Janusian, anomic, agent, and steward: how employees perceive the identity of healthcare organizations

Abstract We suggest that healthcare organizations (HCOs) develop a hybrid identity that can be described in a typology made up of four identities: janusian, anomic, agent, and steward. These hybrid identity types result from the combination of two seemingly incompatible identities: utilitarian or business oriented versus normative or community care oriented. We also posit that the perception of HCOs’ identity is related with members’ patient-focused behaviour and organizational identification. To explore these possibilities, we surveyed a sample of 732 members from three very different HCOs: a non-profit mental health provider, a haemodialysis for-profit company, and a state-owned acute hospital. Results show that our typology of HCOs’ identity discriminates between organizations, between occupations and, more importantly, that patient-focused behaviours and organizational identification are higher when members perceive their organizations as janusian, i.e. rate their organizations high in both utilitarian and normative identities, orientations that do not usually go together. Implications for the management of janusian HCO are discussed.


Introduction
Healthcare organizations (HCOs) are complex systems operating in complex environments. Internal pressures to enact different professional and microsystem cultures can lead to reduced quality and efficiency. 1 Owing to competing external demands and fluctuating priorities, HCO members can neglect their focus on the patient. 2 Nevertheless, quality remains a generalized endeavour and a managerial duty in HCO 3 , but the extra effort required to change the organization depends, to a great extent, on the connection between professionals and their organizations in addition to their occupation. 4 How members perceive the identity of their organizations can influence both the degree of patient-focused behaviours and organizational identification.
Organizational identity, defined as members' selfdefinitions of the organization to which they belong, i.e., their answers to the question 'who are we as an organization'? 5 plays a central role in helping organizations deal with both internal differentiation and external pressures. As happens at the individual level, more differentiated identities can give HCO increased adaptive ability to cope with a changing world 6 without compromising their sense of purpose or meaning in life, a fundamental motive for identity construction. 7 Different identities in HCO, such as being a business or a community health promoter, can be seen as more or less compatible and an additional effort might be required to manage this multiplicity. In addition, organizational members develop their understanding of what organizations stand for and use these perceptions to guide their daily activities and to define who they are with reference to what their organization is. 8 The perception of multiple identities, especially if viewed as incompatible, can send ambiguous messages to the focus of members' daily action, or render organizations undesirable objects of identification and thus compromise the current level of care and the future-oriented quality and work safety.
The purpose of this paper is to examine how HCO members perceive their organizations by suggesting a typology of organizational identity based on the combination of two well-known apparently irreconcilable identities: utilitarian or business oriented versus normative or community care oriented. As a result of the combination of high and low levels of utilitarian and normative identities, we propose the existence of four types of perceived organizational identityjanusian, anomic, agent, and stewardand explore the degree to which the perception of these four identity types affects the intensity of organizational identification and the level of patient-focused behaviours reported by organizational members. The empirical study is based on a sample of 732 members of three very different HCOs. Results show that both patientfocused behaviours and organizational identification are higher when members perceive their organizations as janusian, i.e. when rating their organizations as high in both utilitarian and normative identities.

