Inter-organizational cooperation : A rehabilitation project based on cooperation between health care and three social service agencies

Purpose: Cooperation between organizations is an often-suggested remedy for handling unsolved borderland problems. However, actual projects aiming at cooperation are seldom very successful. The purpose here is to highlight obstacles related to cooperation between different organizations based on a case study of a rehabilitation project where health care and several social service organizations (social insurance, social welfare, and the local employment agency) were involved. Data were gathered through participation and interviews. Findings: It seems that efficient cooperation requires an understanding of the participating organizations’ differences in work logic as well as work practices. Furthermore, only certain fairly standardized “normal” problems may be handled through organized cooperation while non-routine exceptional problem requires a more fully integrated work organization. Implications: Obstacles to cooperation are highlighted and ways to improve the possibilities of cooperation between organizations are suggested although such possibilities are generally hampered by differences in work logic.


COOPERATION IN THEORY AND PRACTICE 1.Cooperation as Ideology and in Reality
The notion of "cooperation" is here used to denote that the actual work tasks are performed together at the operational level while "coordination" refers to the intellectual planning for the common activities that different people or groups perform-with the intention to combine different competences into an integrated wholeness.
Cooperation is generally regarded as a remedy for coping with most problems at the borderlands between organizations and the notion of cooperation has also become a fashion [1].Whatever the reason, for many years, there has been a general tendency towards uncritical acceptance of increased cooperation between public organizations as a remedy for various problems in society and as something of a universal cure [2][3][4].Within the public sector there has been recurring demands for increased inter-organizational cooperation between health care and various social services and several projects have been launched over the years-although none of them have been very successful they have provided enhanced insights in the obstacles to cooperation.
As a well-known early project of cooperation between primary health care, social services and hospitals [5], it was noted that cultural work logic created some problemsmainly because the "acute care" mind-set of hospital employees tended to take over and hamper the possibility of cooperation with both primary care and service agencies.
Several other projects have been launched and some of the main problems have been identified and summarized in a study by FRISAM [6].The problems described in that study were mainly related to differences in values and goals, lack of flexibility in routines and reward systems, and economic restrictions with a narrow focus on each specific organization.
These problems do not seem to have diminished over the years because later project studies have confirmed that the problems encountered are clashing cultures, such as the cultural differences between providers of medical services and the long-term care services [7] and between physicians and other service providers [8].Other reported problems are lack of role clarity [9], professional self-interest, competing ideologies, lack of mutual trust, and conflicting views about client interests and roles [10].
Reviews to assess current knowledge have been provided [11] as a starting point for such evidence-based decisions that have been advocated by Cookson [12].However, serious doubts about the effectiveness of this approach have been expressed [13].In addition to factual issues, a study by Tubin and Levin-Rozalis [14] shows that in most cooperation projects there is generally also a marked distrustfulness between the participating members from different organizations and that such lack of trust seems to be a major obstacle.
A more recent systematic review by Suter et al. [15] summarizes the current research literature on (obstacles to) cooperation into ten key elements as a conceptual framework.However, the conclusion is that much more needs to be learned about specific structures and mechanisms to be reasonably sure of successful inter-organizational cooperation.

Managerial Level Ideas and Base Level Practice
In most public organizations there are three different institutional systems-professional, administrative, and political.They emanate from different parts of environment and they are governed by different work logics.Political work logic is e.g. based on ideas of fairness in the distribution of resources while administrative work logic is based on ideas of structuring, ordering, and routinizing to make efficient work flows.Professional work logic (at base level) focuses on problem solving and decision processes in relation to individual problems.Most public organizations have to cope with these different work logics and incompatible value systems within the same organizational unit [16].
Agreements on cooperation projects are, generally, negotiated between managers (at a managerial level above base level).At that level there is often a limited knowledge of the detailed work processes at the base level.Therefore, the idea of cooperation may appear as simple and self-evident at managerial level-while the reality of cooperation may not be quite so simple at base levelespecially since the cooperating organizations' different production processes seldom are specified [17].
Public organizations generally work in an institutionalized environment.In such an environment, organizations tend to standardize their structure to conform to social norms, and activities are generally directed to-wards predefined situations and problems [1].In public organizations the range of possible actions is limited, mainly through the demand for uniformity since uniformity in relation to institutionalised norms is the main source of legitimacy.So, in the reality at base level, legitimacy might be expected to become more important than efficiency.
A relatively common phenomenon seems to be that people at the management level in organizations make agreements on coordination in conceptual and abstract terms.Then, later on when people at grass-roots level shall try to implement the practical everyday work tasks it often turns out that various details in the cooperating organizations' structures are opposing each other and inhibit or even prevent cooperation [18].
In cooperation projects there is often some tension between cooperation as an idea and as a practice.At base level, this tension is mainly handled through combinations of reinterpretations of the initial ideas and some marginal adaptation of activities.For example, inhis study, Lindberg (2000) [19] displays how processes are renegotiated and organizational borderlines are changed in order to give the appearance of following the management's intentions.As is pointed out by Blomquist and Jacobsson [20], the actual cooperation is generally turned into continuous negotiations because the preferences and conditions are seldom given in advance but have to be redefined as new situations emerge.To accomplish cooperation in new types of situations involves complex negotiations and need for mutually good intent but also for mutual deference between all the involved parties at base level [14].
So, at the administrative level there is a tendency to regard all problems at base level to be easily solved through standardised procedures while at the actual base level problem-solving requires complex negotiations and good intents from all involved participants.

