Pursuing collaborative advantage in Swedish care for older people: stakeholders ’ views on trust

Purpose – The purpose of this paper is to explore stakeholder views on the policy of integrated health and social care for older people with complex needs in Sweden and the issue of trust in implementing the policy. Design/methodology/approach – The study used a qualitative interview design and interviews with nine strategically selected stakeholders. A thematic analysis focused on trust, as defined in the theory of collaborative advantage, was used. Findings – This study of health and social care exposed a lack of trust on political, strategic and inter-professional levels. Two opposing lines of argument were identified in the interviews. One advocated a single government authority for health and social care. The other was in accordance with recently implemented national policies, which entailed more collaboration between local government authorities, obliging them to make joint local agreements. The Swedish experience is discussed in an international context, examining the need for collaboration in integrated care services for older people. Research limitations/implications – Although the findings are important for the current adjustment in health and socialcare for olderpeople, the number ofintervieweesare limited.Future studieswill includemore regions and longitudinal studies. Originality/value – Sweden is currently undergoing an extensive adjustment in line with recent national government policy which involves more primary health care and a corresponding reduction in the number of hospital beds. The restructuring of the care system for older people with complex needs is a paradox, as it simultaneously increases the need for centralisation while also increasing coordination and collaboration on a local basis.


Introduction
Across Europe, increases in the ageing population have led to national government initiatives in the care of older people (including home and residential care), advancing different models of integrated care. These include increased coordination and collaboration between health and adult social care services (Exworthy et al., 2017;Leichsenring and Alaszewski, 2004;Vabo and Burau, 2019). Sweden is particularly interesting from this perspective, as it has a universal long-term care model, involving both health and social services (Ranci and Pavolini, 2015). Despite a decline in the provision of services for older people, public resources allocated to this sector remain high in comparison with most European countries, making Sweden one of the most generous countries in the Organisation for Economic Co-operation and Development (OECD) in this area (Meagher and Szebehely, 2012). In Sweden, coordination and collaboration between health and social services have been at the forefront of national policies for eldercare (SOU, 2016, p. 2). Cross-sectional coordination and inter-organisational collaboration in the care of older people with complex needs have been widely advanced as a means to the end of integrated care (Coleman and Glendinning, 2015;SAHCSA, 2016). Through well-implemented coordination, different levels of national and local government are able to provide efficient and safe care to growing numbers of older people, and efficient collaboration between health and social care organisations generates continuity of care (Klinga et al., 2018).
However, many barriers to optimal collaboration in integrated care have been identified. There are, for example, different definitions of integrated care and varying ways of applying such care. There is also a general lack of understanding about the aims and objectives of integration and a lack of trust between the collaborating parties meant to provide it (Cameron et al., 2014). In particular, the issue of trust has been stressed as one of the main factors for initiating and sustaining successful collaboration (Ansell and Gash, 2008;Vangen and Huxham, 2003). But, although good collaborative relationships are essential for integrated care, they can be difficult to achieve, with inherent contradictions caused by differences between the collaborating agencies. However, these differences also create the potential for collaborative advantage, in the form of the synergy, which means it is possible to achieve things that could not be attained by any one organisation acting alone (Huxham and Vangen, 2005). The theory of collaborative advantage is about managing collaborations, and it is structured in overlapping themes that include goals, power, trust, culture and leadership, describing complexities in collaborative settings (Huxhamn and Vangen, 2005). Two factors are important for initiating trusting relationships, described as a trust-building loop. They are, firstly, the formation of appropriate expectations, and secondly, the willingness to take the risk that collaboration involves (Vangen and Huxham, 2003). When both of these factors are present, trust can be built gradually through modest aims. This leads to the reinforcement of trusting attitudes necessary for more ambitious collaboration (Vangen and Huxham, 2003).
However, such a small-wins approach is inappropriate if there are major social issues at stake, such as the reorganisation of care services for older people, or if there is a history of mistrust between the collaborating organisations. Then, a more comprehensive approach to trust-building is needed (Vangen and Huxham, 2003). Managing power imbalances is particularly crucial for sustaining trust. Usually, some partners are more essential to the collaborative enactment than others, and understanding and being open about this imbalance, and being aware of the way the power balance may change, is central to maintaining trust (Huxham and Vangen, 2005).
The purpose of this article is to explore contemporary policy responses, where coordination and collaboration are important features in dealing with the complexities of integrated social and health care for older people in Sweden. Different perspectives on the provision of care for older people are obtained in interviews with stakeholders and are compared with the most recent national government eldercare policy. The focus of the analysis is on cross-sector and inter-organisational collaboration.