Background
Healthcare organizations: a plural nature in a plural context A diversity of literature underscores the plural nature of HCO. Acute hospitals are described as four worlds based on different unreconciled mind sets and organizing features: the world of cure represents the medical community and is organized with chiefs and committees; the world of care represents mainly nurses and also other professionals who provide care, with their own hierarchy; the world of control represents conventional administration, organized in a hierarchy; the world of community is represented by the hospitals' trustees, not formally connected to the hospital. 'The hospital ends up being not one organization, but four, as each part structures itself in an independent way' ( p. 58). 9 Healthcare networks are viewed as examples of organizations that face high levels of pressure from both organizing and strategizing pluralism. Organizing pluralism tensions come from the fact that these organizations are populated by strong professional groups often antithetical to management, and consequently, internal pressures to enact multiple cultures and identities are high. Pluralistic strategizing tensions arise from the need to meet multiple demands like increased service quality, cost containment, and an increasing business orientation. 10 Another source of HCO plurality comes from their different ownership structures. In most markets, public, private, and non-profit providers co-exist. 11 For instance, non-profit organizations are seen as especially relevant healthcare providers because they do not have to face the same incentive for profits to maximize revenue for shareholders and owners. At the same time, they have primarily social objectives, are naturally client-led and responsive to patients' needs 12 and do not suffer the bureaucracy of self-centered public organizations. 13 The entire healthcare field is defined as a prototype of a diversified context, in which multiple logics, such as market, professional, corporate, and state co-exist and shape the dynamics of the organizations and professions that belong to this field. 14,15 When organizations encounter prescriptions that are different from multiple institutional logics they experience institutional complexity, an enduring feature of the healthcare field, 16 which induces internal differentiation because different organizational actors will support distinct external logics. 17 HCOs are social actors that are a part of this turbulent scenario. Like all social actors, HCOs develop and express their identity and, by doing so, can be identified appropriately by society, expected to act in a self-directed way, set their own goals, and be accountable for their actions. 18 Organizational identity is a global property of organizational actors 19 not necessarily linked to particular individuals who are able to perceive organizations as entities. This property becomes global and a reified taken for granted reality through an externalization process. Externalization begins with powerful individuals who think about the essence of the organization (I think), who are followed by interactive episodes of sense-making negotiation between organizational members leading to shared understandings ('We think') until the meanings become institutionalized at an organizational level ('It is'), 20 and become an object of perception by individuals. At any time, individuals can be invited to answer questions about how they perceive this organizational property. In this case, they can look at the organization and say how It is.
As plural organizations, we can expect members of HCO to provide distinct responses to the question of who is their organization, thus revealing the hybrid nature of the perception of their organizations' identity. For instance, a specific organization can be viewed as a business in the quest for better efficiency and results, or as an entity devoted to promoting the health of the community. This duality regarding the essential meaning of the organization structured around business versus altruistic value systems has been called the generic utilitarian-normative hybrid. 21 Because organizational identity acts as a sense-giving device for organizational self-conception and action, it influences several individual and organizational outcomes, such as identification, reputation, or performance. 22 If members perceive that their organizations adopt more than one identity, they can experience ambiguity and conflict in the focus of their action and their identification with the organization can be compromised. This could happen in a context in which pressures to increase efficiency and the use of more business-like practices would lead organizations to enhance their utilitarian identity dimension and diminish the normative one. Organizational members could start to be less concerned with patients and more with cost containment or become less identified with an organization that does not properly fulfil its normative caring role, a core feature for healthcare professionals. These conceivable results are important, because if members lack focus on patients in their daily activities, the level of service will be undermined in the short term 23 and if members do not identify with the organization, the long-term service will be compromised too, because improvement requires the extra effort that comes from the connection between members and organizations. 4 Thus, dealing with the hybrid nature of HCO becomes a core managerial challenge.
Janusian, anomic, agent, and steward: towards a typology of hybrid organizational identity In their seminal work on organizational identity, Albert and Whetten 5 propose the existence of hybrid organizations, defined as organizations which define themselves with more than one identity that do not usually go together. A particular type of hybrid organization combines two apparently discordant value systems: a normative system, reflecting the internalization of a collective interest and altruistic ideology, like a church or a family; a utilitarian system, characterized by an economic rationality of self-interest, like a business. HCOs are this kind of hybrid organization. While observing a group of doctors in a meeting aimed at launching a new clinic, foundational authors documented the discussion of the fundamental question: 'are we a business or a humanitarian organization?'. In a hybrid organization, the answer would be 'we are both'.
