Inter-organisational integration for rehabilitation in Sweden - variation in views on long-term goals.

Abstract Purpose The aim was to study variation in views on long-term goals expressed by members in a project concerning rehabilitation of individuals in an area in mid Sweden. The organisations participating were the municipal social services, the county council's health care organisation, the governmental social insurance office, the national employment service and the national labour market institute. Theory A number of different meanings of the concept of integration are accounted for and the background of inter-organisational rehabilitation in Sweden is described. Methods Qualitative interviews using a semi-structured interview guide with altogether 20 leading representatives of the various organisations, all members of the Steering Committee of the project. The analysis consisted of two parts: an in-depth analysis using a phenomenographically inspired approach and qualitative content analysis. Results Two main views regarding the goal of inter-organisational integration were identified in the in-depth analysis: Category A, which regards integration as a new approach to rehabilitation, and Category B, which regards integration as a way of rendering the existing organisations more efficient. The content analysis showed a wide variation in spontaneously mentioned topics e.g. on the task of the group. Conclusions and discussion There exists a risk that divergent views in the Steering Committee concerning the future direction of the collaboration may send mixed messages providing uneven support for the professionals within the different organisations. The study points to the importance of having a common vision and common, well-defined goals at the outset of a collaborative project in order to support micro-level cooperation.


Introduction and purpose
Side by side with Norway and the Netherlands, Sweden has the highest costs for absence from work due to long-term sickness w1x. Whether or not this reflects a situation of inadequate rehabilitation is a matter for debate. However, in cases where a person's problems are diffuse and changing, heyshe runs the risk of ending up in a 'grey zone', due to a lack of certitude among the organisations involved about which one has the main responsibility for the rehabilitation w2, 3x.
Integration between organisations has increasingly been recognised as a means to improve the quality of rehabilitation, especially for individuals with complex needs that require contributions from several different professions and organisations. Different organisations have different operational goals, but in integrated work it is important to clarify the common aim and responsibilities w4x.
To develop integration between different organisations means to work across both administrative and professional boundaries, and within as well as between different complex systems of organisations. In Great Britain, for example, Primary Care Trusts (PCTs) with members from various professional backgrounds and organisations are linked to a general practitioner and have a common responsibility for the population in a catchment area. Members of the different professions are obliged to integrate their work regarding the population within the area w5-7x.

This article is published in a peer reviewed section of the International Journal of Integrated Care
This situation contrasts with that in the USA, where collaboration often occurs in networks on a voluntary basis, for example within community health partnerships, a form of voluntary integration between different organisations aimed at improving public health in a given area w8x.
The concept of integration is not unambiguous. It has generally acquired a positive connotation, although evidence is lacking in terms of its efficiency w9-11x. Literature on integration is found in various disciplines such as management science, political science, public health, psychology and sociology. In this literature, the word integration is often used synonymously with other terms like coordination, cooperation and collaboration. In our study, we use the concept of integration as defined by Konrad w12x. She describes integration as having five levels of 'intensity', from informal contacts to a common authority: • Information sharing and communication. Informal contacts • Cooperation and coordination. Joint action between services directed at individuals • Collaboration. Common activities for attaining common goals • Consolidation. A common umbrella organisation for activities in an integration project • Full integration. One common authority In Sweden, statutory experimental work concerning integration in the field of rehabilitation was launched in 1993. It was conducted on a local basis and financed by two authorities, the national Social Insurance Office and the county council's health care services w13x. There was an expansion of the financial base in 1994, by also including the municipal social services w14x. In addition, the National Employment Service and the National Labour Market Institute participated in the integrated work, although they did not contribute to it financially.
The rehabilitation project described here started in 1998 and comprised the above five organisations. A separate project organisation was developed that included professionals from the different authorities involved. The overall goals of the project were to improve the situation of the individual and to achieve national economic gains by utilising the common resources in a more efficient way. The purpose of the rehabilitation project was to cooperate to find ways for the rehabilitation of individuals, or groups of individuals, with complex problems that affect more than one organisation. The project was targeted towards municipal citizens between 16 and 64 years of age with diffuse or multiple health problems. A Steering Com-mittee consisting of managers and experts from the different authorities had the task to support the project leader and to review project proposals prior to decisions by a board of politicians.
The purpose of this study was to attain in-depth knowledge concerning variation in views of the goals of the rehabilitation project among the members of the Steering Committee, and to map out their opinions concerning their commitment to the project and its management.