Background and context
In Sweden, the integrated care for older people involves decentralised coordination between the regional (county) and local (municipal) authorities and inter-organisational collaboration between providers of regional healthcare agencies and local municipal social care. Regional authorities are responsible for all healthcare, including hospitals and primary health care. Local authorities are responsible for care for older people, but private entrepreneurs may be care providers after a process of public procurement by the local authorities. Interprofessional collaboration takes place mainly in multi-professional teams, e.g. between physicians in primary healthcare and professionals in specialised care and rehabilitation, and in the interface between home healthcare, multi-professional teams and municipal home care staff.
Community Care Reform (CCR) in 1992 was a turning point for the overall organisation of care for older people, and decentralisation became a key word. During the same period, marketisation of the public sector took place, opening up what had been a government monopoly to private entrepreneurs (Edebalk, 2016;Meagher and Szebehely, 2012). The vision was that local municipalities should generally take a greater responsibility for citizens' interests, and in particular, for eldercare. From then on, the regional authority was responsible for medical expertise and health care, and the local authority for social care, including home and residential care for older people. Distribution of responsibility between the regional and the local municipal care services also involved their cooperation in providing day-to-day services and in joint local planning (Government Bill, 1990/91:14). The CCR also contained administrative reforms, including a new law, the Liability Payment Act (1990:1404) ( Table 1). This forced local municipalities to pay for hospital care for older people ready for discharge if the municipalities could not make appropriate provision for these older people within five days. The idea was that with coordinated operational care planning, there would be much better collaboration between the regional and local authorities. How this should be implemented was, to a large degree, left to the local authorities and regions, which were expected to reach their own mutual agreements.
However, in 2002, a new Government Bill (2002/03:20) suggested a revision of the Liability for Payment Act (1990Act ( :1404 because of an urgent need to strengthen collaboration between the regional authorities and local municipalities. Stricter regulations for care planning before patients were discharged were introduced. Further reinforcement of collaboration and coordination between regional and local authorities was deemed necessary again in 2010. A policy was adopted that included coordinated care plans for patients in need of both health and social care and also included a registered healthcare contact prior to hospital discharge. Nonetheless, evaluations showed that this had little effect (SAHCSA, 2016). Finally, a new act came into effect in 2018; Collaboration in Discharge from Hospital Care (Act 2017:612, in short the Collaboration Act) applied to all patients, although older people are usually in more need

Legislation
Year Change Liability of Payment Act 1990 Local municipalities liable for payment within 5 days for older people ready for hospital discharge CCR 1992 Local municipalities responsible for social care of older people, regional authorities responsible for healthcare Government Bill 2002 Stricter rules of care planning and information from hospitals to community care before discharge from hospital Government Bill 2010 Policy of coordinated care plans and registered healthcare contact before discharge from hospital Collaboration Act 2017 Primary healthcare centres responsible for coordination of older people discharged from hospital. Local municipalities liable for payment within 3 days for older people ready for hospital discharge, unless local and regional authorities make up local agreement One of the main reasons for failure to collaborate is explained as stemming from a lack of trust. There is: a lack of trust and collaborative spirit which prevails between the different types of authorities and neither of them counts on the other to put the patient's need in focus. The national government's opinion is that both the regions and the municipalities should have the patient's well-being as their prime focus (Government Bill, 2016/17:106, p. 105f).
With the new legislation, the regional and local authorities are now forced to agree on their shared responsibilities and payment for patients in hospital care; otherwise, the "back-up" solution described in the new legislation will be applied. Where older people are considered ready to be discharged from hospital, this legislation shortens the time before the relevant local municipality becomes liable for payment of hospital charges from five to three days. This means that the local municipality needs to arrange home or residential care for the discharged older person within three days. It is not intended that the back-up solution will be used as standard, as it is anticipated that the two parties will arrive at agreements so as to avoid charges. There are also some specific requirements in the legislation addressing the regions' primary healthcare centres, which will be given the responsibility for coordinating care for older patients leaving hospital. This will rectify a long-held complaint about primary healthcare that it is not sufficiently pro-active, and that it does not assume any responsibility for the care chain (Government Bill, 2016/17:106). It will put further pressure on primary healthcare providers to comply with the legislation, including the provision of coordinated care plans and registered healthcare contact.
The first assessment of the impact of the Collaboration Act (SAHCSA, 2018) showed that the new act had clear aims, but statistics on a national level, necessary to fully evaluate the effects of the act, were lacking. Another problem raised in the assessment is the lack of a common definition of a patient "ready for discharge". This influences the way the number of days the patient is waiting to leave the hospital is measured.

