Introduction
The provision of mental health services is highly entrenched in political and moral disputes. Hence, service users may be portrayed as policy victims abandoned by society but also as policy risks seen as being unpredictable and violent. Nevertheless, the body of research focusing on policy issues in this field is surprisingly small. Previous research has primarily approached the mental health field as a medical issue, or alternatively, in sociological terms as a matter of social control (Maycraft Kall, Reference Maycraft Kall2010). This article discusses the domestic implementation of a policy idea that has diffused over the industrialised world during the last couple of decades – compulsory community care (CCC)Footnote 1 for people with mental illness. The case we are investigating is its application in Sweden, a country where local adoption of policies is conditioned by a high degree of independence for the local and regional authorities that provide social and psychiatric services.
Since mental illness is a complex phenomenon affecting numerous aspects of the individual's life, an equally complex array of organisations might be involved in addressing his/her problems. Given this complexity, it is not surprising that the coordination of various kinds of treatments and supports provided by different providers has been identified as a major challenge in mental health (McDaid et al., Reference McDaid, Oliveira, Jurczak, Knapp and Group2007; Melke, Reference Melke2010; Shorter, Reference Shorter, Knapp, McDaid, Mossialos and Thornicroft2007). In Sweden, this problem is expressed in the interface between the three major agents: state (national), health care (regional) and social services (municipal). The division of responsibilities between these agents has been a long-standing problem. During the era of the large institutions, the overarching responsibilities for patients rested largely with the state. When the de-institutionalisation process began, the responsibility for medical care was transferred to the regional health authorities (county councils), and treatment provided at general hospitals (Markström et al., Reference Markström, Sandlund and Lindqvist2004). The number of hospital beds consistently increased until the 1960s. In the 1970s emphasis began to shift towards outpatient care and community services, which resulted in a reduction of hospital beds that has continued into this millennium. This development in Sweden is representative of a wider international policy trend (Shorter, Reference Shorter, Knapp, McDaid, Mossialos and Thornicroft2007). This shift in ideology and responsibilities has produced a number of new welfare agents and professional groups. The added complexity has further contributed to the problems of coordination of services discussed above (Goodwin, Reference Goodwin1997).
Maycraft Kall (Reference Maycraft Kall2010) depicts the main strategies for government and steering of the mental health field in Sweden as ‘soft and loose.’ The constitution guarantees local authorities a high level of autonomy. When the government wants to implement nationwide policies, its regulatory steering is typically accomplished softly through so called frame legislation. Financial steering is accomplished indirectly through financial incentives (seeding money) and information steering is effected through mild persuasion forms based on learning strategies (e.g. guidelines, conferences, education, reports). Compared to international practices, the Swedish strategy emphasizes ‘carrots’ and ‘sermons,’ rather than ‘sticks.’ This means that local agencies are granted vast freedoms and discretion to formulate implementation strategies. Consequently, there is variation in both access to and quality of services. In response to the problem of inter-organisational co-operation, the Swedish government has primarily attempted to instigate frame legislation to clarify the responsibilities of both regional health care authorities and municipal social service authorities. Largely, the regional mental health care services (hereinafter called psychiatric services) are to provide for psychiatric treatment of mental illness while community mental health services within municipal social service organisations (hereinafter called social services) are assigned social support, housing, providing daily activities, and so on (Maycraft Kall, Reference Maycraft Kall2010). Central policy initiatives thus need to take into account how the responsibility for people with mental disability is distributed between different institutions which also cater for different catchment areas. Despite repeated efforts to clarify boundaries, there are still contested areas with regard to division of labor between authorities. Two prolific examples are rehabilitation services and support for those who suffer from both mental illness and drug/alcohol abuse (Rosenberg, Reference Rosenberg2009), both of which are relevant to compulsory community care.