Methods
A qualitative approach inspired by phenomenography was used. An essential focus in phenomenography is describing the variation in qualitatively different views and in how people perceive the surrounding world w15x. Any phenomenon from specific experiences to experiences of more general aspects of the surrounding world can be investigated using the phenomenographic approach.
What the researcher seeks to find is not the 'objective' truth, but rather how different people experience a certain phenomenon from their own perspective w16x. Phenomenography was originally developed by a research group at the Department of Pedagogy at Gothenburg University w17x and has its roots in pedagogic research on learning. The field, however, has subsequently been expanded to include the description of conceptions also within other areas, including health care w18x. In performing a phenomenographic study, approximately 20 informants is typically regarded as sufficient, as data saturation is usually achieved within this number.
The Steering Committee consisting of heads and specialists, played a central role in the project. Heads were executive managers responsible for staff and for managing a budget. Specialists were individuals with some kind of special knowledge, but without responsibility for managing budget. The Steering Committee rejected or encouraged various ideas, which then would spread to the different organisations. The Steering Committee and the project leader met once a month, to review proposals for future rehabilitation projects and to come to an agreement as to how to divide the costs between the organisations involved, prior to decisions by politicians. Hence, the members of the Steering Committee were chosen as objects for the study. The participants who had been members of the Committee at the end of 1998 were interviewed during the first half of 1999 in order to capture the different views of the rehabilitation project and the work that was done. Table 1 provides an outline of the 21 individuals who were interviewed.
SIOsSocial Insurance Offices; HCsHealth Care; Soc.sSocial Services; LMIsLabour Market Institute; PESsPublic Employment Services. Semi-structured interviews with open-ended questions were used for data collection (see Table 2). The interview guide was developed after some exploratory interviews. Based on the responses of the respondents, the question sequence in the interview guide was sometimes modified.
All of the interviews, which lasted from 50 minutes to 2 hours with an average time of 76 minutes, were conducted by the first author (US) and were taperecorded. The tape recordings were then transcribed verbatim by a typist outside the group of researchers and checked for accuracy. Seventeen interviews were conducted on the premises of the regional social insurance office, which was also used as a meeting place for the project; two interviews were carried out at the first author's workplace and two in the offices of interviewed persons.
The analysis was conducted according to the phenomenographic approach and followed various steps as described by Dahlgren and Fallsberg w19x. The first part of the analysis was done by the first author (US). Transcripts were carefully read through and significant quotes were selected concerning the theme 'the goal of the rehabilitation project'. The quotes were then compared and compiled into different categories. Summaries were made of the main content of each category and classified accordingly. Summaries of the interviews and categorisations were passed on to the last author (CSL), who filed the interview transcripts under the various categories. In a few cases where the authors were in disagreement, the interviews were reread in their entirety. The categorisation was reevaluated during discussions between the first and last authors until consensus was reached. This way of ensuring reliability or intersubjective agreement is common in qualitative research w20, 21x.
The interviews also produced a rich material of a more factual nature, which could not be described as related to conceptions. Thus, in addition, a qualitative contentoriented analysis with no pre-determined categories was performed w22, 23x. The absence of a statement regarding a certain issue may be explained by the fact that the subject in question was not spontaneously mentioned, and in some cases several opinions were expressed by one and the same person.
This article is published in a peer reviewed section of the International Journal of Integrated Care Prior to the interviews, informed consent was obtained and interview persons were guaranteed complete confidentiality. For this reason, the conceptions of the project leader have not been accounted for in tables where he would be recognised, since he cannot be grouped under any of the organisational headings.

