Prioritising own organisation and C Lack of communication. The themes are interconnected in an intricate web and hence not mutually exclusive.


CONCLUSIONS AND DISCUSSION
The barriers found are all related partly to organisational change in general and partly to the specific development of organisational integration. Prioritising of own organisation led to flaws in communication, which in turn led to a high degree of uncertainty within the project. This can be seen as a circular relationship, since uncertainty might increase focus on own organisation and lack of communication. A way to overcome these barriers would be to take the needs of the clients as a point of departure in the development of joint services and to also involve them in the development of inter-organisational integration.


Introduction and purpose
The term 'integration' often has positive connotations, but the definition is unclear w1-4x. Development of inter-organisational integration is a continuous process across professional boundaries as well as complex systems made up of the organisations involved. The effectiveness of integration is, therefore, difficult to evaluate w5-7x. Integration is sometimes described as a continuum from the lowest level of informal contacts, to the highest level where a common authority is established as responsible for management and operational decisions w4x.
In rehabilitation, many different actors of the welfare system are involved. The division of responsibilities between the organisations is often unclear and the clients may 'fall between stools'. Research in the area of inter-organisational integration between health care and other community agencies is also a relatively new field of interest w3x.
To improve vocational rehabilitation, a number of experiments have been made in Sweden since the early 1990s aiming at an increased integration between different public organisations. In 1993, an experiment involving financial co-ordination was launched and conducted on a local basis by two organisations: the state run social insurance offices and the health services of the county councils w8x. The financial co-ordination was expanded in 1994 through a special legislation w9x, which also included the municipal social services. This study is about a project included in such an experiment.
Mur-Veeman and her colleagues w10x suggest three concepts important for the analysis of integration as shared care: Structure is described in terms of organising for the implementation process. Power means to steer people into a desired direction and the one that is most powerful in integration work is the one whose resources are most needed by others. Culture is seen as a common base for values and understanding within a profession or organisation. These concepts are interdependent. Changes in organisational structures might for example imply changes in both cultures and power relations. Daily contacts are regarded as important in developing mutual trust and a common organisational culture. Involving the actors in developing the structure of a project may also increase the possibilities of a positive development. Structures often change during the process, and a good leadership is essential to achieve organisational change w11x.
A new culture cannot be implemented by leadership alone, however. It evolves through common experiences of daily working w12x. Trust building is an important part of this process. It can be achieved when different organisations strive together towards a common goal. It is then important to rely on each other and have the needs of the individual in focus w13x. Hitherto there is little evidence of results in interorganisational work. According to Huxham w14x this is due to the process of developing integration, which is difficult and resource-consuming. Previous studies have also shown that inter-organisational work often fails w15x.
There is a lack of empirical studies of the implementation of innovations in service organisations since they have often been published in the so-called ''grey'' literature. Greenhalgh and colleagues w16 p.2x define implementation as ''active and planned efforts to mainstream an innovation within an organisation''. The project studied can be regarded as an innovation in the organisation of human services according to the definition of the same authors: ''a new set of behaviours, routines, and ways of working that are directed at improving health outcomes, administrative efficiency, cost effectiveness, or users' experiences and that are implemented by planned and coordinated actions.'' According to Krause and Lund w17x there is also a lack of research regarding processes and barriers to implementation of programmes of vocational rehabilitation.
Barriers that impede how the implementation is done and barriers in inter-organisational and inter-professional work are intertwined in implementing a project of organisational integration. The results of a complex intervention are dependent on the structures, the processes as well as the context in which it is implemented w16x. As the world is continuously changing, it is impossible to see organisational development as a linear development through different stages w12x. Changing ways of working demand changing strategies on all levels: individual, group and organisation w18x. Inter-organisational integration mostly implies also an inter-professional integration at group level, as establishing groups with actors from different organisations involved is a common way of working in practise.
The pit-falls of developing inter-organisational integration are many and it is important to be aware of the difficulties that may arise during the process. In accordance with this the purpose of this paper was to explore the perceived barriers in a Swedish project of vocational rehabilitation aiming at development of inter-organisational integration.