Since the early 1990s, CCC has been introduced in most western democracies. The concept has often been controversial. Proponents of CCC have argued that this is a so-called ‘least restrictive alternative’ to inpatient compulsory care, and that it facilitates adaptation to community life while at the same time protecting the public from potentially dangerous individuals. Others have pointed out the risk of net-widening (i.e. more people might be subjected to coercion) and the risk of human rights abuse and how coercive practices in people's homes may constitute new ethical costs (Sjöström et al., Reference Sjöström, Zetterberg and Markström2011). In Sweden, CCC can only be initiated upon discharge from hospital-based compulsory care, and similar rules apply to forensic patients and those whose compulsory care is motivated by need-of-care.Footnote 2 Each patient under CCC is subjected to certain ‘special provisions.’ A study of all court decisions in Sweden (541 decisions regarding 516 non-forensic patients) during a six-month period in 2009 found that the most commonly applied provisions were: medication (79 per cent), staying in touch with community mental health services (51 per cent) and/or social services (27 per cent), staying in designated housing arrangements (25 per cent), and abstaining from drugs or alcohol (15 per cent) (Zetterberg et al., Reference Zetterberg, Sjöström and Markström2014). The Swedish CCC regulations have been described as toothless, since there are no powers to force patients to actually comply with special provisions. Decisions about CCC are made by an administrative court after an application from a treating psychiatrist. The treating psychiatrist must submit a coordinated care plan detailing the services and interventions provided by both psychiatric services and social services respectively (Proposition 2007/08:70). In an evaluation report, the National Board of Health and Welfare concluded that the reform was successful in targeting the group of patients that the legislators were concerned with, i.e. those with long-term problems and repeated hospitalizations for whom voluntary care has failed to prevent relapse (Socialstyrelsen, 2010). The reform appears not to have resulted in so-called net-widening, ie it hasn't recruited new patient groups to compulsory care. The report did not address outcomes for patients in terms of improved health or reduced need for hospitalization.
A notable feature of the new form of legal coercion that CCC comprises is how the major decision-making responsibilities rest within the psychiatric services, while the provision of everyday services is largely executed by social service organisations. This article aims to investigate what happens when a new policy is introduced that assumes complex co-operation of two organisations with high degrees of discretion. On a local level, what happens in and between these organisations as they begin to apply the new legislation? To understand such processes, we adopt a translation perspective. The process of institutionalisation of CCC will be analysed in terms of how ideas are collectively translated, legitimized and reproduced into particular institutional contexts.Footnote 3
Translating ideas
We adopt a constructivist form of institutional theory that treats organisational change as the travel of ideas. Latour (Reference Latour and Law1986) has argued that the notion of translation offers a more dynamic understanding than the otherwise often-used term diffusion (cf. Callon, Reference Callon and Law1986). Drawing from Czarniawska and Sevón (Reference Czarniawska, Sevón and Czarniawska1996), we are concerned with the process – rather than the results – of institutionalisation. Ideas are dispersed through an on-going process of translation in which reality is created and recreated. Czarniawska-Joerges (Reference Czarniawska-Joerges and Czarniawska1996) argues that the inherent content of an idea and the problems it is supposed to solve should not be the starting point of analysis. Instead, she shifts the focus to how the problem is presented, understood and negotiated. The fit between the idea and the problem is a result of the translation process. Ideas do not contain any internal energy – they require purposeful agents that transfer them in time and space to active recipients. The momentum in the dispersion of an idea is thus not contingent on the idea itself, but rather on how a network of people interprets and promotes ideas in terms of solutions to problems. In this matter, the dispersion of ideas will always be a process of a collective accomplishment (Czarniawska and Sevón, Reference Czarniawska, Sevón, Czarniawska and Sevón2005; Erlingsdottir, Reference Erlingsdottir1999; Johnson and Hagström, Reference Johnson and Hagström2005). When an idea travels to a new location, its content will be affected –positioning something into a new context always involves recreation (Czarniawska and Sevón, Reference Czarniawska, Sevón, Czarniawska and Sevón2005). However, at the same time people, organisations and actions will be affected by their exposure to the idea (Erlingsdottir, Reference Erlingsdottir1999).