Method
Study design and data collection The findings were part of a larger research project addressing sustainable quality coordination in health and social care for frail, older people in Sweden. This study used a qualitative interview design to obtain stakeholder views on the governance strategies for care for older people with complex needs. Stakeholders were strategically selected to represent national government, regional and local authorities and non-government organisations. Eight semi-structured interviews with nine stakeholders were held ( Table 2). The interviewees were from the same municipality in one region, although the municipality is large and divided in several districts. The purpose of the interviews was to compare and analyse the views of stakeholders currently engaged in the organisation of care for older people with complex needs. The interviews were carried out during autumn 2018, nine months after the new Collaboration Act (2017) was adopted. At this time, there had not yet been any financial transfers from hospitals to primary healthcare, nor was it clear how such transfers should be developed. The interview guide focused on the recent changes in national policies, on the challenges and advantages of collaboration between health and social care, as well as quality issues in the care of older people. There are specific and overlapping terms associated with older people with complex needs, such as "frailty" or "multi-morbidity" (see Cesari et al., 2017). In the interviews, the definition "older people with complex needs" was used, alternatively "older people in need of both health and social care".
The interviews varied between 1 and 2 h, and they were recorded and transcribed verbatim. In one interview, two representatives from one region participated at the same time.
Ethical approval was obtained from the Regional Ethics Committee in (Gothenburg,, and informed consent was received verbally by all participants.

Analysis
A thematic analysis in six steps was carried out (Braun and Clarke, 2006).
Step (1) Familiarisation with the data; the research group read the transcripts individually and then discussed the interpretations. The significance of trust versus mistrust in the interviews was identified in the group discussion.
Step (2) Generating a coding frame; a theoretical coding frame using a trust theme following guidelines taken from the theory of collaborative advantage (Vangen and Huxham, 2003) was set up.
Step (3) Searching for themes; the interviews were carefully read and quotes sorted under the themes (for example: power imbalances, trust, mistrust, common aims) in the coding frame.
Step (4) Refining themes; the initial coding frame was revised and codes were merged to be more comprehensive (Table 3). Relevant quotes under each theme were selected and translated (by the first author, IK).
Step (5) Defining and naming themes; three main themes were delineated in light of the research questions and the application of theory. The final Step (6) in Braun and Clarke's (2006) scheme is writing the report. The thematic analysis showed that most of the interviews included comments on continuing mistrust between regions and local municipalities on political, strategic and interprofessional levels. Nonetheless, there were also examples of trust and trust-building, as well as aspects agreed upon regarding strategies and implications for the future of integrated eldercare. The findings are presented under three main themes: mistrust, prospects of trustbuilding and integration.

Mistrust
In spite of all the efforts geared at more coordination and collaboration, most of the interviewees were frustrated by pleas for collaboration and coordinationthey just did not seem to work. The exception was interviewee IP2, who acknowledged the difficulties, but seemed convinced that the new Collaboration Act (2017:612) would address many issues. Mistrust and a power imbalance between the regional and local authorities was pervasive and went back a long way. Regions were thought of as more resourceful and with more higher-status professionals, while municipalities were described as being smaller and less important, "It's a little bit like "big brother" and "little brother". The municipalities think that nobody else should set the agenda, while they also believe that the regions and the health care sector do so quite a lot" (IP4).
It was said by one interviewee that antagonism and differences between regions and local municipalities still existed, which gave rise to a blame game. Hospitals, "want to get rid of these patients and there is someone else [the municipality] who must bear that cost. It turns out to be a blame game, where someone is pushing the costs somewhere else. That's my experience" (IP6). In this quote, the interviewee blamed the healthcare sector, but there were other examples where the situation was reversed. The reasons behind such antagonism was explained as authorities having different understandings of their own missions and designs and a history of mistrust.
Although there were meetings and processes implemented to promote collaboration, not everybody accepted responsibility: We're not in pace with each other, municipalities and regions. Yes, you sit in those meetings, but it's not really their [municipalities] job [it's always somebody else's]. How can that ever be a joint project? The municipalities do not really give many answers, either. It's quite comfortable to sit in the back seat and say, "I'm not the driver" (IP3).
This comment is interesting, as it is actually the municipalities that are responsible for eldercare.
The power imbalance was exacerbated by different budgets, where everyone was trying to shuffle their problem over to another budget area and where it was claimed the regions had more resources. IP7 asserted that hospital care devoured most capital, and the social care in the local municipalities received much less funding, as did the primary healthcare units, "Look at the hospitals -they just scoop up millions there, and then look at primary carejust look at it. There's a huge financial difference which is advantageous to hospitals" (IP7). With the new Collaboration Act (2017:612), there will be an increased focus on primary healthcare and changes in financial distribution, with more resources allocated to local primary healthcare centres, although this was something that IP7 expressed doubts about.
There were references to inter-professional mistrust; the different cultures and statuses of healthcare professionals compared with social care staff in the local municipalities. This applied particularly to the levels of expert knowledge and professional culture. There was a general understanding that health professionals had the authority to make decisions because of their professional knowledge. Still, there were also interviewees who underlined the competence of home care staff, as they were close to the older people and knew them well, "Hospital staff should acknowledge, and trust, the competence of home care staff" (IP8). Having said that, the importance of education, care skills and inter-professional working as well as the increased status of professional carers, which could lead to an increased willingness to work with older people, were issues raised as future challenges in the care for older people.

