Legitimising Inter-Sectoral Public Health Policies: A Challenge for Professional Identities?

that all Norwegian municipalities need to integrate public health concerns in their decision-making processes at all policy levels. Based on a Health in All Policies (HiAP) approach, population health and health equity are seen as whole-of-government responsibilities, making all municipal actors across sectors and professional boundaries responsible for health issues. Although many municipalities are well on their way towards implementing this goal, several experience a lack of legitimacy and inter-sectoral collaboration, as well as encounter conflicting profes sional identities. Theory and Methods: In this interview-based case study, we investigate the legitimacy of the HiAP approach in three Norwegian municipalities. We use an institutional perspective to analyse legitimacy, and we discuss how professional identities might relate to the implementation of this inter-sectoral collaboration. Results: Our findings suggest that the three municipalities are in the process of legitimising HiAP. Further, that legitimacy based on the integration of HiAP in planning and management structures and in formal documents seems easier or less complicated to achieve than other types of legitimacy related to personal understanding, values and norms. Conclusions and discussion: We argue that these findings may be related to the possible risk of identity conflicts, which could potentially pose challenges to collaborations, such as HiAP, and then again challenge the implementation of integrated care.

experience as complex and challenging. For example, some struggle to make public health a whole-of-government responsibility that is owned by the entire organisation across its boundaries and supported by leaders in different sectors, as opposed to being a concern for the health sector alone [6,7]. Gaining acceptance, support or legitimacy for public health issues within sectors, such as transport, education, culture and planning, is therefore a central but difficult concern, making the implementation of HiAP particularly interesting to investigate. We regard HiAP to be dependent on municipalities' legitimisation processes so that the policy results in action promoting health and equity instead of being left on paper in a plan. Our investigation is thereby relevant for the achievement of integrated care.

