Coordination of services for dual diagnosis clients in the interface between specialist and community care

Introduction

One of the main challenges for coordination in the health and welfare sector is service provision for people with concurrent severe mental disorders and serious alcohol and/or drug issues (dual diagnosis).¹ The purpose of this article is to study the coordination of services for dual diagnosis clients.

In Norway, services for this group consist of specialist health services in substance abuse, specialist mental health services, various community health and social services, and those offered by general practitioners. This target group has long been regarded as one of the most clinically complex.² The clients show wide variation in issues, needs, and goals, and knowledge of causes and effective therapies for their disorders is seen as uncertain and controversial.^1,3 The outcomes for patients with dual diagnosis are worse than for those with single diagnosis. There are also challenges in the organization of services for this client group.⁴ Accordingly, both treatment and follow-up care become a complex matter. There is currently no combined register of the prevalence of mental health disorders and addiction in Norway. The municipalities have gradually upgraded their services for this target group over the past 20 years and have increased the number of employees involved. In 2017, there were 17.6 full-time equivalents dedicated to patients with severe and prolonged disorders.⁵

There are several factors involved in achieving comprehensive and coordinated services: Who is to collaborate, how they will collaborate, and how the collaboration will be organized.^6–8 The terms coordination and collaboration are often used synonymously, although they have different definitions. Here coordination is to be understood as top-down steering in a hierarchical organization, while collaboration indicates a willingness to collaborate. Such collaboration may take place in professional networks.⁹

This article analyzes experiences of collaboration on this client group from a state hospital and a large municipality in Norway, based on the following question: How are services for dual diagnosis clients coordinated and how appropriate is this coordination?

In Norway, the state is responsible for specialist health services. These services are organized in health trusts. In recent years, as in many other countries, there has been deinstitutionalization of substance abuse and mental health care. The majority of specialist health services are therefore outpatient.¹⁰ The health trust in question had organized its outpatient facilities in two separate clinics, one for substance abuse and one for mental health. An important challenge in outpatient care is that clients must cater for their own basic needs such as income, food, and housing. If they fail to do so, the municipality must provide these services. The municipality must also provide home-based services, such as cleaning and cooking, for those who cannot manage themselves or do not have close relatives who can assist them. Municipalities also distribute medicines in cases where clients cannot be relied on to obtain the medicines themselves. In addition to the treatment provided by the health trust, municipalities also provide various forms of treatment such as supportive interviews and social skills training. The municipality studied here used the purchaser/provider model. This implied that the municipal purchasing office would normally assess each client’s needs for services and make a decision on the services to be provided. The services were provided by various municipal departments. Follow-up care with support interviews and social training was provided by one department, another was responsible for home-based services, and home nursing was responsible for medication management and care outside of normal working hours. For a number of clients, it was therefore a challenge to receive coordinated services.

Research shows that in many cases it is difficult to achieve good coordination in services for dual diagnosis clients. Ådnanes and Steihaug^11,12 describe how clients experience a lack of coordination of services. Hansen¹³ shows that even when different service providers meet in so-called “responsibility teams” (a variety of case management team which often consists of the client and all others involved), the individual providers are more concerned with setting limits for their own interventions than contributing to comprehensive and coordinated service provision. On the other hand, Hansen and Ramsdal¹⁴ show that it is possible to organize meetings between different services to create comprehensive and coordinated services for this client group. However, they also found it challenging that specialist health services were divided into separate organizations for substance abuse and mental disorders. Results from some international and national studies suggest that permanent teams with members from both municipal and state services, such as assertive community treatment teams, have been successful, but other researchers have not been able to demonstrate such a relationship.^15–17

Most evaluations of collaboration models are based on the idea that there is an optimal model regardless of the context, and that clinical results for clients can be used as success criteria for such models.¹⁶ One explanation of the difficulty in proving that one collaboration model is better than another may be that evaluations do not take sufficient account of the fact that different models fit into different contexts. It is therefore necessary to point out some of the factors that should determine the collaboration models to be used.

Different problems, different coordination needs

Firstly, there is a relationship between the collaboration model and the type of issues to be solved. Problems can be described using two extremes on an axis, namely tame and wicked problems.^18,19 Tame problems demonstrate a clear connection between cause and effect and can be isolated and solved separately. Wicked problems, on the other hand, show no clear relationship between cause and effect and no obvious goals; they may involve disagreement about what will help and the problem is often intertwined with other similar problems. Many of the challenges of the welfare state are wicked problems.¹⁸ Concurrent substance abuse and mental health disorders are typical examples of wicked problems as they are clinically complex. There is incomplete knowledge of what causes these disorders and no decisive knowledge about how to solve the issues involved.³ The outcomes for clients with concurrent substance abuse and mental health disorders are uncertain and worse than for those with single disorders.