A CASE STUDY OF COOPERATION IN REHABILITATION OF PATIENTS
The background to the study was that people from the social insurance organization had sporadic meetings with people from the primary health care organization in the catchment area to discuss problem-solving in relation to difficulties in the rehabilitation of some individual cases.At the meetings such cases were brought up that earlier had been subject to various rehabilitating efforts to no avail and now no options seemed to remain.Early retirement was generally suggested as the only possible decision left.These meetings were considered as the end point of the rehabilitation efforts related to the specific case.
To find out if something more constructive could be done, a project was initiated to develop a more systematic cooperation between the four organizations representing primary health care, social insurance, social welfare, and employment agency.The main aim of the project was to prevent social maladjustment and drop-out of patients/clients that had been on long-term sick leave.

Methods
A pilot study was initiated by the county council and in a first step contacts were made between the health care organization and the regional social insurance organization.An investigation started to search for early indications of social maladjustment and drop-out by studying client files at the social insurance organization.Of the clients with long-term sick leave and who were in active rehabilitation were 135 chosen.Their journals were analysed and data compared with corresponding files from primary care (113 journals), social welfare (20 files), and employment agency (35 files).
The main result was that in about half the cases the actual cause for being on sick leave was unknown.In a few cases there were suspicions of simulation but in most cases there was genuine uncertainty about the problem and what could be done about it.Few rehabilitation activities had been undertaken.Lack of knowledge, incompatible treatment ideas, and low priority of certain health problems accounted for most of the inactivity regarding the rehabilitation efforts.There were also indications of organizational problems in the form of complicated routines and bottle-necks in the production that had resulted in long periods of waiting between scheduled rehabilitating activities.
In the majority of cases, the single most important problem seemed to be to obtain an unambiguous clearcut diagnosis and this tended to cause most of the time consumption in the drawn out processes.Actually, the main conclusion of this initial study was that the rehabilitation processes seemed to have contributed to social drop out instead of counteracting it.
A rehabilitation project was started based on this pilot study, and a rehabilitation group was formed with participants from all the four organizations; primary health care, social insurance, social welfare and employment agency.The rehabilitation project aimed at cooperation of the various efforts made in the rehabilitation of people who had been on long term sick leave.It concerns a work situation where the involved technology has relatively low degree of standardization and where there are frequent unforeseen events.
The actual work process in the project was scheduled for preliminary discussions of patients/clients at the beginning of each month.At the following meeting a week later the patient/client was invited and it was decided what steps to take and who would be responsible for the continuous contact with the patient/client.There was also continuous follow-up on all cases in the following week.The immediate changes were that the number of discussed cases doubled and the level of ambition increased.
The proceedings of this project was studied based on participating observation and on interviews with personnel from the four organizations (primary health care, social insurance, social welfare, and employment agency) participating in the rehabilitation project.Another source of information was studies of personal files of patients with long term sick leave.The interviews were mainly open conversations aiming to catch the experience of the entire work situation as a complement to the more limited data gathered from the patient/client files.The interviews were later followed up by further discussions with key personnel to more specifically trace changes and developments in the cooperation project.