Theoretical perspective
Our aim with this article is not an instrumental evaluation of the input, output and outcome of HiAP in municipalities but a study of collaboration as a legitimising process within the HiAP framework. By using institutional theory, we aim to gain an understanding of agent-structure relations in the process of policy implementation, making sense of how and why institutions, including the actors therein, operate. Inspired by new institutionalism, we regard organisations and their agents to be dependent on their institutional environment and on gaining legitimacy not only to succeed economically but also to survive [8][9][10]. To examine HiAP, other scholars have used institutional perspectives, such as path dependency and rationalised myth [11][12], institutional logics [13] and translation of organisational ideas [14][15]. By contrast, studies that apply a legitimacy perspective have been difficult to find.
We regard HiAP as a national/international reform and thus a structuring force which requires the collaboration of different actors for its implementation. However, the capacity to implement is dependent on the support received from different participating sectors and on the legitimacy that governance structures manage to create. We understand legitimacy as ' a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs and definitions' [8, p. 574]. As a collaborative entity, HiAP therefore becomes a legitimacy-seeking institution. This new institutional perspective indicates that in order to gain legitimacy, municipalities can use their planning and management system as a legitimising process to establish coherence between the norms, values and beliefs in the institution and in the context [16]. However, different municipal departments might have different norms and values, as well as different views and understandings of what is desirable or appropriate [17]. Legitimising inter-sectoral initiatives might therefore be particularly challenging because these may, for example, be in conflict with actors' professional identities [18].
In this paper, we analyse the legitimacy of HiAP by using a combination of Scott's [9] theoretical framework of institutional pillars and Suchman's [8] further development of legitimacy types, a combination used previously [28][29]. In this framework, legitimacy has four different types: regulative, cognitive, normative and pragmatic [8,9]. Regulative legitimacy is related to the regulative mechanisms of institutions and organisations, representing different regulatory processes, such as rule-setting, monitoring and sanctioning activities that constrain and regulate behaviour. These may be constitutions, laws, codes, rules, directives, regulations and formal structures of control. Regulative legitimacy depends on conformity to relevant regulations and to legal or quasi-legal requirements [9]. Cognitive legitimacy is founded on the cultural-cognitive pillar and rests on understandings or conceptions that are pre-conscious or taken for granted. Individuals in organisations create cognitive pictures that represent the world and how to act in it [9]. Suchman [8] presented two types of cognitive legitimacy. The first is the existence of cognitive explanation models or an understanding of the organisation and its work. In the second form, the organisation and its work have gained legitimacy that is taken for granted, in which the organisation's legitimacy represents a cognitive picture in itself. Normative legitimacy involves institutions as being normative systems that include both values and norms. Values are the perceptions of the preferred or the desirable, whereas norms specify how things should be done, or they are the legitimate means to reach the desired ends. Normative legitimacy is related to goals and objectives and how to achieve them [9]. Pragmatic legitimacy rests on the self-interested calculations of an organisation's most immediate audiences. It is based on a perspective of usefulness, benefits and interests. An organisation or its ways of working obtain pragmatic legitimacy if they are considered useful or beneficial [8]. This might represent usefulness related to society, in general, or to someone, in particular. Although separate, the four types of legitimacy are interrelated, overlapping and possibly even contradictory. Gaining legitimacy is therefore not seen as a linear process in which one form of legitimacy leads to another in a certain order and with a definite end point. It is a continuous process, maintaining the different types of legitimacy [8,9]. Furthermore, in this paper, we use a paradox presented by Huxham and Vangen [18] to understand how conflicting professional identities might challenge the legitimisation of HiAP. They argued that in order for collaborations to be successful, collaborative initiatives need to be in line with professional identities. However, gaining collaborative advantages is based on the differences between the collaborating partners, so they seldom share mutual identities, which can be demanding and challenging, an argument supported by other scholars [19][20]. Huxham and Vangen [18] argued that tension exists between the need for participants to identify with their own organisations, departments and professions and the need to identify with the collaboration itself. On the one hand, the process of promoting identification with the collaboration itself (as in the HiAP approach) can be essential to making the collaboration effective because it is linked to the process of gaining acceptance and making partners buy into the collaboration. On the other hand, the goals or initiatives of the collaboration might contradict those of the member organisations, that is, the different municipal departments. Identifying with the collaboration itself may obstruct or counteract effective initiatives in their particular departments. Professional identity is therefore a central aspect of collaborations, albeit related to both challenges and successes.
Scholars investigating the HiAP approach have recognised the risk of health imperialism as a professional challenge in the implementation of inter-sectoral public health policies. They have discussed the demands of gathering diverse partners from different sectors under the umbrella of health, arguing that this might be met with distrust from professionals outside the health discipline [21][22][23][24][25]. However, investigations into professional identities, particularly those in relation to the HiAP approach, seem to be scarce. In this paper, using HiAP as a case, we aim to supplement earlier studies by utilising a legitimacy approach that examines how inter-sectoral collaborations work. To discuss our findings in relation to the potential risk of identity conflicts, we ask the following question: What types of legitimacy characterise the HIAP approach in the three municipalities? We also consider how legitimisation and its different types influence HiAP implementation, as well as how these interact.

Research design
To answer the research question and obtain an in-depth understanding of the legitimacy process of HiAP, we used a qualitative multiple case study design [26]. We strategically selected three Norwegian municipalities based principally on their relatively long experience in implementing the HiAP approach through integrating public health policies into their local planning and management systems. The selection was based on different sources: 1) a national survey on the implementation of public health policies in all Norwegian municipalities, 2) a national supervision conducted by the County Governor, 3) a web page containing municipalities' experiences with implementing national public health policies and 4) information from regional public health advisors. Norwegian municipalities vary widely with regard to demography. Therefore, municipalities representing different geographical areas and population sizes were chosen to ensure that the research material is as rich as possible [27]. See Table 1 for the details of the chosen municipalities.