Methods Setting
A local vocational rehabilitation project in a community in the middle of Sweden was studied between 1999 and 2002. The project aimed at political and financial co-ordination in vocational rehabilitation and included the social insurance office, the local health services and the municipal social service. The office of the state employment service also participated although without financial commitment. A number of politicians, managers and specialists from the different organisations were involved in the project.
The goal of the project was on one hand to reduce the suffering of the individuals and on the other hand to use the joint resources more efficiently. The project targeted long-term sick municipal citizens between 16 and 64 years of age with diffuse or multiple problems, where at least two of the organisations were involved in their vocational rehabilitation. According to Konrad w4x, this type of rehabilitation project could be seen as intermediary form of integration because it was a formalised relationship with common activities.
This article is published in a peer reviewed section of the International Journal of Integrated Care  The project began its work to develop inter-organisational integration in 1998. Innovation and implementation of new ways of working was ongoing in an integrative process. This was done in a separate organisational structure with the aim of transferring the experiences into the participating organisations later. Decisions about how to distribute the economic resources from the different organisations were taken through a joint political board. Inter-organisational groups were formed consisting of specialists and managers at different levels, parallel to the traditional organisations. A simplified picture of the project organisation that was built to accomplish the work is shown in Figure 1.
A vision described in the project plan was to find new ways of working in collaboration in order to provide a more efficient rehabilitation of people with complex problems. There was some variation in views on goals, which has been presented previously w19x. At the end of the project 16 different inter-professional teams had been established for different activities and special localities outside the original organisations were used for these groups. The activities were targeted to different groups of individuals, for example unemployed people with psychiatric disorders. There were no changes of general routines or procedures in the original organisations.

Data collection
Qualitative methods were chosen to get a deeper understanding of the perceptions of the actors within the project. Semi-structured interviews w20, 21x were performed each year 1999-2002. In total 51 individual interviews were performed. In addition, focus group discussions (FGDs) w22x were also performed with existing project groups at different levels of the organisation in 2000 and 2001 ( Table 1). In total 68 persons were involved in 14 FGDs. The number of persons in each focus group varied between 2 and 9, with an average number of 5.
Individual interviews make it possible to explore a phenomenon deeper w21x whereas FGDs are suitable for gaining knowledge of attitudes, experiences, perceptions and wishes w23x. The interaction between the participants also contributes to make different perceptions in a group evident w24x.
The interview questions and the FGDs focused on goals, leadership, processes of organising, barriers and possibilities in developing inter-organisational integration. The interviews were held at the work places of the informants or in localities of the rehabilitation project. The interviews lasted from 40 minutes to 2 hours, with an average of 70 minutes. The FGDs were more even in time, each about 70 minutes. The interviews were conducted by the first author (UW) who was also the moderator of the FGDs.
All interviews and FGDs were tape recorded and transcribed verbatim by an independent typist. The interview guide was somewhat adapted according to the informant or group in focus. During each interview probing questions were asked in order to increase trustworthiness. Criteria for selection of informants were to get a broad range over the collaborating organisations as well as over the different levels of the project-organisation. Focus groups made it possible to involve more people to reflect on the work done. The project leader and key informants from the social service, the social insurance office, the local health care and the employment service were interviewed.
The key informants were professionals, middle managers, top managers and politicians.
Informed consent was obtained from all the informants and confidentiality was assured in presenting the research. According to the in Sweden prevailing ''Act This article is published in a peer reviewed section of the International Journal of Integrated Care Table 2. Soc.
Total 15 on Ethics Review of Research Involving Humans'', approval from an Ethics Committee was not needed, and hence not applied for. Table 2 shows the distribution of the interviewees for each year of the study.

Data analysis
A thematic analysis was used, inspired by Malterud w24x. This method of analysis is suitable for identifying new views within a research area. This was done in a step-wise manner and focused on barriers to integration throughout the project: The transcripts of each of the 51 interviews and the 14 focus group discussions were read as a whole to get an overall impression and preliminary themes were formulated. The material was then coded into different meaning units. The contents of the meaning units was condensed and abstracted into themes and subthemes. This was illustrated by quotations. After that the interviews were reread to confirm that the themes were in accordance with the raw data.
The data were analysed by the first author (UW) while the last author (IH) acted as co-reader. Discussions were held until consensus was reached.

Results
Three different main themes of barriers emerged from the data: A Uncertainty, B Prioritising own organisation and C Lack of communication. The main themes and sub-themes are described in Table 3 and Figure 2.
It is important to point out that the themes are interconnected in an intricate web and hence not mutually exclusive, which is shown in Figure 2. Prioritising own organisation leads to a lack of communication with others. When dialogue and communication is not prevalent it leads to uncertainty about what is the meaning of different messages. Likewise, it is possible to start in any of the three barriers and see how they affect one another.