Results
The result of the case study illustrates the obstacles and problems related to this kind of inter-organizational project as follows: Improved relations.A noticeable change was that the role of social insurance organization diminished.Instead, the representatives from the primary health care and from the employment agency were often assigned to take responsibility for the further care of the patient/client.As a result of the developed cooperation efforts, the group members found that the contacts between the involved organizations were much facilitated.Furthermore, the group members assured that the earlier suspicion and mistrust between the different organization shad diminished drastically.The tendency to refer patients with diffuse symptoms to specialists diminished and the cooperation was also regarded to have put focus on such patients/clients that earlier had been neglected and put aside to give room for clients with more easily defined problems.
Different views on patients/clients.Eventually it became obvious that health care and social insurance on the one hand and social welfare and employment agency on the other hand, had different views on the relations to the patients/clients in the rehabilitation work.Health care and social insurance representatives were less prone to ask the patients to be present at the meetings.They also wanted to be able to present carefully prepared proposals at the meeting with the patient/client.Generally, they also wished for more specific knowledge regarding the more difficult cases.
Organizational obstacles.Beside the discussions directly related to rehabilitation of patients/clients, organizational hindrances were often discussed at the meetings.
Various problems emanating from different rules and regulations and from the allocation of quotas for allowances were highlighted.The rehabilitation group experienced numerous obstacles related to the rules and regulations at different levels in the planning of the rehabilitation activities.They often had great difficulties in obtaining access to the regular resources within their own organizations-and it was even harder to get individual adaptations and accommodations to rules and regulations when sudden needs occurred.
Throughout the project, the local conditions and restrictions in the rehabilitation work became more obvious.The people in the rehabilitation group became frustrated and wrote several letters to the municipal council.When nothing happened, letters were also sent to the Ministry of Health and Social Affairs and to the Ministry of Labour to attract attention to their problems.
Relations between different levels.Although the project was initiated at management level, after some time dissension became apparent between the management and the project members at base level.The relations between different levels within each organization were also frequently brought up as problems to the rehabilitation group.Group members often complained about management: "they don't understand our specific work conditions" or "they have their models of how work should be organised but those models don't fit here".
Actually, the local social insurance organization came on contra course to its regional principal and a similar situation occurred within the health care organization.The group members were frequently accused to have identified themselves too strongly with their rehabilitation project instead of complying with the general procedures in their permanent organization.Apart from this dissension between management level and base level within the participating organizations, the rehabilitation project was by the group members regarded as a very positive experience.
Project results.There are no clear indications that the changes induced through the project actually resulted in more efficient rehabilitation of patients/clients.Most of the initial problems remained unsolved.Still, the group members seemed to regard the project as successful.
Their explanation for this is that although there were few real changes, the problems had been more clearly identified and some tangible actions had been initiated to address those problems.They said that they felt that something more than just talks had been accomplishedlike e.g. the sending of letters.
The project at large seems to have resulted in an increased feeling of meaningfulness in the rehabilitation process.Actually, the most tangible result of the project seems to be the structuring and translating of "talk" into orderly routines and the establishing of clear roles for the participants.Even if most of the actual activities just were talk, the structuring of it seemed to enhance the feeling of having done something worthwhile.Furthermore, the situation with mutual exchange of views seems to have lead not only to increased sense of meaningfulness but also to hopefulness and increased commitment.