Data collection
This case study is based on interviews with municipal employees and politicians from the three selected municipalities. The informants were selected strategically on the basis of advice from local public health coordinators and on the informants' positions and experience in implementing public health policies at strategic levels in their municipal organisations. In total, 30 interviews were conducted with 31 informants (13 women and 18 men) representing different types of administrative positions, politicians and leaders in different departments ( Table 1). Twenty-eight of the interviews were conducted in person, and two were conducted by telephone. They were all completed between May 2015 and October 2015, mostly at the informants' offices, and lasted from 45 minutes to two hours. All the interviews were recorded and then transcribed verbatim. The interviews were based on a semi-structured interview guide with three main themes: public health terminology, internal processes of inter-sectoral action and planning, and legitimacy of public health policies. This paper consists of responses to all these themes.
To prepare for the interviews and to supplement the interview data, we performed a document review. The main author read through the three municipalities' master plans, action/financial plans, annual reports, health overviews and public health plans to get an impression of how they had integrated public health policies into their planning and management systems. The document analysis was used as background material for the interviews and to supplement the understanding of the results; however, this did not involve a separate analysis with separate results.

Data analysis
An interview analysis was performed, inspired by Braun and Clarke's [28] thematic analysis. The thematic analysis was based on the different types of legitimacy [8,9]. It was also conducted in a manner that compares the three municipalities. The material was read through, and initial ideas were noted. Then, the dataset was coded in a systematic fashion, in which codes represented interesting features or meaningful units of the data that were suitable for the four main themes (the four types of legitimacy). Next, the codes were collected into potential sub-themes. Finally, detailed analyses of the themes were conducted, and suitable quotes were selected. When the results section was written up, the themes, codes and text were reviewed and revised to ensure conformity with the material and the text. NVivo software was used to organise the analysis. In order to supplement the thematic analysis, we performed word counts in NVivo to investigate the use of particular terms in the interviews.

Ethical considerations
The Norwegian Centre for Research Data approved the study, and the informants provided informed consent. All the quotations used were validated and approved by the representative informants.

Regulative legitimacy
The informants in all the three municipalities stated that as a regulative mechanism promoting regulative legitimacy [9], the NPHA has been essential in the process of legitimising the HiAP approach. They explained that the NPHA places the responsibility for public health issues with administrative and political leadership. However, we asked if the municipalities have developed local regulations facilitating HiAP implementation besides using national laws and regulations. In all the three municipalities, there has been eagerness to develop local regulations and structures, which serve as requirements to promote regulative legitimacy, in order to integrate public health policies into their planning and management systems in different ways. The municipalities use several different regulatory guidelines, plans and management documents to increase the regulative legitimacy of the HiAP approach. For example, all the municipalities require every sector to include public health issues in their annual action plans and yearly reports. Public health is also a main issue in their municipal master plans, or they have made a particular public health plan. Health impact assessments, representing the consequences for public health, are considered in all administrative and political issues in the three municipalities. With regard to the implementation of structural changes and the integration of public health policies into structural procedures and regulations, we therefore argue that the regulative legitimacy of the HiAP approach is quite high in all the three municipalities.