When applying this translation perspective on empirical data, focus will be directed at a particular idea and the process in which it is potentially institutionalised (Erlingsdottir, Reference Erlingsdottir1999). In our case, we will investigate if and how CCC is institutionalised in Swedish social service and health care authorities. The journey that leads towards the institutionalisation of an idea can be arranged into five phases. In the first phase, the idea is disembedded when it is brought from one context to another. Since behaviors, institutions and practices are stationary, it is necessary to isolate core components of the idea from the organisational context in which it originates (Czarniawska and Sevón, Reference Czarniawska, Sevón, Czarniawska and Sevón2005; Sahlin-Andersson, Reference Sahlin-Andersson and Czarniawska1996). In the second phase the idea is packaged in terms of models, ideals or prototypes. This will facilitate the third phase where the idea travels in time and space in order to be implemented in the fourth phase and materialised in conjunction with existing practices (Johnson and Hagström, Reference Johnson and Hagström2005). Finally, the idea may become institutionalised as a taken-for-granted feature of the practices in the new organisational environment. However, this does not necessarily represent the endpoint. Translation activities will continue to reproduce and possibly reform ideas into new models and practices (Czarniawska-Joerges, Reference Czarniawska-Joerges and Czarniawska1996; Erlingsdottir, Reference Erlingsdottir1999; Sahlin-Andersson, Reference Sahlin-Andersson and Czarniawska1996).
Two prerequisites need to be satisfied before an idea can be implemented into a new organisational context. First, the timing needs to be right, and second, the idea has to be perceived as normatively sound. Ideas consistent with prevailing master ideas – professional norms that fit into institutional modes of action – are more likely to be accepted (Erlingsdottir, Reference Erlingsdottir1999; Ponnert and Svensson, Reference Ponnert and Svensson2011). Master ideas are fundamental and taken-for-granted in an organisation and, as such, vital for organisational identity (Czarniawska-Joerges, Reference Czarniawska-Joerges and Czarniawska1996). Organisational identity is defined by comparing, referencing and imitating other organisations (Sahlin-Andersson, Reference Sahlin-Andersson and Czarniawska1996).
Our ambition is to analyse how the idea of compulsory community care is translated in light of the master ideas, agents and practices in health care and social service organisations. Sweden provides a case that is particular fruitful for such analysis due to its tradition of soft governance and the high degree of discretion enjoyed by municipalities and county councils. When the same legislation is to be implemented in two different types of organisation, it becomes possible to gain further understanding into how organisational features impact on the implementation process.
Methods and material
The translation process is central to our theoretical approach, and here analysed by studying three different locations in Sweden where we have interviewed key informants. The three locations were purposively selected to capture variation in organisational conditions that are characteristic of Swedish mental health services such as scale, how mental health services are organised within municipalities, and how psychiatric and social services are organised in relation to each other (cf. Ritchie et al., Reference Ritchie, Lewis, Elam, Ritchie and Lewis2011). These considerations resulted in the selection of one metropolitan area, one mid-sized municipality, and one cluster of neighbouring small municipalities within a single region. These locations were situated in different parts of Sweden. Our goal is to cast light on how organisational conditions impinge upon processes of implementation.
Metropolis has a population of about 300,000 and is the centre of the region responsible for psychiatric services. The municipal social services are divided into ten administrative districts, each responsible for community mental health services, among other things. There is a large university hospital in Metropolis, with a general psychiatric clinic, a forensic clinic and a number of outpatient teams. Middletown, with a population of about 120,000, is the main city in a large, scarcely populated region. The community mental health services are managed by a specialised branch within the municipal social service organisation. Just as is in Metropolis, there is a large university hospital, with similar services. Smalltowns consists of four small municipalities (population ranging from 15,000 to 42,000) within the same region, each with its own social service organisation. In one of the municipalities, there is a small hospital with a psychiatric clinic including outpatient teams. For two of the others, the hospital has local psychiatric outpatient units that also care for patients in the fourth municipality.