The kind of problem to be dealt with will affect the need for cooperation. If the problem is tame, with a clear connection between cause and effect, there may be no need for complex forms of coordination. By contrast, if comprehensive services are to be provided to a person with a substance abuse and/or mental health disorder, there may be a need for expertise in substance abuse, mental health care, housing, and social assistance. The need for collaboration will then be much greater than in the case of tame problems.²⁰

The tamer the problem, the simpler the coordination needed, while the more wicked the problem is, the greater the need for comprehensive coordination.

Different ways of organizing coordination

The term coordination often refers to processes where different parts are interrelated, prioritized, and adapted to each other and where collaboration between different service providers is involved. Collaboration may be described in terms of a number of elements, such as structure, process, roles, and relationships.^7,9 Coordination can be organized in different ways, such as using process and network models. In process models, more or less standardized, professionally grounded work processes are established, commonly described as successive measures that regulate a form of patient flow.²¹ This implies a conception of a kind of sequential responsibility and service provision. First, one agency is responsible for making an offer, then another unit takes over, and so forth, until all the issues have been resolved.²¹ If the issues to be resolved can be divided into a number of independent sub-processes, as with tame problems, a process model will often be the most effective way to organize the work. Elements are then placed together in a path which is known to have a lasting effect.²²

Network models are used in case management. A venue is created where people from different services, such as mental health, substance abuse, employment, and housing, meet to create a comprehensive and coordinated service offer.²³ Individual client needs determine who will participate in the collaboration. Case management can therefore be described as a form of matrix organization involving a variety of service providers. The participants collaborate on services for the client, as seen in responsibility teams drawing up individual plans. Responsibility teams can be loosely organized networks meeting a few times a year to collaborate on a comprehensive and coordinated offer of services,^13,24 or they may be more permanent teams such as peripatetic or assertive community treatment teams.¹⁷ In the case of wicked problems, which involve a need for collaboration on making regular adjustments to the services, network models are usually more appropriate than process models.

Several studies have used scales to categorize different forms of network collaboration.^25,26 We have taken Andersson et al’s²⁵ scale from simple to complex forms as our starting point. Information exchange is the simplest form, often involving casual contact between employees from different agencies, but it may also be more formalized, as in the use of different electronic messaging systems. Case coordination usually indicates that an employee of one of the agencies providing services works across agencies to create a common and comprehensive program for a client. Interagency meetings involving staff from different organizations are used to discuss and coordinate measures for common clients. In multidisciplinary teams, employees work together for a long period. Different professional groups contribute different expertise and solutions to the team. A partnership requires a formal agreement between two or more parties. In addition, formal communication and information structures are often established. Co-location can enhance collaboration, since employees are in close physical proximity, and clients will often get “one door” to access services. Pooled budgets are the most complex form of collaboration and generally also include the other forms.

Coordination implies more than different ways to meet and communicate. An important challenge is how to reach agreement on coordinated service provision.¹³ In practice, this means making joint decisions on the content of the services. Abbott²⁷ emphasizes three aspects of professional decisions. The first is to identify problems. Identification and categorization of the information obtained is governed by the professional’s understanding of what is interesting and relevant. Information is linked and the problem is organized and understood according to a professional classification system.²⁷ Involvement of others will provide a basis for collaboration on the exchange of relevant information and on determining what is interesting information. The second aspect is to draw conclusions, analyze, and decide on the best measures. Finally, action or treatment is chosen. Collaboration between different professions and more services on information gathering, mapping, and analysis of needs and implementation of measures is an important factor for achieving more integrated services.^17,28

The abovementioned presentation of factors that affect the quality of service coordination forms the background for the question discussed here: How are services for dual diagnosis clients coordinated and how appropriate is this coordination?

Data and methods

This article is based on data from a study of transitions involved in collaboration on providing comprehensive services for dual diagnosis clients. The study included a project group with participants from primary and specialist health services, organizations for clients and relatives, and researchers from a university college.

Data collection

The data collection took place during a full-day workshop. The aim of the workshop was to elicit the participants’ experiences; we therefore chose qualitative methods for the data collection.²⁹ There were five to six participants from each of the following organizations and agencies: a client organization, a relatives’ organization, a district psychiatric center, a substance abuse clinic, the municipal purchasing office, and the municipal department of service provision (31 informants in total, all interviewed twice). The workshop was divided into two sessions. We chose to use group interviews in both sessions, but with different compositions, as described in more detail in the following section.

Group interviews may make participants feel group pressure to agree and are therefore of limited utility for eliciting controversial viewpoints.³⁰ On the other hand, they are a suitable method to reveal what participants agree on and their common experiences, while different perceptions and experiences will also emerge. Group interviews mobilize and activate participants in a way that is impossible in individual interviews.³¹ Interaction in the groups provides new insights that we cannot access by other methods. Since we aimed to reveal actual challenges in collaboration and understand these, we considered the benefits of allowing the participants jointly to describe the challenges to outweigh the disadvantages. It was also important to us to study collaboration as a common, not an individual, challenge.³¹ Further, experiences could be shared in both homogeneous and heterogeneous groups.