Normative legitimacy
The normative legitimacy of HiAP is related to its alignment with values, norms and roles [9]. Is the HiAP approach on the policy agenda representing an important issue? Is public health seen as an important part of different sectors' roles and responsibilities, as a part of their professional identities? According to the informants' experiences in this study, the HiAP approach seems to be on the policy agenda in all the three municipalities. The informants described public health as becoming an issue of interest both at the political and administrative levels; it is acknowledged as an important field. For example, in municipality 1, the mayor described the HiAP approach as a field of interest that is now a natural part of everyday work in the municipal organisation.
However, some informants, particularly those in municipality 2, questioned the meaning of having public health on the policy agenda and the consequences this brings. Although public health represents an issue of interest and importance and is discussed and included as a main issue in their master plan, some informants questioned whether politicians and administrative employees really understand the consequences of including public health as a main issue for future development. Some perceived public health as a term that is used by many and promoted as a central issue but that ends in public speeches and planning documents without further results or actions. One informant described this issue as follows: 'I believe that public health issues are well integrated. It has become more visual. It's got legitimacy because it's included in our municipal master plan. (…) However, as to what this inclusion of public health actually means, that's something different. What it implies. I believe that public health is integrated into our plans, but it's mostly left on paper in the plan'. (Public health coordinator, municipality 2) In this municipality, the HiAP approach might be legitimised because of its presence on the policy agenda and its regulative legitimacy, as it is represented in planning documents and regulative systems. However, similar to the informants from this municipality, we might ask whether this legitimacy is somewhat challenged if, despite being on the policy agenda, the HiAP approach does not achieve any result and if public health intentions remain merely on paper. Plans that do not lead to actions are a common challenge in planning practice and literature [29], and this municipality is no exception. This issue might be related to the legitimacy of the HiAP approach. A key element for translating plans into action is ownership; the argumentation in the plan needs to be experienced as meaningful by those using it [16]. When asking the informants in this municipality about their regulative public health mechanisms, they did not appear to have much ownership. For example, their main public health goals are included in their municipal master plan; however, when asked about their main public health goals, several did not know about them or where they were written. They did not use them, or they expressed that they do not have ownership of them. They did not seem to experience these goals as particularly relevant to their daily work. This result is somewhat surprising and represents a paradox, as the informants argued that the inclusion of public health in their municipal master plan is an important part of the legitimacy of public health in their municipality.
In general, many informants in the three municipalities embraced the values of the HiAP approach, recognising that public health is a part of their roles, although they worked in education, culture, planning and other areas. For example, in municipality 1, the chief executive officer considered himself/herself as a health minister and the principal director of public health issues, stating the following: 'When I started, I acknowledged the need to highlight this field. I need to see myself as a health minister in our municipality, to see this as a part of my role, that public health is not something that the health sector alone should work with'. (Chief executive officer, municipality 1) Whilst acknowledging that public health is a part of their roles or is relevant to their tasks, the informants pointed out that public health is not necessarily acknowledged as the main issue, as the main drive for action or as the main reason for their work. For many, public health issues naturally seemed to represent a less-significant factor in their professional identities compared with other issues.
Discussions about roles and responsibilities evoked certain feelings amongst some of the informants compared with discussions related to the other forms of legitimacy. According to Scott [9] this is natural because normative legitimacy involves discussions about professional identity. For example, one informant perceived public health as a natural element within the cultural sector. However, he/she stated that public health actors are now working with issues previously handled by the cultural sector without acknowledging and respecting the cultural sector's initiatives. He/she explained this issue in the following: 'It's something about respecting the things that exist from before. Can we build on that, instead of finding things that are not good enough? Because they might not know what's actually been done in the cultural field (…) We might discuss that "we still need this and that"; however, we also need to acknowledge the fact that the cultural discipline has been given less financial priority and that we've been working with these issues for a long time'. (Leader of the cultural department, municipality 2) This informant touched upon an issue mentioned by the other informants related to their roles and responsibilities and to the boundaries across professions and disciplines. One informant was more direct in criticising the HiAP approach, indicating that the public health profession seems to claim ownership of some of the cultural disciplines or arenas which he/she represents. He/she stated the following: 'Some think that public health takes culture's arena, that they cross the line too much. And they do'. (Leader of the cultural department, municipality 1) Although only a few informants expressed their criticism this clearly, we recognise such experiences as important and find it interesting that the most critical informants were from the cultural department. We questioned whether these results are related to the interplay and boundaries between the different disciplines and professions and their respective identities, opening possibilities for conflict. However, one public health coordinator argued that conflicts could be avoided if people were aware of how they promoted the HiAP approach. He/she recognised the importance of clear roles in the collaboration and specified that the public health coordinator and his/her team should not perform other sectors' roles and that they should recognise the importance of not telling others how to do their job. He/she explained this in the following: 'In some cases, it seems as though public health competes with other tasks-education, for example. For me, this does not represent a real conflict. My team and I are here to help, not to do everything else. We should not be executing public health initiatives. I know that in the education sector, they're sick and tired of all the different things they have to do, and I understand that. And this is one of the reasons that our public health work is not about the public health coordinator coming up with initiatives and then trying to get someone else to implement them for him'. (Public health coordinator, municipality 3)