Multiple identities describe the existence of diversified views of the organization that can be more or less in conflict, or place more or less contradictory demands on members. A common approach to describe these possibilities implies distinguishing ideographic and holographic organizations. 21 In an ideographic organization, different groups hold diverse conceptualizations of an organization's identity. This would be the case of a hospital's identity being described in fundamentally different ways by doctors, nurses, and managers. In a holographic organization, multiple identities are held by all organizational members, who look at the same organization as having two or more defining core features, and the possible tension coming from this duality is espoused to some extent by all members. HCOs are good examples of holographic organizations because more and more of them have to deal with the seemingly irresolvable conflict between seeking to reduce costs and being financially viable while maintaining an increasingly demanding level of quality care for patients. But more importantly, if one organization fails to accomplish just one of these demands, the survival of the entire organization will be at risk. 22 Thus, to somehow conciliate the multiple identities of HCO becomes a central management concern.
We follow the lead of Foreman et al. 22 about the requirement of HCOs to fulfil both efficiency and high quality care expectations, and propose a typology of hybrid HCOs. Instead of viewing utilitarian and normative identities as independent and seemingly irreconcilable dimensions, we suggest that these two identities can be perceived in conjunction by organizational members. Moreover, the combination of utilitarian and normative identities will influence members' identification and action, instead of each dimension separately, as studied by Foreman and Whetten. 24 Figure 1 depicts the typology we propose. Janusian HCOs are perceived by members as having high levels in both normative and utilitarian dimensions. Like the Roman god Janus, represented with two faces integrating the past and the future, a janusian HCO holds together apparently conflicting values of collective interest and economic rationality. On the contrary, we have organizations characterized by members as having low levels of both utilitarian and normative identities, thus failing to give their members higher aims to belong and promoting individuals' mismatch. In accordance with Durkheim, 25 we call these organizations anomic. Steward organizations are portrayed by members as having high levels of normative identity, but low levels of utilitarian identity. Like people who hold stewardship motives, 26 steward organizations give priority to collective long-term goals. Then again, we have agent organizations, perceived by members as having high levels of utilitarian identity and low levels of normative identity. The same as individuals who hold agency motives and are individualistic utility maximizers, 27 agent organizations are more concerned with financial gains. At the individual level, both fundamental motivations can coexist, with people seeking individual economic benefits and, at the same time, showing fairness, reciprocity, and altruistic values. 28 The same can happen at the organizational level, and HCOs can be viewed as hybrid actors and show several combinations of various levels of utilitarian and normative identities.
The perception of these four types of hybrid HCOs will have some influence on how members orient their actions and also on their identification with organizations. Organizational identity is a sense-giving template for individuals and organizational membership and requires individuals to enact a role on behalf of the organization. 29 Accordingly, for instance, if an organization is perceived as steward, this stimulates in members a greater concern with the care delivered to patients regardless of the resources invested. In addition, and because organizational identification is usually defined as the overlap between individuals' and organizations' sense of self 30 and organizations have the power to set organizational goals and engage in actions not necessarily in line with their members' individual interests and desires, 18 members are likely to be very sensitive to organizations' ability to enact a role consistent with their own sense of self. 29 For instance, the perception of an agent organization would lead professionals to reduce their identification if economic goals are pursued at the expense of high quality service to patients.
The existence of hybrid identities has advantages and disadvantages. Having more than one identity gives organizations greater potential to respond to the expectations of multiple stakeholders. 31 Besides increased flexibility, hybridity can foster deeper reflection about relevant choices and thus organizations can come up with more thoughtful strategic decisions. 32 High levels of internal conflict coming from competing priorities and internal supporters of one identity and not the other is a major disadvantage. Great ambiguity about what the organization is can lead to reduced organizational performance 33,34 and ultimately, strategic paralysis. 31 At the individual level, reduced identification and ambiguity regarding the proper focus of action can also occur, with impact on short-and long-term performance. If we assume that internal conflict is the normal state of any organization, then dealing with tensions coming from multiple identities becomes a core management activity.