We chose different group compositions for the two sessions of the workshop. In the first session, all participants were divided into six homogeneous groups, each containing every participant from one agency/organization (district psychiatric center, substance abuse clinic, purchasing office, department of service provision, clients, and relatives). The aim was to gain knowledge of how the different groups perceived transitions and collaboration. This approach was based on the assumption that perceptions of collaboration would be influenced by where a person worked and what role he or she had. We were also particularly interested in ensuring that clients and relatives could express their opinions without any objections by service providers. This kind of division into groups that are distinct from one another, but have members with much in common, enabled us to make comparisons across groups.³¹ In the second session, participants were again divided into six groups, but this time the groups were heterogeneous, having only one member from each agency or organization. This combination enabled the different service providers to exchange views across agencies/organizations and thus create different group dynamics from the previous division.³²

Participants were selected by the agencies and organizations themselves. Recruitment to the workshop and the interviews was simultaneous. Criteria and principles for participation were discussed in the project team and emphasis was placed on including staff from different parts of the agencies. The relatives included both parents and siblings.

Analysis

Interviews were recorded digitally and transcribed. The transcripts formed the basis for further analysis and the quotations used in this article. The interviews were conducted in Norwegian. The quotations were translated by a professional translator. To reduce the risk of misinterpretations of the data due to the researchers’ pre-understanding, other researchers discussed in detail the findings and their systematization during the analysis process.³³ The analysis was based on the four steps of systematic text condensation.³⁴ First, we read all the material to gain an overall impression, with a particular focus on the participants’ experience of transitions. Here, the researchers read the transcribed data several times. Then we identified meaning units relevant to the purpose of the study and coordination of services for dual diagnosis clients. These units were processed and coded, and then arranged into code groups. Then we analyzed and condensed the content of each code group. Finally, we contracted the condensed text into an analysis that constitutes our results. Interpretations were discussed in the entire team to maximize the validity of our findings.

We do not intend to generalize, but wish to illustrate some of the challenges of coproduction. The results provide insight into the challenges experienced by relatives regarding service provision for dual diagnosis clients and can give an indication of what will happen in similar situations. There is thus a basis for analytical generalization.³⁵

At the start of the workshop, the participants in the group interviews were informed about the purpose of the project and that they could withdraw at any time. All participants signed a written informed consent. We attempted to describe the findings in such a way that the information could not identify individual participants. The project was reported to and recommended by the Norwegian Centre for Research Data, project number 43638.

Results

In the following section, we present the findings of the study regarding the coordination of services for dual diagnosis clients. The results can be summarized under the following headings: information exchange, appropriateness of the services, coordinators, and responsibility teams.

Information exchange

When specialist health services plan the handover of clients to community care, they inform the municipal purchasing office in writing. Unfortunately, this is not always seen in a positive light. Employees in the specialist health services argued that this was easier before the purchasing offices were established; then, they could contact the local substance abuse consultant directly. Now the process takes longer, because an application has to be made and a decision reached before the case is sent to the consultant. Waiting time may be up to 3 months and during that time, the specialist health services do not have any cooperation with the municipality. It is not uncommon for the specialist health services to have many clients in the same situation, which many employees found tiresome. Even if they specifically request feedback, it is not generally forthcoming from the purchasing office.

Municipal employees were also dissatisfied with the exchange of information, particularly when the specialist health services promise more than the municipality can fulfill, which creates problems for both users and the municipality. It was also felt to be a problem that the specialist health services provide guidelines for the content of the municipal services. One of the municipal employees said:

And sometimes we feel that the hospital has one opinion about the needs, and we have another opinion about the needs. So we’re told that it’s staffed sheltered housing in this case. But we don’t have much of that. And we think that we need to help them where they live and what they need to have an okay life where they are. [...]. Because the thing is, we’ve got nothing to offer. There’s not enough housing. And I find it difficult when psychiatrists and other people who are good at things, right, telling us what to do, and maybe they think a bit differently too [...]. But they aren’t good at what the municipality can do. What a client needs when he comes home to community care, I reckon that’s quite simply what we know best.

Within the specialist health services, it is also difficult to establish cooperation. There was much discussion of the division between substance abuse and mental health services, which seems to be especially evident in specialist services in general and between the social medical outpatient clinic and the district psychiatric center in particular. There was discussion of whether to treat substance abuse or mental health issues first. Clients with concurrent substance abuse and anorexia were mentioned as an example. There is a requirement for simultaneous treatment, but it is highly impracticable. The reason for this is the further requirement that a client must be drug-free in order to receive psychiatric treatment for anorexia. It was stated that some departments of the specialist health services lack knowledge about addiction and what good treatment involves. The following statement illustrates this:

Yes, in some departments, they lack knowledge about addiction and what’s important for good inpatient care, for example. And sometimes I think they haven’t made the effort to learn more about it either.