A Uncertainty
The aim of the project was diffuse from the start, and the overall goal was never discussed and made concrete. The length of the project was first only two years but was later extended twice which led to uncertainty in all parts of the project throughout the project time, even if it was more pronounced during the first two years.
This article is published in a peer reviewed section of the International Journal of Integrated Care  In addition, the target group of the project was not clear. Specific target groups were identified by administrators and professionals, but the groups were unclear and overlapping. In 2002, some actors reflected on whether the project had reached the individuals most in need: ''«was the selection wof activitiesx a coincidence due to the fact that there happens to be an enthusiast in a specific organisation who is extremely keen about a specific matter«is that specific target group the one who is most in need?'' (Middle manager 2002)

A2 Unclear leadership
There was an uncertainty throughout the project about who was leading who. This was particularly evident between the project leader and the advisory group, where a struggle for power took place. The members of the advisory group wanted to be above the project leader hierarchically, but he claimed that his position was directly below the politicians. The project leader was recruited from one of the organisations involved. Early in the project there were also discussions about who was to pay the salary of the project leader, the organisation where he was previously employed or the political board of the project.
The professionals were also not sure of who was in command and who could decide over their working tasks. On one hand they had their ordinary manager and on the other hand they had the project leader.

A3 Unclear roles
The lack of clarity concerning the project also made the mission of the appointed groups unclear. The actors were uncertain about their responsibilities and their authorities. At the outset of the project there were some co-operation groups at the professional level that had the task of bringing forward new ideas on how to handle certain groups of individuals, for example unemployed people with psychiatric disorders. The task was very broad and because the mandate of the group was unclear, the participants had to work all by themselves and did not know in what direction to go. After two years this co-operation group ceased.
The roles of the different groups in the project were diffuse. Sometimes the tasks of the groups overlapped and no one except the project management seemed to have a comprehensive view of the project. How different groups were interconnected was not evident. Groups changed names, came into being or ceased without anyone knowing how or why. It was also unclear how joint projects were chosen since no criteria for selection were decided.

B Prioritising own organisation
Throughout the project it was evident that the actors kept the perspective of their own organisation. The managers had a hard time raising themselves above their own budgets and seeing the benefits of the situation as a whole. The different groups within the project met about once a month. Some expressed that co-operation existed during the meetings but not in between.

B1 Territoriality
Both organisational and professional territoriality was seen. The age range of the target group, for example, was decided on the basis of the demands of the social insurance office due to their own organisation's age range in focus.
Guarding one's budget was important in 1999. There was a fear that someone else would get advantages at one's own cost. One manager expressed: The same person had a similar opinion still in 2002.
In most situations, the own organisation and its activities were prioritised. For example, some managers at high level were responsible for services delivered in several communities and they sometimes prioritised other meetings. When professionals were busy and lacking time, it was their ordinary tasks that were most important to them. Some of the participants felt alienated, particularly those from the office of the employment service, possibly due to not being involved in the financial coordination. They were also introduced into the project later than the other actors, which was perceived as a negative factor.

B2 Formal and financial focus
Economic distributions between the organisations were made for each activity within the project according to which organisation would win the most from it. Importance was also given to mobilise support for each task at every hierarchical level of the organisations involved. A project organisation was, therefore, established as a parallel separate structure. A suggestion of a new activity within the project was to go through all of the groups at different levels to get opinions and a formal report from each one. In 2002, it was decided to create a minor steering group to each single activity, which also meant one more hierarchical level within the project.
Mobilisation of support led to a focus on formal and financial aspects. There were large groups and long meetings. The focus on the individuals served was of no priority: ''still«we talked about it today, we have somewhat lost focus on those who fall between stools.'' (Project management 2002) A financial distribution of costs between the organisations involved was done according to the benefits from each specific project that was discussed. Several calculations were performed before decisions could be made to start an activity. The financial co-ordination system was regarded as complicated.

''The financial matters are so complicated that they almost need a controller, employed full-time.'' (Top manager 2001)
In almost every activity, one organisation had the main responsibility due to the financial decisions about who was to have most benefits from it. This made the other organisations lose interest in an activity where they were little involved.
Although the project was directed to target groups that fell 'between stools', the territoriality made the financial transactions the most complicated when no specific organisation could be regarded as responsible.