Results
Two different categories of goal-related views, called A and B, were identified in the phenomenographic analysis. In both categories, there was a tendency towards a more 'holistic' view w10x. In category A, two subcategories of views-A1 and A2-were identified. The distribution is presented in Table 3. Below, descriptions of the various categories, including illustrative quotes, are given. The consecutive numbers refer to the the interview persons (IP).

Category A: developing a new approach to rehabilitation
According to the category A view, the goal of the project was to promote a new approach, which had not previously existed, in terms of working methods andyor organisational forms of rehabilitation. Several participants expressed the notion that, unlike the current situation characterized by separate organisations, in the long run the project would lead to the passing of new, uniform legislation and to the creation of a common agency or a merger of organisations.
'If this will in any way be successful, since it is now only experimental work, it will lead to new legislation. I believe it will lead to one joint agency for the entire population in the area. In the process, agencies will be dissolved.' (IP 5) It was important for the people in this group that longterm results would be achieved. Their view was that currently used methods make it impossible to contribute the efforts required for complex situations where no organisation is held responsible. Their opinion was that entirely new actors and new working methods must be introduced. 'I would like to see that in the end « we will have found new ways to work with those who are most complex « much of what lies behind the problems involving long-term hospitalisation and long-term sickleave, has to do with the individual having slid so far down a negative spiral, that they perceive society as their provider« And this is where I believe we come into the picture, regarding this complex of problems, that even then we are able to perceive: how can we, in a long-term view«?' (IP 16)

Category A1: individual perspective
The category A1 view was characterized by a new approach from the individual's viewpoint. The goal of the rehabilitation project was to develop new organisational forms based upon the individual problems or the appointed target groups of the project. The joint efforts would be organised from the overall perspective of the individual. Contemporary working methods were considered inefficient in tackling the basic causes of individual dysfunctional situations. It was considered to be important to contribute to the appropriate efforts in order to achieve long-term results. New working methods and new actors were deemed necessary. A consolidation of organisations may be the result. 'The needs of the individual become the focal point«, then we must adjust to that and maybe it will imply that certain boundaries are erased and new ways of working will have to be found.' (IP 17) Category A2: population perspective The category A2 view was characterised by a population perspective, which encompassed all ages. The goal of the rehabilitation project was viewed from the needs of the population in the municipality, where a common responsibility and a socio-economic perspective was important. Some respondents expressed the opinion that problems that are a threat to public health-e.g. loneliness or social isolation-must be addressed. In order to promote public health, it would also be necessary to apply preventive measures. The aim would be a higher quality of life for the entire population. New working models and new actors were seen as necessary.
This article is published in a peer reviewed section of the International Journal of Integrated Care 'Sometimes I think, 'is it possible, are we the ones to help these people or is there something else'? Because, actually, I would also like to add preventive work and the like. In that case, all age groups must be included, you can't have 16-64 (years) « if something is to be started up, then one must begin with the children, so that they don't eventually end up in these different groups. The question is what one can do for these types of specific problem groups. « But, somehow, this is another dilemma in society: they have no social network, they have neither one nor the other, who will take care of this, so to speak « if their quality of life is to be improved, then maybe completely other measures are required.' (IP 14) Category B: rendering existing organisations more efficient A characteristic feature of the Category B view was that integration would render existing organisations more efficient by, for example, facilitating the transfer of information between the organisations. In this category, only the individual perspective was articulated. Joint resources were to be used in the best way, from a holistic view of the individual. In order to achieve faster decision-making, a quicker process for handling cases would be attempted, which would lead to shorter queues and waiting times for individuals. According to this view, administrators should work more together and plan their efforts jointly. Passing on cases between professionals within different organisations should be done more flexibly, without 'gaps' or waiting periods.
'« As it is now, it takes a long time before the regional social insurance office reacts and begins to consider early retirement pension, or that they're classified as being unfit for work and you get the feeling that this person is being tossed into different activities just so they have somewhere « there should be a more rapid assessment of the person's complete situation « that a decision is reached more quickly.' (IP 15)