It has been specified that drug-induced psychosis is a matter for mental health care. But it was questioned whether there may be other reasons for the psychosis in some cases. One group argued as follows:

Specialist health services with substance abuse expertise are involved as needed, but it varies how much this actually happens. Some agencies are seen as more inflexible than others.

Lack of coordination between substance abuse and mental health services created an experience of fragmentation. Relatives talked about experiences where clients were sent between outpatient clinics without any obvious plan or progress. One relative said:

So now that’s over [...] like the first months when she was sent around, she was sent to drug treatment. Then the focus was on the drug treatment. Then they couldn’t see the mental illness. Then they worked according to the substance abuse principles, which are not quite like those in mental health care. Then she got sick and was sent back to mental health. Because then she was too sick to be in substance abuse. And then they couldn’t do any other psychiatric treatment, of course, than stuffing her with dozens of different medications without any plan, and [...] or any assessment or anything.

This suggests that the various agencies do not realize what the other service providers do and what expertise they have. Lack of insight into each other’s expertise may create challenges regarding what to do and who makes decisions. One service provider stated:

One problem is that the different services don’t think in the same way and disagree on the way forward. I think that’s frustrating for clients.

It was seen as challenging that the specialist health services give guidance and that responsibilities are not clarified. The municipality attaches great importance to clarifying what responsibilities it considers it has and has prepared descriptions of the various services. These are intended to curb expectations as to the possible services clients can receive. Municipal employees pointed out that it is also important to describe clearly what the municipality can offer to partners and applicants.

The specialist health service staff reported asking for feedback from the municipality, but stated that they get too little. They were concerned about the flow of information between themselves and the purchasing office and found that cases took too long to process and had too many stages. The following assessment reflects the attitude of many:

I think it’s generally a matter of becoming less formal – there’s too much bureaucracy – I believe that if you have too much bureaucracy, the natural dynamics of communication between people come to a halt. We need good systems for informal contact.

Personal relationships were therefore highlighted by many interviewees as important in enhancing collaboration. This involved knowing the other person, his or her functions and the kind of treatment he or she can provide. Many also emphasized the importance of good communication, which implies that employees needed to:

[…] find channels to each other that work.

However, many felt it might be difficult to find such channels. Whom a person is in contact with and how good that contact is will depend on the individual, since the contact is informal.

It is therefore important to have a broad network of contacts outside one’s own organization. Seminars in one’s own field and with other organizations are central to building networks and relationships with service providers in other agencies or departments than one’s own.

Appropriateness of the services

Dual diagnosis clients show considerable individual variations and different responses to interventions. The purchasing office receives written applications or referrals to assess the needs of clients based on a decision on services. It is difficult to use standardized services or solutions when clients differ greatly. Consequently, in determining the services to be provided, it is challenging to adapt available resources to a client’s perceived needs. The purchasing office staff described how they try to relate to clients and other municipal services as follows:

We find out what we think the user needs. And if we’re a bit uncertain, we cooperate and ask the department of substance abuse and mental health, but we’re the ones that make the decision and we determine the help to be provided.

The first challenge is thus to find out what interventions the client is believed to benefit from. However, when the client’s needs have been ascertained, these are not always compatible with the interventions that the municipality can offer.

For some, we maybe can’t manage to find the right measures, and can’t find good enough services, or they’re not available right now.

Well, yes [...] or some people think it’s not their business to provide that service, or well [...] they think it’s not the service portfolio we have, or can offer now, for different reasons.

Sometimes the municipality lacks a suitable service, or service providers think the most suitable service is not one they are supposed to provide. A particular problem is that some services are only provided at certain opening times.

Because it’s been pointed out that dual diagnosis patients don’t only need help from 8 am to 3.30 pm. They need help 24 hours a day. And that service isn’t arranged like that today.

Service providers said that difficulty in getting hold of people hindered cooperation. It was particularly difficult to get the right people and the right time. The challenge was to find communication and collaboration channels that worked. The informants highlighted the use of coordinators and responsibility teams as collaboration channels.

Coordinators

An important point for clients was that they needed a coordinator who could be with them in transitions between services and over time. The coordinator could also be a lived experience consultant or peer who would still be available even if service providers should leave their job. The importance of coordinating work around individual clients was also emphasized by service providers, one of whom said:

[Individual plans] will have lower priority than coordinators and coordination [...]. If you get a patient, a client and a close partner, you can start to achieve something that works well.

Staff of the purchasing office play a significant role in clients’ transition from specialist to community care. Meetings with the hospital are attended by employees of the purchasing office acting as coordinators.

A coordinator is also key in building a good relationship with clients. One explained it as follows:

And this is a lot about relationships. It really is. If we manage to create good relationships and start working with patients already when they’re in hospital before the transition. And if we’re successful, then it’s often [...]. Then it gets to be an IP [individual plan] and then you know who’s the substance abuse consultant and maybe the patient’s met him even before he’s discharged. These patient groups really need good relationships, otherwise they feel insecure if things are unstable. That’s not how you get your life straight. There must be someone there.