B3 Distrust and different perspectives
A common opinion among the actors was that territoriality already existed as a barrier within health care among different professions and different specialities, for example between psychiatry and primary care. There were also internal professional barriers perceived within the local community, for example between social and medical care for elderly people.
Health care was seen by some actors as a knowledgeoriented organisation with academic professionals. Some informants perceived a lack of co-operation from the doctors in health care. However, most work seemed to function well at group level. Outspoken conflicts occurred seldom but existed beneath the surface. The interviews once a year gave some of the participants moments of reflection on their work. During a focus group discussion in 2001 it became evident to a group of professionals that they had many different perspectives: The commitment was deeper if an activity involved one's own organisation. In 2002, some signs of withdrawing from the co-operation groups were shown from the social care and the office of the employment services.
''People still fall between stools. Whose pigeon is this, we ask, who owns the responsibility?'' (Middle manager 2002) In 2002, it was revealed that distrust had prevailed from the start, especially among managers, against the project leader, since he was recruited from one of the organisations involved. It was said that there had been a tight relationship between the project leader and the chairman of the political board, who was also a politician within the same organisation as the project leader came from. As a consequence some of the interviewees felt that the focus of the whole project was on that same organisation.
It was also obvious that there was a distrust between the project management and the top managers of the different organisations involved, and between the managersyprofessionals and the politicians. Some of the top managers suspected that their comments on proposals were censured before decisions were taken by the political board. They also felt sometimes that they were not being listened to.

C Lack of communication
Throughout the project there were flaws in communication, both vertically between different hierarchical levels and horizontally between groups. The lack of dialogue, the lack of participation and the lack of learning influenced each other in a vicious circle, most evident at the beginning of the project.

C1 Lack of dialogue
The formal structure of the project organisation made it slow to arrive at decisions. As mentioned before, every suggestion of an activity was supposed to pass

C2 Lack of participation
The project management had an idea that the project was governed according to a bottom-up approach. The ideas were supposed to come from the professionals at the grass-root level and the managers were waiting for them to come. The professionals, however, felt that the project was governed top-down and they were waiting for the managers to act and give directions. As there was a lack of communication vertically, both managers and professionals waited for the others to take the initiative. The results were passivity.

''«we were to take care of the suggestions and ideas that arose in the interaction between the administrators and the clients« It is a good and a fine idea, but it did not work.'' (Project management 2002)
Most interviewees agreed about missing the employment service in the economic co-ordination. There was a common perception among the participants that this had led to an alienation of the employment service from the project.
''« there have been some intricate matters about who was to make certain decisions when it came to money. The employment service cannot, since they do not put any money into the project. And my opinion is that this has locked them in«the problem is that they have been sitting there somewhat set aside.'' (Top manager 2002) There was no involvement of service users and the politicians did not see any need of discussion with them either. The initial study of potential target groups was oriented towards professionals, and no groups of clients or consumers were asked about their needs.