Other aspects
Beside the goal of the rehabilitation project, other important aspects of participation and involvement emerged, which are illustrated in Table 4.
The majority of the actors participated in the project as part of their jobs and their role in the group was mainly to participate in discussions. The task of the group was unclear for most of the participants, and for nearly one third of the participants the point of departure in the discussions was the interest of their own organisations. Half of the participants perceived the climate during the group's discussions as open, while the other half did not. Nearly all of the participants considered it the responsibility of the project leader to propel the project forwards. However, the focus so far had been mainly on issues of how to split costs between the organisations involved concerning activities within the project as well as how to organise the overall rehabilitation project.
It was also evident that no one perceived the interests of the project's target groups as having been in focus. There were varying opinions concerning what age groups the project should be directed towards, and which target groups that were most important. Approximately two thirds of the actors envisioned the project leading towards a change in the present organisational boundaries. However, almost half of the interview persons feared that the project would be seen as a threat to existing structures, and about the same fraction felt that that the current direction of the project was wrong.

Discussion
This study shows that the participants of the Steering Committee, after almost a year's work within the rehabilitation project, had qualitatively different views of the long-term goals of the project. The overall goal of the project was to integrate the organisations involved, but the main focus of the Category A view was that this development should signify something completely new, e.g. innovative organisational forms or novel actors, and include efforts to tackle the underlying causes of the individuals' problems. In this group, some respondents advanced the fifth level of intensity described by Konrad w12x, i.e. full integration as being desirable.
In contrast to the Category A1 and B views, the group representing the Category A2 view felt that the rehabilitation project comprises a population perspective. The focus is on integration at an overall level between organisations whose goal is to provide citizens with efficient service. In a similar way, Boklund w10x distinguishes between a holistic view of care in a micro perspective and a macro perspective. In the micro perspective, the encounter between the individual and the professional is essential. The professional makes an assessment of the individual's needs in its social context. In the macro perspective, on the other hand, the focus is on integration between organisations, with the aim of providing more efficient service and efforts aimed at the individual.
According to Boklund w10x, integration at the operational level can be described as having four different states of intensity. The lowest state, 'separation', This article is published in a peer reviewed section of the International Journal of Integrated Care  means a lack of integration. It is considered to exist when professionals are operating in a parallel fashion. The second state is integration through 'coordination', e.g. when an administrator sends a referral to someone without any meeting or interaction having taken place. The third state of intensity, 'collaboration', signifies interaction and common problem solving. The fourth state is 'consolidation', which means that organisations and sometimes even groups of professionals are integrated with each other to such a high degree that the same tasks are carried out.
The groups representing the Category A1 and B views have a micro perspective, where the focus is on how work integration targeting individuals should be conducted on a professional level. Many of the individuals in these groups were of the opinion that the long-term goal of the rehabilitation project was to promote uniform legislation and an organisational merger. Certain participants might view such developments as threatening, since a merger or consolidation of organisations constitutes a threat to existing organisational structures and especially to the positions of middle managers. Such a turn of events runs the risk of being curbed by the people concerned. Svensson and his colleagues w24x describe how the hierarchy of an organisation can serve as a preserving element and counteract development, since the individuals who have nothing to gain from change are usually the ones who hold the power. In the terms of Konrad w12x, the views in the Steering Committee differed on the goal concerning the intensity of integration from the second level, cooperation and coordination, to the fifth level, full integration.
Similar to Boklund's w10x classification, the interview persons also spoke of operational integration at different levels of intensity. It varied from the fourth level of integration, 'consolidation' (group A), to the second level, 'coordination' of activities between participants from different organisations (group B). Consequently, the participants in the Steering Committee had different views on the importance of integration within the rehabilitation project. If, as managers of their respective organisations, they gave the importance of integration different significance, it would in the long run imply that resources such as, e.g. time to participate in integrated groups would differ between professionals from the different organisations.
Uncertainty about the goal of the rehabilitation project would also mean uncertainty in terms of the tasks of the project. If the participants have different goals, it means that they will be heading in different directions.
As an instrument in the project organisation, the Steering Committee, would then become inefficient. This means that there is a risk that a great deal of energy will be spent on trying to understand what in fact the task is w2x. The uncertainties, which have been found to exist at the managerial level also risk shifting the responsibility for the development of the integrative work over to the professional level w25x. ued integration process. It may turn out that some participants in an activity at the local level have a complete mandate to make decisions regarding individual patients, while others must first present the issue to their manager for approval and subsequently refer it back to the group.
It takes time to get an extensive change process involving a rehabilitation project of such a size as the one described here w24x. Typically, the first phase in such a process is more or less chaotic w26x. It is important, however, to provide the actors with enough time to reflect over their experiences if long-term, goal-oriented change is to be achieved. The individuals learn lessons through the interaction between the actors involved w24x. Such learning may lead to changes in attitudes and behaviour, which are prerequisites for successfully implementing a project like this w27x.
A process of change initiated on the local level has greater prospects of being successful, if all organisational levels are involved and management is supportive w24x. In this project, the leaders participated by virtue of the fact that they were appointed and not as a result of their own choice, a circumstance which may have impacted on their commitment. How the work is conducted also determines to which extent the various affected individuals acquire a common interpretation of the reality to come w26x. Hence, in starting new integrative efforts, it is important to survey, e.g. values, ambitions and motivations of the actors involved to determine the basis, if any, for integrated work across organisational boundaries w28x.
An organisational change always includes a cultural change, and clarifying the visions of those involved constitutes is a basis for a common understanding w29x. The fact that there was such a wide variation in views within the project indicates flaws in communication between the participants. It seems that the aim, the language, and the terms of participation had not been defined and clarified enough during the early dialogue. This is particularly important in a project of inter-organisational integration, where it is known that terms can be interpreted differently. A doctoral thesis on integrative organising w30x points to the difficulties experienced by individuals from different organisational cultures in comprehending each other's vocabulary.
As this is a qualitative study, it should be recognised that the generalisability of the results is of a conceptual nature. In the analysis phase of this study, two of the three authors independently filed interview transcripts under various categories in order to enhance the trustworthiness of the study. This work method and the fact that categorisation was discussed until consensus was reached means that inter-rater reliability and trustworthiness should be regarded as high w31x.
The content analysis displayed a wide variation in views on a large number of issues central to the work of the group. In order to illustrate these differences, the numbers in Table 4 are presented as a 'snapshot' of the views in this particular Steering Committee, one year after its creation.
A difficulty encountered in the study was that few concrete situations had so far arisen. Several of the persons interviewed expressed themselves on an abstract level and had difficulty exemplifying their statements. An advantage, however, is that the study presents the situation at the starting point, before the project had really begun, which provides a good opportunity to follow how the integrative work subsequently developed on various levels. Konrad's w12x five categories have been included as a theoretical frame of reference in this manuscript. This framework could be further developed, but that would require a more extensive study of inter-organisational integration.