We have chosen to use qualitative interviews with key informants to capture the nuances and active interpretative aspects of how ideas are translated. We were targeting people in key positions holding a rich amount of information, and those individuals were easy to identify in the organisations. All interviewees had some sort of managing or coordinating position. Most respondents from psychiatric services also had some direct involvement in clinical work with patients. The interviews were semi-structured and grounded in a thematic guide that was slightly adapted to the position of each respondent and the nature of her/his organisation. A total of 15 interviews were conducted – five in each location. Eight respondents were employed in psychiatric services (six psychiatrists, one nurse and one social worker), and seven in social services (all social workers). Unfortunately, it was not possible to obtain anyone to interview in one of the small municipalities. Most interviews were undertaken at the office of the informant, except for three that were carried out via telephone. The length of interviews ranged from 35 to 65 minutes, and they were audio-recorded and subsequently transcribed verbatim. Participation was voluntary, and no individual patients were discussed. The project has been approved by the Regional Ethical Review Board in Umeå.
Data have been analysed by means of directed content analysis with the aim of validating or conceptually extending a theoretical framework or theory – in our case a translation perspective on organisational change (cf. Hsieh and Shannon, Reference Hsieh and Shannon2005). In the first stage, data have been organised according to the five phases of translation described in the previous section. Within each phase, data are discussed in terms of theoretical notions such as the matching between idea and problem, legitimating an idea, organisational identity, master ideas and the materialisation of ideas.
Results and analysis
Phases 1 and 2: Disembedding and packaging the idea
The first phases of the translation of CCC primarily occurred on a national level. Before turning to the interviews that mainly deal with the later phases, this section discusses how the idea of CCC was disembedded and packaged before it reached the local levels.
Compulsory community care was enacted in Sweden in 2008, following a period of intense critique of mental health services. The spark that ignited this debate was a series of five incidents where persons with mental illness had acted violently in public. These events attracted considerable attention in the mass media, especially the case where Sweden's minister of foreign affairs was killed in an attack at a department store in Stockholm. A prominent theme of the debate in the aftermath of these events was a sense of the failures of community mental health. The Swedish government appointed a National Coordinator of Mental Health to investigate the problems. The coordinator was provided with extensive fiscal resources and a mandate to propose a wide array of new policies (Nationell psykiatrisamordning, 2006). One of the proposals from the national coordinator, that was eventually passed by parliament, was the introduction of CCC which was partly inspired by an international trend whereby a large number of countries had introduced the concept. In the bill where the government laid out its proposal, the stated intention was to facilitate the return of hospitalized patients into community life, but also to better protect the public from forensic patients after discharge. The new proposal also introduced mandatory coordinated care plans to improve the collaboration between psychiatric services and social services. Finally, there was an intention to improve the legal protection and rights of patients compared to the previous practice where patients were given long-term leave from compulsory hospital care while living in the community (Kjellin and Pelto-Piri, Reference Kjellin and Pelto-Piri2014; Sjöström et al., Reference Sjöström, Zetterberg and Markström2011).
The proposals above thus represent the packaging of the CCC idea. The package itself can be seen as the new legal regulation together with the rationale behind its specific design as described in the introduction of this article. But the packaging process also involves the actors that convey the founding ideas to practitioners as well as the steering forms applied by the government. In our case, the State operated through the National Board of Health and Welfare, which organised information activities, arranged regional level seminars and provisioned guidelines. The government also dispersed seed money to the local service providers. Typically such resources were dispersed without clear guidelines about their specific use, which left considerable leeway in how the final wrapping of the package was realised locally.
Phase 3: The idea travels to the practice organisations
During the third stage toward the possible institutionalisation of CCC, the idea was dispersed into the practice organisations where it would be realised. The meaning of the idea was contingent on the relationships between national promoters, local promoters and the practitioners who were the ultimate recipients of the dispersion of the idea. Our focus in this article has been on the local translators, a group that was crucial in how the idea ultimately would be received, accepted and materialised (Czarniawska-Joerges, Reference Czarniawska-Joerges and Czarniawska1996). Within the psychiatric services, local promoters were easy to identify. They served a relatively distinct function connected to managerial responsibilities. In contrast, local promoters within social service organisations had less formal authority within their organisation. This reflects how the idea was originally designed, where the crucial aspects of decision-making were put in the hands of psychiatrists.