One of the clients concluded the discussion about relationships with the following statement:

Everyone who’s moved on can point to the person who he realized cared about him.

Such an experience of someone taking particular care of them is clearly central to many of the clients’ positive experiences.

Responsibility teams

In some cases, the specialist health services routinely invite others to responsibility team meetings or collaboration meetings. It was clear that much could be done by meeting and talking about a particular client.

[…] experience of a complex patient; similar issues, we’re very good at calling a special responsibility team meeting or collaboration meeting with all the others involved with the patient. So that’s a routine we keep to very closely. And there’s a lot that can be achieved by just meeting the others and talking about the person.

The employees spoke of responsibility teams as central to collaboration.

We tend to have responsibility teams for the different clients.

The substance abuse clinic has a responsibility team meeting before the patient is discharged to bring together the patient and service providers. This shows that responsibility teams have become a common form of collaboration, but that does not necessarily mean that they function as expected or intended. The following quotes would suggest this:

We have responsibility teams, but they’re the client’s teams. There are a great many agencies providing services and when clients only have a passing relationship with some of them, well, they don’t want to have them included in a responsibility team.

The responsibility team is where this could be solved. I also think that the user should own it, but that’s not always possible because the person may be too sick.

This statement suggests a need to renew procedures and working methods associated with IPs and responsibility teams. Responsibility teams are a venue for cooperation, but it requires considerable resources to call and conduct a responsibility team meeting.

Discussion

In the following section, we discuss how services for dual diagnosis clients are coordinated and the appropriateness of this coordination. The presentation is structured under the following headings: information exchange, coordinators, and interagency meeting points.

Information exchange

Much of the communication between the various agencies is in writing and it is partly one way. It may be discussed whether such unilateral information transfer is included in the scale of Andersson et al.²⁵ However, there seems to be good reason to believe that the people who send these messages think that they enhance service coordination. There are also a number of other reasons for sending such written messages. The health service is not merely providing treatment facilities; it also forms a part of general public bureaucracy. This implies that basic bureaucratic principles are followed.³⁶ In practice, it also means that the health service safeguards values such as neutrality and independence through impersonal written communication. In addition, the health service has a statutory duty to document in writing its assessments and the treatment it provides.³⁷

Our data show that written communication between the various specialist health services also creates some problems for coordination. An important factor for developing good collaboration between individuals is good mutual trust.^7,38 The analysis reveals that employees at the two outpatient clinics have stereotypical perceptions of each other as having inadequate knowledge of, respectively, mental health and substance abuse issues. When such stereotypes have been established, people tend to relate to others on the basis of these, in this case that the others do not have the necessary knowledge.⁶ In practice, this can mean a swift rejection of the other’s written arguments based on a perception that he or she does not know anything about the matter.

An interesting question in this context is, of course, why the services are not merged, given the fact that so many need services from both outpatient clinics. The reasons are partly historical; substance abuse clinics were previously not under the specialist health services but part of social services. The main aim was to integrate the substance abuse services in the specialized health care structure, from both a legal and organizational perspective.³⁹ Although more than 10 years have passed since substance abuse became part of the specialist health services, the old division between the substance abuse clinic and the mental health clinic is maintained in several health trusts.¹⁴ One explanation for the division is that health trusts are largely organized according to what are commonly called “medical reductionist principles”.⁴ This means that health care is organized according to medical specialties and diagnoses, not according to patient needs. Fragmentation of the services as a result of this splitting up of the specialist health services is a general challenge.¹¹ The fragmentation can be understood as the result of work being organized and distributed in line with the principles of a process model. Work is divided between the clinics as independent sub-processes and arranged in a linear fashion.²²

Written messages from the specialist health services to the municipal purchasing office are also unsatisfactory. There is every reason to believe that “orders” from the specialist health services regarding the services the municipality should provide are based on a professional assessment of how specialist health care can best be followed up. Yet the purchasing office often rejects these, arguing that the specialist health services are unfamiliar with community services or that the service “ordered” is unavailable. Although the municipality performs its own assessment of suitable services for a client within the framework of those available, it may not capture the same challenges as the specialist health services. It is precisely the fact that the specialist health services and the municipality have different insights into a client’s situation that calls for a joint, broader perspective on suitable services.⁴⁰

In practice, parts of the written assessments are rejected by the recipients. This can be perceived as a kind of defense of one’s own territory, as a form of jurisdiction.^25,27,41 However, the arguments are largely based on perceptions of the others as stereotypes with particular attitudes and knowledge.⁶ Such stereotypes do not become modified by written communications received, but on the other hand, many informants pointed out that the “others” they had got to know personally did not fit into the stereotypes. Developing trusting relationships is an important contribution to weakening stereotypes and territoriality.⁴¹

As we have pointed out earlier, developing effective, coordinated services for dual diagnosis clients involves wicked problems. It requires the exchange of different professional perspectives with a view to creating an optimal basis for decision making.³⁸ If parts of the information exchanged are rejected, the breadth of the basis for decision making will also be reduced and any decision taken will not be based on a diversity of professional principles. There is therefore reason to conclude in our case that the exchange of written information did not provide a basis for developing good coordinated services for clients.