Data gathering and sample
We collected data from three distinct HCOs, a nonprofit mental healthcare provider located in the surroundings of Lisbon, a state-owned acute care hospital situated in Lisbon, and a haemodialysis for-profit company with several facilities throughout the country. Our strategy was to maximize the differences between organizations' ownership, allowing for various combinations of both utilitarian and normative identities. After obtaining the consent of the administration of each site and ethical approval from the university research department (BRU-IUL), a professional was designated to coordinate the data gathering process. The study was presented as an academic project conducted under the responsibility of the authors. Questionnaires and informed consent forms were distributed to employees who volunteered to participate and a box was used to obtain confidential responses. Because the study implies the search for differences in the perception of organizations, in each organization we tried to involve different occupations. In total, we received 732 completed questionnaires, 28.4% from the nonprofit mental healthcare organization, 29.5% from the for-profit haemodialysis company, and 42.1% from the acute state-owned hospital. Table 1 describes the sample's demographic characteristics.

Measures
As measures of interest variables, namely normative and utilitarian identities, organizational identification, and patient-focused behaviours, we rely on existing validated measures. Since we assumed a social actor view of organizational identity, we used Bunderson's 35 professional and administrative organizational role obligations to measure normative and utilitarian identities. Acting in accordance with the administrative ideology requires organizations to play a bureaucratic system role and a market enterprise role. To act according to the professional ideology, organizations are required to play a professional group role and a community servant role. Respondents are asked to indicate, on a five-point scale (from 1 = not at all to 5 = to a great extent), to what extent their organization emphasizes characteristics aimed at measuring each of the two ideologies. 'Competitive with other similar organizations' and 'Concerned with community health' are examples of items for utilitarian (utilitarian) and normative ( professional) identities, respectively. In both measures, acceptable reliability was achieved (Cronbach α = 0.79 for utilitarian identity and 0.86 for normative identity).
To measure organizational identification, the sixitem scale developed by Mael and Ashforth 30 was used. This scale has been widely utilized in the literature and shows good metric characteristics. 36 Items have a standard five-point Likert-type scale. A sample item is 'When I talk about this organization, I usually say "we" rather than "they"'. The reliability of the scale is acceptable (Cronbach α = 0.81) Patient-focused behaviour was measured by the four-item scale developed by Paulin, Ferguson, and Bergeron 23 to test the impact of the level of service on customer-linkage behaviour and on organizational citizenship behaviour in the hospital context. The items, set at 1 (nothing) and 5 (highest), assess to what extent members' work is perceived as contributing towards patients' comfort and satisfaction, and the probability of recommending the organization. The reliability of this scale is good (Cronbach α = 0.86).

Results
Means, standard deviation, and intercorrelations of study variables are presented in Table 2. Globally, individuals perceive higher levels of normative (M = 3.43, Sd = 0.87) than utilitarian identity (M = 3.05, Sd = 0.91). These two variables are significantly correlated (r = 0.45, P ≤ 0.00), and means are statistically different (t-test = 11.18; P ≤ 0.00), consistent with the strongly institutionalized context of the healthcare field, which expects from HCOs mainly a community health promoter role. Relationships between utilitarian and normative identities and organizational identification and patient-focused behaviours are significant but low.
We suggest that the hybrid nature of HCOs can be described by the four major categories of janusian, anomic, agent, and steward, which result from the combination of high and low levels of utilitarian and normative identities. Because there are no parameters to determine these thresholds, we used our data to calculate limits above and under the mean of both variables. Then, we created a new variable combining the four possibilities. Figure 2 presents the results of this procedure. Differences in  the combination of utilitarian and normative identities are clear. The janusian type contains individuals who rated their organizations highly in both utilitarian and normative identities (32.8% of cases). The anomic type comprises the opposite position of low levels of both utilitarian and normative identities (32.2%). The agent type includes individuals who rated their organization high in utilitarian identity and low in normative identity (15.6%). The steward type corresponds to individuals who rated their organization low in utilitarian identity but high in normative identity (19.4%). Our typology should be sensitive to different organizations and also to distinct occupations of members within them. In order to explore these possibilities, we cross-tabulated the hybrid identity types with the three organizations and also with five health-related occupations. Tables 3 and 4 present the results, respectively, for organizations and occupations.