Prospects of trust-building
Much work has been done to implement integrated care, such as the establishment of multidisciplinary teams and individual care planning. IP9 had contact with many older people, and according to her, much had improved: They [older people] have had discussions before discharge from hospital, and the special teams have a lot of knowledge, and they also look after the next of kin [. . .]. They [older people] are very pleased. Even when they had to go into the Emergency Department they were admitted immediately and did not have to wait for hours (IP9).
IP2 was a strong believer in the new Collaboration Act (2017:612) and referred to situations where there were signs of overcoming disagreements through dialogue and moving towards common goals. He remarked how effective collaboration was closely related to trust, "We actually wanted the same thing, to build trust, build arenas for collaboration. All of the organisation [SALAR] was behind the Government Bill [to the Collaboration Act] [. . .]. You have to be good at joint planning and collaboration to build trust" (IP2).
All the interviewees agreed on certain issues. Up to now, primary healthcare had not been involved in the care of older people to a sufficient degree. Clearly defined boundaries for responsibilities and expanded information and communications technology (ICT) were things that would make a change in the care of all people and specifically those with complex issues. Previous legislation to ensure increased coordination and collaboration had not been complied with, and consumer choice was inappropriate for the older target group.

Integration
There were two main lines of argument in the interviews, one including an integration of health and social care authorities and the other arguing for more collaboration between regional and local authorities. The first point involved the integration of health and social care for older people into one governmental body, "I think there must be a clearly defined task for a single government agency, and if that's not possible there's got to be one authority which has primary responsibility and which decides over the other authorities" (IP5). Another view involved nationalised health care, "In my darkest moments I pondernationalize all hospitals and let home health care be governed by the regions, like in Norway" (IP7). The reasoning behind such points of view originated in the ingrained mistrust and power imbalance between regions and local municipalities, the non-compliance with previous legislation and vague agreements, which did not work properly.
The second line of argument reinforced the need for more collaboration between regional and municipal authorities and obligations for them to collaborate and coordinate their services in local agreements. Investments in long-term change, together with the implementation of new ways of working and collaborating, were proposed. There were also expressions of uncertainty: Since 2010 more collaboration has been initiated, still it's moving too slowly and the older people suffer because of that. I do not know what more to do, maybe more legislation, I do not know [. . .]. There's a huge potential for improvement in collaboration, team work, coordinationwell, everything (IP1).