The idea and the problem
The translation process needs to be understood in terms of how the idea was presented, negotiated and a problem constructed (Czarniawska-Joerges, Reference Czarniawska-Joerges and Czarniawska1996). The interviewees described a variety of opinions about the idea as well as the problem. The meanings attached to the idea had also changed during the process. Most interviewees first heard about the new regulations in informal contexts, such as chatting with colleagues. Prior to the introduction the idea seemed attractive to most:
I could see that it was a means to shorten the long-terms leaves, a way for people to remain in the system while also being out. That would be a more acceptable way of life.
(Social services, Smalltown)
Generally, I was in favour of this. I don't think you can rehabilitate someone into a life at home from the hospital. You have to offer rehabilitation in the person's natural environment.
(Psychiatric services, Metropolis)
When the interviewees first heard of the new proposals, they primarily saw it as an opportunity for patients who otherwise would have been locked up in hospital to be in the community. In this phase, most of the interviewees thus welcomed CCC as a ‘least restrictive alternative’ to compulsory hospital care.
The National Board of Health and Welfare was the most important national promoter of the idea and arranged a number of local and regional seminars and conferences. Most of our interviewees reported that they had taken part in such activities. Some of them were then responsible for the further dispersion of the idea through various educational activities within their own local organisations. Some of the seed money provided by the state was used for this purpose. As more knowledge was gained on the local level, the interpretations of the idea changed. At the point when CCC was enacted, a division of opinion between social services and psychiatric services had opened up. Within psychiatric services, the initial positive attitude was partly replaced with concerns about the practical applicability of CCC. In fact, the law did not provide any coercive measures within the community context, which staff in psychiatry saw as a problem.
My spontaneous reaction was that the law lacked teeth. (Psychiatric services, Middletown)
The first reaction was: ‘How are we going to use this?’ You know, the law had no edge /. . ./ so there were questions about the potential of the new legislation. (Psychiatric services, Middletown)
According to our interviewees, staff within psychiatric services were concerned with the lack of coercive powers. However, within social services there were the opposite reservations that the new legislation would entail more coercion. In the municipal settings, the new form of coercion collided with the conception of social services as firmly grounded in values about voluntariness:
All our interventions and services are based on free choice, but people [on CCC] know that there are degrees of freedom. They accept help even if they don't want to. (Social services, Middletown)
At this stage, the evaluation of the idea went from positive to negative, but for different reasons depending on the organisation. Within psychiatric services, it is a taken-for-granted master idea that coercion is sometimes necessary. Interviewees from psychiatric services did not thus see it as problematic per se that new forms of coercive practices were introduced. On the contrary, they were concerned about limitations in applying coercion. A contrasting master idea within social services stipulates that all services and interventions should be occasioned by a voluntary request from a service user.
The fitting between an idea and a problem is likewise dependent on organisational context. The interviewees from psychiatric services portrayed the new form of coercive care as a means to increase legal security on behalf of patients. They described how the previous practice of putting involuntary hospital patients on long-term leave could be in violation of the law, and one psychiatrist saw CCC as a way to avoid ‘stretching the limits of the law’ (Psychiatric services, Smalltown). It is unsurprising that respondents from social services did not mention this aspect, since the legal responsibility for CCC primarily rests within the psychiatric services – this is simply not a problem within the social services organisation.
In social services, the reform was expected to entail more substantive changes than perceived within psychiatry. The new involvement in coercive care was not seen as a change for the better.
There were sort of general concerns, and people talked a lot about this half-way form of care. How should social services solve this? And should social services really pay for this? It was as if social services would take over responsibility from psychiatry, like now they are putting all this on us? And do we really get compensated for this? (Social services, Metropolis)
The concern about new responsibilities and more expenses for a new group of clients was common in the interviews. Fundamental master ideas within social services were thus challenged, which in turn could be seen as a threat to organisational identity. It is therefore not surprising that the idea gained less legitimacy in social services than in psychiatric services.
Phases 4 and 5: Implementation and institutionalisation
During the fourth phase of an idea's journey from conception to institutionalisation, it is established in local practice. Before it can be realised in practice the idea has to be materialised and reproduced in the local context (Czarniawska and Sevón, Reference Czarniawska, Sevón, Czarniawska and Sevón2005). After the mixed reactions in the previous phases, the law was enacted on 1 September 2008, and the implementation phase began. At the time of the interviews in 2012, the law had thus been in place for about four years.