Cognitive legitimacy
The cognitive legitimacy of the HiAP approach is related to individuals' understanding and knowledge of what the HiAP approach is and whether this common understanding across departments has become a truth that is taken for granted [8,9]. For example, in municipal-ity 3, living conditions seem to be a central issue in the informants' understanding of public health. As our earlier investigation in this municipality showed, almost every informant mentioned the term living conditions during the interviews when talking about public health issues. In comparison, no one in municipalities 1 and 2 did. Consequently, living conditions were also used together with the terms public health and public health work in their planning documents [15]. When asked if public health is a central area of interest amongst the politicians, one informant said the following: 'Yes, that might be, at least if you say "living conditions". As I believe you should. I don't believe that public health is a main issue of interest amongst people, but living conditions are more likely to be'. (Advisor, municipality 3) In this municipality, living conditions seem to have become a natural theme related to public health issues, which might represent an accepted truth [8] about the themes being related to one another. However, in municipality 2, an understanding of which public health issues constitute the significant aspects of the HiAP approach seems less unified. In this municipality, a mismatch seems to exist between what some central public health actors promote as central priority areas, what is written in their municipal master plan as key areas and what is experienced as key priority areas amongst the different professionals. According to the public health coordinator, the main priority area for operationalising the complex field of public health in their municipality is reducing social inequalities in health, which is stated in their municipal master plan as one of three priority public health goals. However, when questioning the different professionals in this municipality, not one of the other informants spontaneously mentioned social inequality as a central aspect of what constitutes public health work. Compared with municipality 3, municipality 2 did not seem to have gained a common understanding of social inequalities in health as a common cognitive picture of what public health is about. The public health coordinator was aware of this situation, acknowledging that promoting social inequalities as a central concern in the municipality is challenging. In municipality 2, other informants also argued that some sectors of the organisation lack an understanding or cognitive picture of what public health is. One informant confirmed the limited legitimacy in his/her sector and linked this to the staff's limited understanding of what public health is about. The leader said the following: 'I believe that this represents some of our challenges. If you had approached a teacher and asked, "What do you do in relation to public health?", I believe that they wouldn't have perceived they do anything at all. I don't know because I've never asked them. (…) I believe that the public health approach does not necessarily have much legitimacy because they do not understand what it is'. (Leader for school and childcare, municipality 2) Municipality 3 seemed to have a shared understanding and cognitive picture of what public health work is in their municipality, particularly compared with municipality 2 which seemed to struggle more. According to Scott [9], the legitimacy of an organisation is negatively affected when several actors have a conflicting understanding of their organisation and how it should function.