Discussion
The aim of this article was to explore contemporary policy on integrated care and stakeholder responses to it, focusing on policy coordination and collaboration. Trust between the authorities involved is one of the most important features of collaborative advantage and crucial for integrated care. The findings show a prevailing mistrust between health and social care sectors on political, strategic and inter-professional levels. Mistrust is not uncommon in collaborative relationships, and there is frequently a need to proceed with collaborative agendas, even when trust is lacking (Vangen and Huxham, 2003). The way forward is through time-consuming and continuous attention to a range of activities such as communication about common goals, making agreements and sharing resources. In particular, it has been suggested that sharing resources is an integral part of trust-building and coping, when trust is otherwise lacking (Vangen and Huxham, 2003). Sharing resources was also put forth as an urgent issue by the interviewees as was joint recordkeeping. This is in line with reported challenges with implementing integrated care in the European Union (EU) (European Commission, 2017) such as limitations of ICT, information structure and organisational structures relating to the divided roles between department and professionals. Moreover, evaluating outcomes from models of integrated care, and from patient experiences, has proved challenging; there is no clear evidence that integrated health and social care models reduce hospital admissions or rates of bed days in hospital, nor that they are more cost-effective (Baxter et al., 2018;Mackie and Darville, 2016). However, from a patient safety perspective integrated care has shown some positive effects, and people with complex needs seem to benefit most through an outlined model similar to that proposed in the Swedish legislation: a single entry point in the community, generally in primary health care, individualised care plans and interdisciplinary teams, etc. (Hendry et al., 2018). Despite the obvious conflict areas, power imbalances, lack of ICT and the need for encouraging social care skills described by the stakeholders, one interviewee emphasised that older people themselves described improvements in the coordination of care services. The most recent national government directives attempt to mandate coordination and collaboration through national government policy. The aim is to force the authorities to coordinate and collaborate in health and social care of older people. Such moves are not unique for Sweden, but are also the case in, for example, Denmark and Norway (Vabo and Burau, 2019). How strictly national government policy is enforced varies. In Sweden, the push for coordination and collaboration from national government has increased and enforcement has become stricter over time. However, putting primary healthcare in the driver's seat may not be successful for a number of reasons. These include non-compliance with previous attempts to involve primary healthcare centres, the general shortage of general practitioners and the unintended consequences that may follow government incentives, such as health professionals' feelings of lack of autonomy (Andersson B€ ack, 2016). The findings disclose that there is still antagonism between the regional and local organisations, depending at least partly on their separate budgets. Hospital care consumes most resources, but from an international perspective, Sweden has the lowest number of hospital beds in Europe (Eurostat, 2019) and a comparatively short length of stay in hospitals (Szebehly and Trydeg ard, 2012). The intention of the new Collaboration Act was to shorten the length of hospital stays even more, leading to increased pressure on solutions for home-based care, collaboration between authorities and integrated services.
In view of the future demographic increase in the number of older people, the findings in this study indicate that the group of older people with complex needs might gain from being the responsibility of a single government authority. This was one argument asserted in the interviews, and the solution was seen as aiming at overcoming power imbalances and professional mistrust. Thus, it was a way of enforcing the building of trust. However, an objection to this line of argument is that barriers between different authorities and agencies would not disappear, but only be transferred to other areas. In spite of national policy to push coordination and collaboration between authorities to enhance integrated care for older people, not only in Sweden but also across Europe (Leichsenring and Alaszewski, 2004;Vabo and Burau, 2019), it has not fully succeeded. There are indications that the initial confusion of responsibilities between health and social care has merely moved on, affecting multidisciplinary teams and an array of providers (Thorslund, 2004). Given the increased centralisation of the management of the care sector, and the resignation expressed by interviewees about the fulfilment of collaborative advantage, it is possible that a merger of health and social care sectors would be beneficial in the care of older people with complex needs. The two lines of arguments expressed by the stakeholders in this study may not be mutually exclusive. Across Europe (European Commission, 2017), there are many examples of how such views could be reconciled, e.g. by "joint governance" through an Integrated Management Board made up of representatives of all providers or one organisational body to manage budgets, contracts and facilitate system integration.

Limitations of the study
The number of interviewees are limited, although the findings are important for the current adjustments in health and social care for older people. The findings point at the necessities and complexities of trust in collaboration and add impetus for future development of sustainable coordination. In addition, these results are important because they could have implications on the field. Later stages of the research project will encompass more regions as well as longitudinal studies.

Conclusion
This paper examines the issue of trust in cross-sector and inter-organisational collaboration in health and social care for older people in Sweden. Nine stakeholders involved at political and strategic levels in the care of older people were interviewed. Three main themes were delineated: mistrust, prospects of trust-building and integration. The findings showed a continuing mistrust between the regional and local municipal authorities despite legislation pushing for increased coordination and collaboration. There were small signs of trustbuilding, including agreements on important issues, such as the need for primary healthcare to shoulder more responsibility. Two contrasting stakeholder views were identified. The first and major perspective urged more centralisation and a single government body responsible for all care of older people, while the second advocated more coordination and collaboration based on mutual agreements, dialogues and networking.