Coordinators

The data show that both clients and service providers emphasized the importance of a coordinator. Several researchers have shown that it is very important for many dual diagnosis clients to have a particular person to whom they can relate and with whom they eventually get such a strong relationship that they can trust the person.^12,42,43 Today, the municipality is obliged to appoint a coordinator for any client who needs coordinated services. Employees of the municipal purchasing office act as coordinators. One of the reasons for having a purchasing office is bureaucratic principles.³⁶ Clients must have a formal decision to indicate the kind of services they are entitled to. This means that written documentation is also important here. Nevertheless, the coordinators at the purchasing office are aware that these clients do not completely fit into regular bureaucratic case processing. It therefore takes time to clarify the kind of services to be provided. During this process, coordinators are in informal contact with some municipal agencies in order to clarify what are the most appropriate services to be offered. The informants described how this involves discussions between different coordinators and different services on possible inputs to a decision on coordinated service provision. Here, we thus have a situation where there is real coordination of different inputs with the aim of providing appropriate services. There is coordination based on information exchange and the work of a coordinator.²⁵

Therefore, some employees of both municipal and specialist health services choose to bypass the formal channels and directly contact other service providers. In this way, they can exchange views and discuss how the service offer is progressing. The problem is that such contacts tend to be random and are generally with those the person knows and already has a good relationship with. The positive nature of such informal contact is underlined by the fact that many of the participants wanted to establish more meeting points between service providers from specialist and community health services to build informal, personal relationships.

Interagency meeting points

The data show that service providers, relatives, and users all had good experience of responsibility teams. At responsibility team meetings, the service providers involved can meet and discuss a common strategy. Here, it is possible to reach agreement on coordinated services and adjust them to the wishes and development of the user.³⁸ However, experiences of responsibility teams were not always positive.¹³ The crucial point is not the existence of such teams, but what they actually do. The negative experiences pointed out by Hansen¹³ were due to the fact that the responsibility team members showed little interest in coordinating services. Our data appear to indicate that the participants were genuinely keen to exchange experiences with a view to achieving coordinated services.

Responsibility teams are one way of organizing a network.²⁰ They provide a real opportunity to exchange data and to conduct a joint analysis. This provides a basis for making decisions about service provision in a way that makes use of the different service providers’ varied knowledge and experience.²⁷ The feedback from the participants indicated that this actually happens. A challenge in this context is that wicked problems also require regular (re)evaluation of the services provided.³ Evaluations conducted by the individual agencies or service providers will be insufficient; the coordination as a whole must also be evaluated.⁴⁴

Clients and, in many cases, relatives should also be strongly involved in the coordination of comprehensive services. In Norway, the concept of co-production has become increasingly important. Co-production is often used as a loose term to cover many different, but related ideas. There is no consensus about what co-production is or implies, but the following definition may be a starting point that seems to cover the use of the term in various contexts that emphasize “delivering public services in an equal and reciprocal relationship between professionals, people using services, their families, and their neighbours”.⁴⁵ Previously, the term co-production was applied to cooperation between clients and service providers. More recently, relatives and NGOs have also been included.⁴⁶ “Shared decision making” is a form of decision-making process that may include both clients and relatives. Storm and Edwards ⁴⁷ describe this as a process of active participation by users, relatives, and service providers, where information sharing and real choice of solutions are key elements.

Our data show that neither information exchange nor a coordinator is sufficient to create shared decision making. Even with good and open exchange of information, the various service providers still have the possibility to make their own decisions independently of the others. The goal should therefore be a process where important decisions are taken during interactions between clients, relatives, and the various service providers, which will represent shared decision making.⁴⁷ Co-production can be a way of involving clients and relatives to achieve more comprehensive services and a more appropriate way to coordinate services for dual diagnosis clients.⁴⁸

Conclusion

Our findings suggest that neither information exchange nor a coordinator function as organized by the municipality or specialist health services in question are arrangements that ensure satisfactory coordination of services for dual diagnosis clients. Coordination is nevertheless possible by establishing informal contact outside of formal arrangements, but this seems to be a random and unsystematic solution. Interagency venues, such as responsibility team meetings, appear to be the best-functioning solution, but this requires a greater degree of networking than today.

The organizational context of the services is also important. Public administration in Norway is largely structured as hierarchical line organizations. Within the framework of such an organization, it is challenging to achieve good collaboration across the organization and between different organizations.

When services are to be provided to clients with complex needs, involving problems that can be described as wicked, there will be a constant need to exchange information about how the various services affect the client. This must be followed by an analysis to clarify whether there is a need to adjust or change the services. To achieve this, it will be necessary to coordinate decisions on the content of the services in joint meetings.