Pragmatic legitimacy
Organisations or their ways of working obtain pragmatic legitimacy if these are seen to be useful or beneficial [8]. To analyse our results, we asked whether the HiAP approach was considered useful in promoting health in the municipalities, in general, and whether it was useful to the informants or their departments, in particular.
In all the three municipalities, there was a general impression and experience of the HiAP approach as being useful in developing their municipality, in particular, and society, in general. The informants mostly acknowledged HiAP as a useful and important strategy for promoting health and developing well-functioning societies. However, concerning their own work, not all the informants considered HiAP as useful to their own departments or daily actions. For example, some did not see why they should prioritise cross-sectoral public health group meetings or planning processes or why they should report on public health goals or make new ones. For example, one leader did not recognise the advantage of reporting on public health issues and setting annual goals. He/she only continued doing so out of obligation. Instead of developing new goals, therefore, he/she just copied the text from the previous year. Some informants also expressed difficulties in getting actors to prioritise public health group meetings or other public health arenas or tasks not only because of the limited time and resources but also because these actors do not see what they gain from doing so (i.e. what's in this for me?) Some informants expressed quite clearly that they do not see the personal, professional or departmental gain in engaging in the different public health processes. For example, one informant expressed the following: 'I don't see any advantages. We assess the consequences for public health because we're obliged to do so; yes, we're obliged to'. (Leader of the cultural department, municipality 1) This leader does not see the benefit of considering the health consequences of all municipal initiatives for his/her own work.

Discussion
The results of our study show that all the three municipalities are in the process of legitimising the HiAP approach, although municipality 2 seems to struggle somewhat more than municipalities 1 and 3. In general, all the three municipalities seem to have gained regulative legitimacy, in which the HiAP approach is integrated into their planning documents, health impact assessments and management systems. However, the other types of legitimacy seem to be somewhat more difficult to achieve. Still, all the three municipalities are in the process of legitimising the HiAP approach, supporting earlier research that the implementation of NPHA goals is progressing [30,31].
According to Buanes and Jentoft [17], interdisciplinary approaches, such as HiAP, will challenge the regulative, normative and cognitive dimensions of the different disciplines. Promoting the legitimacy of the HiAP approach might therefore challenge these different dimensions in different ways. Crossing boundaries can involve breaking rules and questioning established paradigms and norms. However, compared with confronting norms and values, breaking rules and regulations can evoke different types of emotions. Breaking norms might involve self-evaluations related to remorse and self-respect, which again provide powerful inducements to comply with prevailing norms. Norms and values are particularly powerful and deep rooted compared with regulative dimensions [9]. When discussing norms and values, some informants get irritated, excited or frustrated, arguing that public health takes the place of and does not respect the cultural department's efforts. Norms and values are connected to what one is or who one is as a professional, and interdisciplinarity might raise questions concerning professional identities [17]. Furthermore, understandings and cognitive pictures are even more elusive to obtain, so the cognitive legitimacy of HiAP and the cognitive pictures of one's roles, responsibilities and identities might be particularly difficult to change [8].
In line with this argumentation, we ask whether cognitive and normative legitimacy might be particularly difficult to achieve in the case of HiAP because it could challenge established professional identities. One possible reason for actors not prioritising public health activities or acknowledging them as useful could be that public health issues are not included in their own understanding or cognitive picture of their work/profession. We might ask if this challenge to established professional identities may further explain the few, albeit strong, critical voices and emotions of the informants who perceive the HiAP approach as usurping the cultural sector or not respecting what has already been achieved. Other scholars recognise similar challenges as some form of health imperialism, representing challenges to the implementation of HiAP [21][22][23]. In line with the paradox presented by Huxham and Vangen [18], promoting the legitimacy of HiAP might be in conflict with or even contradictory to professionals' own identity or the effectiveness of their own departments. McNeil, Mitchell, and Parker [32] may help us understand the paradox in relation to our findings and the implementation of the HiAP approach. They argued that there are possible triggers to professional identity conflicts related to inter-professional/inter-sectoral practice and that differential treatment amongst groups within an organisation might be present in terms of different statuses, pay or opportunities. Furthermore, inter-professional or inter-sectoral work may threaten professional identities if individuals or groups experience the requirements to blend into the dominant or favoured culture or group as a devaluation of their existing roles. In the context of our findings, we might ask if the health sector represents a favoured group within the municipal organisation because of its larger size and budget relative to those of the other sectors. We might also question whether the challenge of gaining legitimacy for the HiAP approach might be related to the experience of health being dominant, making representatives from the cultural sector, in particular, which is often represented by low budgets, feel devaluated. Other scholars have recognised the challenge regarding the health sector's large size and budget in collaborations, such as the HiAP initiative, and understand that this might be met with distrust [25,33]. In line with Huxham and Vangen's [18] paradox, individuals might experience HiAP as a requirement to blend into a dominant group which economically threatens their discipline, profession or department, as well as their professional identity. If this is the case, it is understandable that embracing HiAP might not be considered wise or favourable and might be seen as a hindrance to their department's effectiveness.
However, although the health sector is dominant in many Norwegian municipalities, public health does not represent a defined discipline or profession. There are ongoing debates in the field about its aims, content, priority areas, procedures and ways of working [30,34,12], and public health coordinators and others find this disorientation or complexity challenging [35,36]. Interestingly, we might therefore ask whether the actors promoting HiAP (which might threaten collaborators' professional identities) actually struggle with finding their own professional identities themselves. One of the public health coordinators in our study argued that professional conflicts can be avoided by having clear and defined roles between the public health coordinators and other partners in the collaboration. We might ask if these clear roles are missing and might further challenge the legitimisation and implementation of HiAP. However, this question is outside the scope of this paper and needs further investigation and debate.
Our discussion shows that the cognitive and normative legitimacy of HiAP might seem more challenging to achieve than, for example, regulative legitimacy. Other researchers find similar results, showing the need for and challenge of gaining ownership and setting a common understanding and common goals when implementing public health policies [37][38][39]. However, this does not mean that regulative legitimacy is easily established or is non-significant. The same researchers find that the integration of public health in regulative structures and government tools, such as in plans, budgets and health impact assessments, is both important and challenging. In this context, the challenges regarding professional identity might represent one of the several factors needed to understand the process of legitimising HiAP. However, normative and cognitive legitimacy could be seen as particularly important to promote when legitimising HiAP, supplementing regulative legitimacy, which seems to be most easily established in the three municipalities investigated. Although professional identities, norms, values and cognitive pictures can tend to be particularly deep rooted [8,9,18], they are changeable. Promoting normative and cognitive legitimacy and promoting HiAP as a part of the different professionals' identities might also present rich opportunities. If established first, identities might again be difficult to change. Some scholars have argued that the challenges regarding professional identity and collaboration might be met through inter-professional education programs. Through inter-professional socialisation processes, developing dual identities based on an understanding of interconnectivity and the complementarity of roles may be possible, thus promoting collaboration in practice [40].