Legitimation through the template form for a coordinated care plan
When a psychiatrist submits an application for CCC, the law stipulates that a coordinated care plan is attached. This coordinated care plan – manifested as a physical object – appeared to be vital in how CCC was materialised according to the interviewees from both types of organisations. Apart from the educational activities – with aims that were often perceived as vague – it was the very design of the template form for the coordinated care plan that induced interaction and meetings between social and psychiatric services. This became a common objective around which a shared understanding could be formed. The need for a template form was identified early by the respondents. The collaboration to design a template was partly prompted by administrative needs. The existing digital information systems for clinical record keeping and administration of services did not provide a function for coordinated care planning. For many of our interviewees, the task of designing this template was the core feature of the practical preparations before the law was enacted.
The interviewees often talked about the template form as almost synonymous to CCC. A well-designed template form was depicted as evidence of good practice. Conversely, failures to co-operate or other shortcomings were often attributed to deficiencies in the template form:
Currently, it doesn't say in the template whom to contact when need arises, but we're going to fix that in the fall. (Social services, Smalltown)
In the respondents' discussions of advantages and failures, the template for the coordinated care plan became the materialised idea of CCC.
Degrees of institutionalisation
The respondents almost unanimously stated that the new legislation had become an integrated part of local practice at the time of the interviews, and that it rarely caused much discussion. It had become routine for psychiatric services to consider CCC when patients under compulsory care were to be discharged from the hospital. It was seen as particularly suitable for patients with psychosis that had been known to the clinic for a long time. However, in two of the smaller municipalities, CCC did not appear to have been institutionalised at all. Prior to the introduction, representatives for these social service organisations did attend a few information meetings, but no initiatives were made to prepare for any new practices. At the time of the interview, CCC was rarely mentioned and the occasional client who had been subjected to CCC had not been managed differently from other clients. In contrast, one of the neighbouring municipalities had developed special procedures in relation to CCC. In the small municipality where CCC had reached the highest degree of institutionalisation, the seed money was mostly used to fund the salary of one person working half-time with implementing the new form of coercion. This person took part in several courses, went on to give courses herself to social service staff, and continued to be an active promotor of the idea of CCC.
Clarifying responsibilities
Respondents from most organisations claimed that the idea of CCC had actually generated new patterns of practice. Two features recur in the interviews. First, the requirements of a coordinated care plan had contributed to the parties clarifying their responsibilities:
It contributes to becoming clearer; you get a better structure since you have to come up with a coordinated care plan. (Psychiatric services, Metropolis)
It clarifies things. You draw up a care plan that is more clear than before, and then it's clearer for the person what to consider, and you know who is responsible for what. (Social services, Metropolis)
Depending on master ideas and organisational context, there were different interpretations of what problems were solved due to the clarification of responsibilities. Respondents from psychiatric services highlighted how it had relieved them from fiscal responsibilities.
Before, when patients were on temporary leave, most often the clinic had to pay. But with the new law it became clearer that the municipality was responsible for daily activities and housing /. . ./ I think that municipalities might have felt that this meant more costs for them. Although from our point of view at psychiatric services, we thought that they had that responsibility all the time, even though we used to pay because patients were formally on temporary leave. (Psychiatric services, Metropolis)
Another recurring theme was the approach to the principle that social services should be voluntary. Several respondents from psychiatric services complained about how municipalities required that patients file applications for services that were part of their special provisions. One psychiatrist reproached social services for hiding behind the notion of voluntariness to get away from their responsibility for this group of clients. Respondents from social services were aware of such criticism, but argued that it was based on a misunderstanding and lack of knowledge regarding the complexities of their work.
Psychiatric services sometimes struggle to understand this – how we work under different frameworks. Because if the person doesn't want to embrace us, it doesn't matter how much we insist or encourage them, because it's a voluntary service. (Social services, Middletown)
Both parties thus acknowledged that the coordinated care plan contributed to clarifying responsibilities, but for slightly different reasons. Cognisant of the expectations that the new legislation would primarily bring more changes to social services, it is not surprising that respondents were concerned with these issues.