Results
The barriers found in this study were intertwined. They were partly related to organisational change and partly to the specific development of organisational integration with problems related to actors from different organisations working together.
Barriers to development of inter-organisational integration have previously been discussed by Huxham and  Vangen w15x who among other things point to problems regarding the ambiguity of interrelationships and the complexity of inter-organisational integration. In the theory of collaborative advantage, they also point to many different themes as important to develop integration w15x: common aims, communication, commitment and determination, compromise, appropriate working processes, accountability, democracy and equality, resources, trust and power. This is somewhat different from the results of this study.
Themes that emerged out of the data as barriers to the development of integration were uncertainty, lack of communication and focus on own organisation in the present study. These themes are overlapping with those of Huxham and Vangen w15x and also the concepts described by Mur-Veeman and her colleagues w10x. They argue that the concepts of power, culture and structure are the most important in the implementation of shared care. Their opinion is that the concepts are interdependent in the sense that any of them may be the point of departure.
The themes in this study offer a somewhat other perspective, as they are intertwined with barriers to organisational change. This might be due to the use of a specific project organisation in this case. This is in agreement with the results of a recent dissertation w13x regarding the importance of change management in developing inter-organisational integration.
There were two separate goals of the project with a focus on clients as well as on financial resources. Focus on clients means in this case to get effective routines in inter-organisational integration to handle the rehabilitation of persons that have been in a grey zone between different organisations' areas of responsibility. Parallel controlling of costs as well as improving inter-organisational systems implies a potential conflict according to Hardy and colleagues w25x. Use of financial resources of each participating organisation was prioritised in this project as everyone was eager to keep one's own budget. This might help explaining the limited results in terms of starting different activities outside the original organisations instead of changing ways of working within and between them. In this way, it was easier to know who was to pay.
A finding specifically related to the development of inter-organisational integration was the focus on mobilisation of support at every level in all the organisations involved, which probably was due to the mistrust that, according to Huxham w15x, often is prevalent in inter organisational work. In integration it should instead be of importance to trust one another in favour of developing joint services to clients. Trust is seldom prevalent from the start but trust-building can be developed through joint working and positive experiences that lead to enhanced trust in a positive loop w15x. In the project studied, there was joint working in inter-professional teams within each activity, but this was not spread into the original organisations.
The success of an innovation is also dependent on how the implementation is done w13, 16, 26x. The implementation of inter-organisational integration as an innovation is to a great extent dependent on leadership and ways of working w2, 13, 15x. Good leadership and communication are needed to facilitate organisational change, but there has not been so much research done in the area of leadership of interorganisational integration w2x. According to Kotter w11x, the leader should provide direction, unite individuals and motivate and inspire them. Leadership should communicate and help the participants to understand the overall vision and how different projects fit into the whole picture. When major change is imminent in an organisation, as is the case in development to interorganisational integration, the leader should be able to coach his co-workers through the process of change. The importance of a project leader and learning from experiences has also been emphasised in studies of integration work w10, 13x.
The leadership was unclear within the project studied, and the professionals were not sure of whom their leader was, the project leader or the manager within the original organisation. An evident barrier was the struggle for power between top managers and the project leader. The top managers were alienated from decisions in the project and later in the project they were reduced to an advisory group instead of a managerial group. This seems to have influenced their commitment to the project in a negative way. This is in contrast to another study of co-operation between public sector welfare organisations in vocational rehabilitation, where a conclusion was that involving higher level managers in collaborative work is important w27x.
Uncertainty prevailed during the four years of study. There was also an absence of dialogue and reflection about factors of success or failures and what to learn from them, which probably contributed to putting the own organisation in focus instead of the service user. Within the project there was an organisational focus at the expense of the service users. The needs of the clients were not taken as a point of departure in the development of joint services, which would be important in order to see the common aim of the integration work w28-30x. Inter-organisational integration is too often carried out top-down. In the project studied this means that the service users were not involved in the developmental work w30x.
It is not uncommon that attempts of change in organisations end up in doing new activities in old ways, due to powerful interests within the old context wanting to keep the status quo w31x. That is why organisational change is so important. Leaders have to contribute to create conditions for framing change and understand that even small changes in a complex context might lead to more important changes in the long run w32x.
Uncertainty is not always negative though. With a flexible approach to new ideas and a continuous monitoring of the development, adjustments of directions are possible to do underway. Thus, with a positive interpretation, uncertainty might imply incompleteness and give way for creativity and testing w33x. However, the lack of reflection upon events and learning from experiences in the project studied made the uncertainty negative. There was no system developed for learning from mistakes, which is important in organisational change. The many organisational levels of the project were also barriers to creativity and learning to achieve new ways of working w31x.
The project studied was huge and complex with many actors involved. A new project organisation was built to ensure participation. The barriers arising within the project made communication more difficult and might be questioned, since integration work is fragile and hard to achieve also without such extra difficulties w25, 29x. Inter-organisational groups dealing with clients need trust and support from leaders and management as a fuel to attain sustainability w29x. Changing structures is a natural part of a process of innovation and implementation, but the participants would certainly benefit from understanding the reasons for it. This was not communicated. However, understanding the barriers that arise is a way to overcome them in the work for inter-organisational integration aiming at improving vocational rehabilitation.

Methods
Since the context is Swedish and only one project was studied, the results cannot be generalised, but they might be transferable to similar situations. Data were collected between 1999 and 2002, but the barriers shown in the study are general and ought to be of value for the implementation of other integration projects.
The monitoring of the project during four years is a strength of the study, and so is the number of interviews and focus group discussions that were continuously performed during the period. The interviewer was the same (UW) every year, but she was otherwise not involved in the project.
This study has a perspective from the professionals of the organisations involved. In the future, it would be of great interest to make a study from a client's perspective on service fragmentation and lack of coordination. It would also be of interest to study what sort of support the clients would like to receive and what the implications of this would be for service delivery and decision-making.

Conclusions
Three general barriers were found in this study: prioritising own organisation, lack of communication and uncertainty. They are all related partly to organisational change in general and partly to the specific development of inter-organisational integration.
The results showed that focus on own organisation led to flaws in communication, which in turn led to uncertainty within the project. This can be seen as a circular relationship as uncertainty might increase the focus on own organisation and the lack of communication. When dialogue and communication is not prevalent, it might lead to uncertainty about the aims and the importance of integration and thus prioritising of the own organisation.
A way to overcome these barriers would be to take the needs of the service users as a point of departure in the development of joint services and to also involve them in the development of inter-organisational integration. This could be done for example through focused group discussions with target groups.