Disclosure

The authors report no conflicts of interest in this work.

References

1.		Hannigan B, Coffey M. Where the wicked problems are: the case of mental health. Health Policy. 2011;101(3):220–227.
2.		Evjen R, Øiern T, Kielland KB. Dobbelt opp: om psykiske lidelser og rusmisbruk [Twice the problem: Mental health disorders and substance abuse]. 3rd ed. Oslo: Universitetsforlaget; 2012. Norwegian.
3.		Einstein S. Substance use(r) treatment and health disparities: some considerations or an askance look at institutionalized substance use(r) intervention disparities. Subst Use Misuse. 2007;42(4):671–686.
4.		Ramsdal H. Styring og organisering av tjenestene [Management and organization of services]. In: Norvoll R, editor. Samfunn og psykisk helse. Samfunnsvitenskapelige perspektiver [Community and mental health. Social science perspectives]. Oslo: Gyldendal Norsk Forlag; 2013:213–242. Norwegian.
5.		Ose SO, Kaspersen SL. Kommunalt psykisk helse-og rusarbeid 2017: Årsverk, kompetanse og innhold i tjenestene [Municipal mental health and substance abuse work. Person years, competencies and content in the services]. Trondheim: SINTEF; 2017. Norwegian.
6.		Reeves S, Lewin S, Espin S, Zwarenstein M. Interprofessional Teamwork for Health and Social Care. Chichester: Wiley-Blackwell; 2010.
7.		San Martín-Rodríguez L, Beaulieu M-D, D’Amour D, Ferrada-Videla M. The determinants of successful collaboration: a review of theoretical and empirical studies. J Interprof Care. 2005;19(Suppl 1):132–147.
8.		Valentijn PP, Schepman SM, Opheij W, Bruijnzeels MA. Understanding integrated care: a comprehensive conceptual framework based on the integrative functions of primary care. Int J Integr Care. 2013; 13(1):e010.
9.		Axelsson R, Axelsson SB. Integration and collaboration in public health – a conceptual framework. Int J Health Plann Manage. 2006;21:75–88.
10.		Rogan T. Et felles løft [A joint effort]. Tidsskrift for psykisk helsearbeid. 2008;5(2):104–110.
11.		Ådnanes M, Steihaug S. Obstacles to continuity of care in young mental health service users’ pathways – an explorative study. Int J Integr Care. 2013;13(3):e031. Norwegian.
12.		Ådnanes M, Steihaug S. “You never know what happens next” – young adult service users’ experience with mental health care and treatment through one year. Int J Integr Care. 2016;16(3):5. Norwegian.
13.		Hansen GV. Samarbeid uten fellesskap: Om individuelle planer i kommunalt psykisk helsearbeid [Collaboration without a feeling of community: individual plans in local mental health care]. Karlstad: Working Life Science, Karlstad University Studies; 2007. Norwegian.
14.		Hansen GV, Ramsdal H. Brukermedvirkning og samarbeid – vanskelige mål å kombinere? [User participation and collaboration: difficult goals to combine?] Tidsskrift for psykisk helsearbeid. 2015;12(3):196–205. Norwegian.
15.		Forsetlund L, Lidal IB, Austvoll-Dahlgren A. Effekter av ulike organiseringstiltak for tjenester innen psykisk helsevern [Effects of organizational interventions on mental health services]. Oslo: Norwegian Knowledge Centre for the Health Services (NOKC). Report from NOKC No. 05-2012; 2012. Norwegian.
16.		Ramsdal H, Hansen GV. The organisation of local mental health services in Norway. Evid Pol. 2017;13(4):605–622.
17.		Burns T. The rise and fall of assertive community treatment? Int Rev Psychiatry. 2010;22(2):130–137.
18.		Vabø M. Dilemmaer i velferdens organisering [Dilemmas in the organisation of welfare]. In: Vabø M, Vabo SI, editors. Velferdens organisering [The organisation of welfare]. Oslo: Oslo University Press; 2014:11–78. Norwegian.
19.		Rittel HW, Webber MM. Dilemmas in a general theory of planning. Policy Sci. 1973;4(2):155–169.
20.		Ferlie E, Fitzgerald L, McGivern G, Dopson S, Bennett CN. Public policy networks and “wicked problems”: a nascent solution? Public Adm. 2011;89(2):307–324.
21.		Ramsdal H, Ramsdal K. Organisatoriske innovasjoner i helsesektoren: prosessbaserte virksomhetsmodeller, evidensbasert praksis og behandlingslinjer [Organizational innovations in the health sector: process-based business models, evidence-based practices and treatment paths]. Halden: Østfold University College; 2007. Norwegian.
22.		Vikkelsø S, Vinge S. Hverdagens arbejde og organisering i sundhedsvæsenet [Everyday work and organization in health care]. Copenhagen: Handelshøjskolens forlag; 2004. Danish.