Comments
Cooperation calls for compatible understandings of a common task shared between the cooperating parties-at least to some degree.One way of obtaining such common understanding might be through socialisation.This can for instance be reached through the physical organization of the work together with communication around myths and anecdotes related to the meaning of the work task.In this way, knowledge may develop through ongoing communication and sense-making [21] and as demonstrated by e.g.Stacey [22], meaningfulness is usually created through casual conversation ("small talk") between co-workers and colleagues.Such small talk seems to have led not only to increased sense of meaningfulness but also to hopefulness and increased commitment.
Cooperation between individuals and groups with fundamentally different perceptions of reality, and who have totally different fields of expertise, may result in very different conceptions of purpose and direction.Normally, the overall coordination of such different perceptions of reality is accomplished through the existence of an overall frame of reference in the form of culture [23].But differences might never be acknowledged and can, in principle, remain and possibly be bridged over with the help of a common hope for the future [24] or by means of collective hypocrisy [25].
Sense-making is heavily dependent on the social and interactive climate within the organization, i.e. what kind of work climate exists [26] and the general culture of the organization-i.e. the norms and values developed within the organization [27].
Culture develops an informal organization that includes what should be done and why and who should be contacted when problems arise.A part of the cultural pattern is also the manner in which members deal with each other, who belongs to the group and who does not-i.e. a division into "we, and the others" [28].Included in the cultural pattern are also the particular language and other symbols through which values and standards are expressed in the organization (such as metaphors, legends, myths).In the rehabilitation project there emerged a clear division line between "we and the others" after some time and the courageous action of "writing letters" becoming common legend as a token of commitment.
A part of the cultural pattern is also the issue of how members organize themselves into different groupings which, in turn, may give rise to different subcultures.Lack of understanding and communication between groups in an organization tend to lead to power struggles about ideology and the right to the interpretations and conflicts often emerge between different subgroups about the "true" interpretation of the work at hand [28,29].A power struggle about interpretation typically occurs in the concrete situation because people have different understanding of the specific situation (as a result of different experiences) and therefore often have difficulties to understand each other's viewpoint.Such difficulties are not always a conscious process, but often a result of social segregation into groups of similar thinking and of dissidents.As disagreement grows and animosity between groups becomes manifest, self-organizing processes of gossip and slander generally emerges [22].
Development in an organization requires that learning is supported by the structure and the work climate.Learning, however, is not always positive, or based on the relevant factors or causes.Instead, it is sometimes based on superstition, wishful thinking or cloudy ideology [30].
Regardless of what led to learning within an organization, it normally results in a practice and eventually to a routine.Once a practice has been established, it creates a sense of stability and certainty (i.e., that the organization's members feel that they know what they are doingand why they do it).Simultaneously, an established practice serves as a barrier to innovation because the established routines might e.g.mean a kind of blindness to other, more complex, ways of understanding and handling the task.
To achieve cooperation between multiple parties requires that the participants understand the task at hand in a somewhat similar manner.The prerequisite for cooperation is that the various moments in a work task must be organized in a way that facilitates continuous mutual information exchange.When different ideas about work task emerge, they tend to relatively promptly result in competing routines, often through misunderstandings, mistakes and failures.
The paradox of cooperation.Cooperation projects (such as the rehabilitation project referred to here) are generally started to find new solutions to old problems where neither organization is committed to take on the problems and regard them as belonging to some other organizations responsibility.When a project is started the participating members will have to bridge the gaps between their different competences and organizational variations (and idiosyncrasies).Generally they do it by "talking" until there are some consensus about how to handle the task at hand.
Early routinization is an obvious way to manage cooperation between organizations because they guarantee uniformity in activities and this standardization also serves as legitimizing devices [31].However, this means that production routines obstructs the possibilities to offer individual exceptions, make local adaptations, give personalized treatment, or in general to have a readiness for the unexpected [32].By early routinization it is very difficult to "go outside the box" and this means that although the organization members are supposed to handle new kinds of problems, they end up in routines of what might be the sum of what the participating members find it safe to agree on.Therefore, cooperative projects easily become ill suited to handle the problems they are supposed to solve.It seems that to handle odd and new or unknown problems there is need for actual merger of competences instead of cooperation between organizations.

GENERAL CONCLUSIONS ABOUT COOPERATION
Two main problems might be distinguished in relation to cooperation between the members representing the four organizations; primary health care, social insurance, social welfare, and employment agency.One problem emanates from the differences in work logic at different levels within an organization.The other problem is related to differences in the applied technology in the problem solving processes at base level.

Differences in Work Logic at Different Levels
Public organizations generally apply similar work logic at political and administrative levels.An explanation for this might be that such organizations normally are in an institutional environment where uniformity appears as a prerequisite to the organizations' legitimacy.
Cooperation between public organizations is generally decided on at a relatively high organizational level and the since public organizations tend to have similar work logic at higher levels, negotiations about cooperation are greatly facilitated.
The administrative logic is based on the institutional presumption of uniform handling of clearly defined tasks.Base level activities are therefore generally controlled through directives and various measures of productivity (e.g.number of handled cases and throughput level).This kind of control also seems to work at base level at the social welfare organization.Actually, within the social welfare organization, the administrative logic with its emphasis on rules and regulations tends to dominate at base level as well as on higher levels and the same goes for the social insurance organization, although control of rule adherence does not seem to be equally strictly upheld-at least not in the studied project.
In health care the situation is different since the professional work logic at base level generally assumes that each patient is unique and must be dealt with according to its specific conditions.Activities are mainly evaluated in relation to what help is provided in the individual case.This means that at base level in health care the administrative logic has no great impact as a means of control.Actually, Lip sky [33] states that in organizations where relations between political and administrative control is diffuse there are often marked control problems-especially when there are strong professional groups working at the base level.Such problems were also clearly noticed within health care and social insurance and especially in such situations there is differences between the goals established by the management and the goals adhered to in practice at base level.