Aside from clarifying the division of responsibilities, most respondents maintained that there had been improvements regarding coordination and shared responsibilities.
On the whole I think it's great and it also forces the coordinated planning for patients. It's a good thing that it enforces a mutual discussion and consensus about patients. This means there is this mutual responsibility for patients, hopefully. (Psychiatric services, Metropolis)
Apparently, the idea of compulsory care had materialised as a template form for coordinated care planning, which ultimately helped clarify responsibilities and improved coordination between the two types of organisations. At this stage, the institutionalised idea became a solution to existing organisational problems.
Some respondents also pointed to the effects this had on patients in general, including voluntary patients, who were not targeted in the legislation.
This wrestling game between two authorities and two approaches with disagreement on who does what is a consistent problem. It does improve from year to year, but it is nevertheless a problem. But having CCC has changed the approach for the better. (Psychiatric services, Middletown)
Thus the idea had effects that extended to other parts of the organisations where they became institutionalised (cf. Erlingsdottir, Reference Erlingsdottir1999). A contributing factor was the training programmes that staff from psychiatric and social services jointly took part in. These became opportunities to get to know each other and also learn about the other party's methods and organisation. As a manager from social services in Metropolis put it, ‘You don't engage in warfare with someone you know.’ According to respondents, CCC also forced service providers to agree on practical solutions that carried over to other groups of clients as well. Due to CCC implementation, the organisational infrastructure was adapted to facilitate coordination. One example of this is their agreeing on using only one fax number for sending documents about the planning process.
Paradoxically, the progress in co-operation may actually discourage psychiatrists from initiating CCC. Two respondents from psychiatric services reported that since the extra paperwork had become a requirement, it sometimes made them refrain from filing an application for CCC.
Discussion
The aim of this article has been to investigate what happens when a new policy is introduced that assumes complex co-operation between two organisations with high degrees of discretion. The interviews were done about four years after the law was enacted, which, to some extent, might have affected respondents’ memories regarding the implementation process. However, the distance provided also made it possible to capture the whole process as well as allowing respondents to be more reflective about the process. Although results cannot be generalized in a statistical sense, the chosen locations taken together represent typical conditions for CCC in Sweden. The analysis here is subjected to theoretical generalisation (Flick, Reference Flick, Flick, von Kardoff and Steinke2005), particularly in understanding how interpretations and translations of policy initiatives are constructed under the particular conditions.
We have shown how the idea of compulsory community care has made a journey in time and space to be implemented and materialised. The results can be summarized in three main findings. The first main finding concerns the role that the template form for the coordinated care plan has played in the implementation process. It was the very design of the template form that induced interaction and meetings between social and psychiatric services in the process and this became a common objective around which a shared understanding could be formed. This coordinated care plan – manifested as a physical object – appeared to be vital in how CCC was materialised and the interviewees often talked about the template form as almost synonymous with CCC. The second main finding highlights how the idea of compulsory care helped clarify responsibilities and improved coordination between social services and psychiatric services. CCC has helped professionals from both types of organisations to clarify their respective responsibilities but depending on organisational affiliation, respondents somewhat differed in detailing the improvements. Here, the translation perspective has helped pinpoint how differences between psychiatric and social services are contingent on organisational context. However, the manner in which CCC was interpreted and put into practice also brought about changes that extended beyond services for the targeted group of service users, which is our third main finding. Respondents described how there has been a general improvement in collaboration. The implementation of the new policy coerced service providers to agree on practical solutions that carried over to other groups of clients as well, as the organisational infrastructure was adapted to facilitate coordination. One contributing factor to this spillover effect was the joint training programmes that were offered to social and psychiatric service providers. These encounters brought professionals closer to each other and created a better understanding of the approaches to and conditions for the other party's work. In essence, respondents described CCC as a pedagogical reform to promote the coordination of services.