23.		Onyett S. Case Management in Mental Health. London: Stanley Thornes (Publishers) Ltd; 1998.
24.		Holum LC. Fra idé til virkelighet! Implementering og bruk av individuelle planer i psykisk helsevern - en kvalitativ studie [From idea to reality! The implementation and use of individual plans in mental health care – a qualitative study]. Oslo: Department of Psychology, University of Oslo; 2013. Norwegian.
25.		Andersson J, Ahgren B, Axelsson SB, Eriksson A, Axelsson R. Organizational approaches to collaboration in vocational rehabilitation - an international literature review. Int J Integr Care. 2011;11(4):1–10.
26.		Hudson B, Hardy B, Henwood M, Wistow G. Strategic alliances: working across professional boundaries: primary health care and social care. Public Money and Management. 1997;17(4):25–30.
27.		Abbott A. The System of Professions. Chicago: University of Chicago Press; 1988.
28.		Burns T, Catty J, Wright C. De-constructing home-based care for mental illness: can one identify the effective ingredients? Acta Psychiatr Scand Suppl. 2006(429):33–35.
29.		Danermark B, Ekström M, Jakobsen L, Karlsson JC. Att förklara samhället [Explaining society]. 2nd ed. Lund: Studentlitteratur; 2003. Swedish.
30.		Johannessen A, Tufte PA, Kristoffersen L. Introduksjon til samfunnsvitenskapelig metode [Introduction to social science methodology]. Oslo: Abstrakt; 2010. Norwegian.
31.		Brandt B. Gruppeintervju: perspektiv, relasjoner og kontekst [Group interviews: perspectives, relations and context]. In: Holter H, Kalleberg R, editors. Kvalitative metoder i samfunnsforskning [Qualitative methods in social science research]. 2nd ed. Oslo: Universitetsforlaget; 1996:145–165. Swedish.
32.		Andvig E. Förändran i hermeneutisk forskning – ett exempel från fokusgruppsamtal. [Wondering in hermeneutic research: an example from focus group interviews]. In: Lassenius E, Severinsson E, editors. Hermeneutik i vårdpraxis: det nära, det flyktiga, det dolda. [Hermeneutics in care practice]. Malmö: Gleerups Utbildning AB; 2014:197–209. Swedish.
33.		Malterud K. The art and science of clinical knowledge: evidence beyond measures and numbers. Lancet. 2001;358(9279):397–400.
34.		Malterud K. Systematic text condensation: a strategy for qualitative analysis. Scand J Public Health. 2012;40(8):795–805.
35.		Kvale S. Det kvalitative forskningsintervju [The qualitative research interview]. Oslo: Ad notam Gyldendal; 1997. Danish.
36.		Christensen T, Egeberg M, Larsen HO, Lægreid P, Roness PG. Forvaltning og politikk [Administration and politics]. Oslo: Universitetsforlaget; 2007. Norwegian.
37.		Kjønstad A. Helserett: pasienters og helsearbeideres rettsstilling [Health law: the rights of patients and health workers]. Oslo: Gyldendal akademisk; 2007. Norwegian.
38.		Head BW, Alford J. Wicked problems: implications for public policy and management. Adm Soc. 2015;47(6):711–739.
39.		Nesvaag S, Lie T. The Norwegian substance treatment reform: between new public management and conditions for good practice. Nord Stud Alcohol Dr. 2010;27(6):655–666.
40.		Hudson B, Hardy B, Henwood M, Wistow G. Strategic alliances: working across professional boundaries – primary health care and social care. Public Money Manage. 1997;17(4):25–30.
41.		Axelsson SB, Axelsson R. From territoriality to altruism in interprofessional collaboration and leadership. J Interprof Care. 2009;23(4):320–330.
42.		Kuluski K, Ho JW, Hans PK, Nelson MLA. Community care for people with complex care needs: bridging the gap between health and social care. Int J Integr Care. 2017;17(4):2.
43.		Topor A, Borg M, Girolamo SD, Davidson L. Not just an individual journey: social aspects of recovery. Int J Soc Psychiatry. 2011;57(1):90–99.
44.		Ketelaars CA. Integrated care requires integrated supervision. Int J Integr Care. 2011;11(7):1–11.
45.		Boyle D, Harris M. The Challenge of Co-production. London: New Economics Foundation; 2009.
46.		Askheim OP. Samproduksjon som velferdssektorens kinderegg [Co-production as the kinder egg of the welfare sector]. Tidsskrift for omsorgsforskning. 2016;2(01):24–36. Norwegian.
47.		Storm M, Edwards A. Models of user involvement in the mental health context: intentions and implementation challenges. Psychiatr Q. 2013;84(3):313–327.
48.		Bjørkquist C, Hansen GV. Samproduksjon med pårørende til personer med ROP-lidelser [Co-production with relatives of people with dual diagnosis]. Tidsskrift for omsorgsforskning. 2017;3(3):194–203. Norwegian.