Conclusions
Among the participants in the Steering Committee, there were qualitatively different views of the longterm goal of the rehabilitation project. This means that they were working towards different goals, which theoretically might threaten to render the Committee ineffective in terms of the project organisation as a whole. There is also a risk that, in the future, various managers will differ in their support to integrated work, generating a fluctuating foundation for collaboration and cooperation across organisational boundaries.
The different views might partly be accounted for by reference to the fact that the project was in its early stages. Furthermore, they may partially be explained by insecurities as a result of changes within the group, as some individuals left and new members joined. Additionally, the varying views might be the result of inadequate communication, whereby the aim and the terms of participation may not have been discussed enough and made sufficiently clear and concrete. In any case, the results of the study emphasise the importance of formulating a common vision and common, well-defined goals at the very outset of an integrative work venture, in order to support integration at the micro-level involving individuals in 'grey zones'. This may be obvious in theory w2, 4x, but it seems to be very difficult to achieve in practice.
This article is published in a peer reviewed section of the International Journal of Integrated Care

Vitae
Ulla Sandströ m has majored in physiotherapy and worked as a physiotherapist in the health care system for many years. Since 1997, she has been active as an evaluator and a Project Manager in the area of organisation and leadership. During four of those years she has followed the progress of a collaborative rehabilitation project in Sweden. The interviews made during the 1 year of the project constitute the material for the research reported here.