Limitations
The results of this study are dependent on context. However, the case study design allowed us to obtain an in-depth understanding of the issue that is transferrable to similar situations [41]. Furthermore, although we discussed professional identities, we did not analyse our material based mainly on the different professions. Rather, we used this perspective to show the possible consequences when implementing an inter-sectoral policy, such as HiAP.

Concluding comments
The results of our study show that the three municipalities investigated are all in the process of legitimising HiAP. Regulative legitimacy seemed to be the most evident type of legitimacy in the municipalities, perhaps because it is highly prioritised at the national level. Regulative legitimacy also seems easier to achieve than the other types of legitimacy, particularly those related to understanding, norms and values. We argue that one explanation for this might be the risk of identity conflicts, which can threaten collaborations, such as HiAP. Identity conflicts may be related to a different understanding of inter-sectoral collaboration and of roles, responsibilities, values and norms, which, in turn, might be related to whether the collaboration is experienced as useful or relevant to individuals' work or professions. We conclude that in the implementation and legitimisation of inter-sectoral collaborations at the municipal level, professional identities and the potential risk of identity conflicts need to be considered as potential factors challenging the legitimising process, and important for policies to result in integrated care.

Note
1 In the Norwegian context, the term public health represents a broad approach to public health, encompassing traditional disease prevention perspectives, newer health promotion ideologies and approaches for reducing social inequalities in health.