During the implementation process, a policy about coercing patients has thus been translated into inter-organisational collaboration. The shared interpretation and materialisation of the idea as a template form could perhaps be seen as a logical consequence of the conditions at hand and the structure and organisational patterns that public organisations such as social and psychiatric services have in common. The shift from coercion into collaboration was a process affected by a number of factors. In their first encounter with the new idea, the interviewees often portrayed it as a reduction of coercion, despite the fact that CCC in legal terms actually expanded the scope of coercive powers to new groups of patients. This is most likely to be a result of how the idea was packaged, the stated purpose being to facilitate the transfer to community life and to improve the legal protection and rights of patients. In this respect, it is revealing that the government's bill was given the title ‘A new form of care’. This parceling shifts the focus away from the coercive nature of the reform. Another factor that enables the downplaying of coercion is the steering strategies involved. The Swedish state applied soft steering strategies to municipal and regional authorities bestowed with a large degree of discretion in solving tasks assigned to them. The soft governance and high degree of organisational discretion are likely to have contributed to the relative lack of conflict during the implementation process. On the other hand, the discretion provided for local translation is also a likely contributor to the extensive regional variation in practice, reflected in Zetterberg et al.’s (Reference Zetterberg, Sjöström and Markström2014) survey of court decisions. However, the most important factor for the shift from coercion to collaboration is that one feature of the new legislation was not negotiable: there was a strict requirement on the treating psychiatrist to attach a coordinated care plan to his/her application for CCC to an administrative court. This sets the conditions for the entire implementation process. It was not possible to start to apply the new legislation without this coordinated care plan. One thing that contributed was probably also that the template forms often serve a vital function in providing a framework for organising work. The coordinated care plan as a template form is compliant with institutional expectations and norms, and can therefore be received as a rational and legitimate interpretation of the original idea. Further, materialisation of the idea as a template form means that there exists a physical object that can be exhibited, assessed and developed when the need arises.
These findings have parallels in international research into CCC. Clinicians interviewed in a study from New Zealand held that the effects of CCC are often structural and indirect, where one important function is to bind together the necessary community services (Romans et al., Reference Romans, Dawson, Mullen and Gibbs2004). Service providers in an English study maintained that it would be possible to reach similar goals without coercion, although that would require more resources and a change in attitude (Canvin et al., Reference Canvin, Rugkåsa, Sinclair and Burns2014). Dawson (Reference Dawson2007) argues that clinicians will be reluctant to recommend patients for community treatment orders unless a community service of sufficient quality and intensity is available. There is a current debate about the effectiveness of CCC in the research community (Burns and Dawson, Reference Burns and Dawson2009; Churchill et al., Reference Churchill, Owen, Singh and Hotopf2007; Kisley et al., Reference Kisley, Campbell and Preston2006; Phelan et al., Reference Phelan, Sinkewcz, Castille, Huz and Link2010; Rugkåsa et al., Reference Rugkåsa, Dawson and Burns2014; Swartz et al., Reference Swartz, Swanson, Steadman, Robbins and Monahan2009). To the extent that positive effects can be shown, it is uncertain that those can be attributed to coercion as such. Results from a recent randomized controlled study in England/Wales suggest that intensive community services, rather than coercion, positively affects hospital outcomes (Rugkåsa and Dawson, Reference Rugkåsa and Dawson2013). In their study of CCC in New York State, Phelan et al. (Reference Phelan, Sinkewcz, Castille, Huz and Link2010) stress that assisted outpatient treatment must be regarded as a ‘package deal’: in addition to coercing patients, CCC provides them with access to enhanced services, such as priority for housing and vocational services. The authors claim that there is no conclusive evidence about the mechanisms behind positive outcomes.
If the main outcome of the introduction of CCC was to coerce service-providers into co-operating with each other, the legitimacy of the reform must be questioned. Should the autonomy of patients be limited for the sake of improving coordination between service providers? Does the benevolent spill-over effect to other patient groups motivate the coercion of some individuals? The Swedish experience reported here, combined with findings from international studies, calls for further investigation into the relationships between coercion and organisational factors such as determinants of the effects of compulsory community care. In what ways does legislation affect the distribution of resources, as well as the organisation and delivery of services? Would it be possible to achieve similar results without coercion?