Differences in the Applied Technology
Work at base level of client organizations is characterized by a dialogue between organisation and patient/ client.The content of this dialogue is, however, largely depending on the organization's view of the patient/client and this in its turn determines what technology is used in the concrete problem solving.In relation to cooperation, two aspects of technology are of importance.The first aspect is to what degree the encountered problems are regarded as standard or exceptional-i.e. if there is some routine solution available or not [32].The second aspect is the amounts of exceptions to such standard procedures that the organization is prepared to handle.
Amount of exceptions.The second aspect is about to what degree the patients/clients are perceived as unique and in need of more or less tailor-made activities or if they are regarded as largely similar and suited for standardised treatments.At management level there is a general tendency to regard base level work as well-defined and standardised.At base level units there is a general tendency to regard most work as ill-defined and not easily standardised.However, the possibility to act according to this view varies at the different organizations.For instance, physicians working in the primary health care have a much more independent role than the insurance investigator and therefore have greater freedom to act in accordance with patients' individual needs.The insurance investigators are required to make judgements in accordance with established sets of rules and are forced to act in a more uniform way.This uniformity may be somewhat circumvented (as happened in the rehabilitation project) but because all other activities at the welfare organization are performed "according to the book", unorthodox handling of exceptional cases will disturb the general handling routines.Furthermore, an investigator does not have the same individual professional responsi-bility as a physician and is, therefore, more dependent on administrative level rule enforcement.
Normal or exceptional cases.The other aspect of technology is to what extent the patient/client's problem is perceived as truly exceptional.If the patient/client's problem is perceived as self-evident and easily understandable (i.e.textbook case) it may be handled according to well-proven work routines.This means that no extensive search is necessary to find appropriate problem solutions.It also means some reasonable assurance that the applied routines will lead to some predefined results.
However, when the patient/client's problem is perceived as an exception it might e.g.be impossible to find a diagnosis that covers the problems.Then, the individuals' needs might be found incomprehensible.In such situations, extensive search for possible solutions will be needed and there will be major problems to find some common ground for action between the participating members in the project.If actions are undertaken they will be characterised by uncertainty and based on vague knowledge, misty experiences of somewhat similar cases, and intuition.It is unlikely that such activities would be generally accepted by all the participating members in the project-and subgroups with different interpretations of the true mission are likely to emerge.

Organizational Requirements for Cooperation
The study indicates that cooperation may be a means to solve some work tasks, provided that the cooperating parties have a reasonable consensus of the task.Primarily, cooperation may be accomplished in relation to fairly standardised and easily comprehended aspects of the common task.Cooperation is much more doubtful when the task is perceived as non-standard or incomprehensible.
Normal problems.In situations with normal problems there is generally an established sequential relation between activities and therefore it is possible to establish production lines and to use e.g.network planning [34].In such situations, cooperation between organizations may be accomplished through traditional planning of the total rehabilitation program.Then, from the view of the patient/client it would seem like just one single organization.This is also the basic idea behind so called "clinical pathways" constructed for certain groups of medical diagnoses [35].The cooperation efforts needed to handle normal problems is primarily a question of accomplish common activity planning.
Exceptional problems.In cases where the problems are perceived as exceptional, production planning is hardly feasible because the patient/client will then probably be put on a waiting list or referred in various different directions.The net result of such sequential planning may just be that the same patients/clients will reappear for treatment without ever being helped.Because of uncertainty in the problem-solving processes, unforeseen mutual dependences between competences tend to occur and this generally calls for common mutual work efforts.
Implications for cooperation.The conclusion is that normal and exceptional problems need to be handled in different ways, in different organizational settings.There are some studies supporting this view.For instance, in their study Panella et al. [36] found that clinical pathways for normal problems appeared to be effective in reducing unnecessary variations and improving outcomes and quality of care.However, the implementation of the clinical pathways for exceptional problems had to be discontinued because the pathways were inadequate.In another study by Sydney et al. [37] it was noted that preplanned guidelines only were useful in relation to well defined situations.
The main implication of this view is that cooperation efforts need to be organized in different ways depending on the perception of the problem.In situations with normal problems, there are mainly sequential dependences between activities and cooperation may be accomplished through planning conferences and other temporary meetings where activities are decided on.The patient/client is then transferred between the different activity centres according to the plan.
In situations with exceptional problems traditional cooperation through common planning is hardly enough.Such problems put demands on professional development where activities are evaluated according to the amount of knowledge gained and not just the number of cases taken care of.A probable mutual dependence between activities puts demand on some more elaborate form of cooperation than the traditional planning conference.An example may be to organize problem solving groups as joint ventures or some similar continuous work group with common responsibility for the activities and results-i.e.what Scaeffer and Loveridge [38] defines as true partnerships.
A final remark.Most hospitals are at the administrative level regarded as a single entity-one organization.However, at the base level a hospital might better be described as a conglomerate of several small organizations developed from specialties mainly using various treatment methods.Different departments within a hospital have differences in technology and work logic and have developed different cultural traits in pretty much the same way as the cooperating organizations described in this article.This means that similar problems might be encountered in attempts to cooperation between different departments within a hospital.