A chi-square test of independence was performed to examine the relationship between identity type and the organization. The relationship between these variables is significant (χ 2 = 114.48, P ≤ 0.01). Standardized residuals indicate that members of the non-profit mental health organization tend to favour a steward view of their organization. Members of the for-profit haemodialysis company are more likely to perceive their organization as janusian, and less as anomic or steward. State-owned hospital members are more prone to perceive their organization as anomic and very few consider it as janusian.
The relationship between identity types and individuals' occupation was tested by a chi-square test of independence. The relationship between these variables was significant (χ 2 = 74.72, P ≤ 0.01). Doctors are more likely to perceive their organizations as anomic. Nurses perceive their organizations more as janusian or agent, and less as anomic and steward. Other health professionals do not show any tendency towards any identity types. Medical auxiliaries consider their organizations more as anomic and less as agent. Members who perform support jobs tend to favour a steward view of their organizations.
One of the aims of this study was to explore the effects of our suggested identity typology on members' organizational identification and patientfocused behaviour. Table 5 reports the results of a multiple analysis of co-variance, with age, tenure, and gender as co-variates. The effect of organizational identity type on organizational identification and on patient-focused behaviour was significant (F = 25.16, P ≤ 0.01 and F = 13.90, P ≤ 0.01, respectively).
Post hoc analysis using the Scheffé criterion for significance indicated that the average organizational identification was significantly higher in janusian identity than in all other identity types. The organizational identification average is no different in agent identity than in anomic and steward types. Finally, the mean of organizational identification is  significantly lower in anomic identity type than in steward type. Post hoc analysis using the same significance criterion shows that the average of patient-focused behaviour is higher in the janusian group than in anomic and steward, but not when compared to agent type. The mean of patientfocused behaviour is higher in agent identity compared to anomic type but not to steward type. The average patient-focused behaviour in anomic identity is not different from the steward type.

Discussion
The purpose of this research was to propose a typology for HCOs' perceived identity composed of four categories named janusian, anomic, agent, and steward, and explore the impact of these categories on members' organizational identification and patient-focused behaviours. Our typology is based on the generic normative-utilitarian hybrid, identities usually viewed as incompatible. However, as HCOs are social actors, they are expected to provide high quality care to the community and simultaneously adopt management practices oriented to efficiency, and the absence of the fulfilment of any of these expectations will compromise the entire organization. This means that most HCOs cannot avoid the task of conciliating apparently incompatible identities. If we take the point of view of organizational members, our results show not only that contradiction does not seem apparent, as 32.8% rated high levels of both identities (M = 3.89 and M = 4.17, for utilitarian and normative identities, respectively) and thus creating what we called a janusian organization, but also that patient-focused behaviours and organizational identification are higher among individuals who have a janusian view of their organization. This result is important because we could expect that a steward type, characterized by a more prominent normative identity, and thus more consistent with professionals' ideology, would generate the highest levels of both organizational identification and patient-focused behaviours, representing a  distance from the utilitarian identity. Organizational identification is higher in the steward group than in agent and anomic ones but is lower than in the janusian group. Results pertaining to patient-focused behaviours go in the same direction, as the mean for the janusian group is higher than that for the steward group. If achieving high levels of patientfocused behaviour is fundamental to maintain a high quality of care in HCOs' daily activities, and increasing organizational identification is a major challenge to guarantee the involvement of professionals in quality improvement efforts, then devising a strategy towards a more janusian dual hybrid identity of HCO becomes a central management endeavour. The hybrid nature of organizations and its effects on individuals' attachment to their organizations has already been subject to empirical research. 24 However, our research took a different view. We combined normative and utilitarian identities in a meaningful typology and show that organizational identification and patient-focused behaviours are higher when members perceive high levels of these two identities, instead of determining the separate effect of each identity on individuals' identification or behaviours.
Our results have important implications for managing HCO. Approaches to managing the highly differentiated nature of healthcare organizations are already offered in the literature. These strategies tend to focus on the practice of mutual adjustment based on flexible and timely communication between professionals 1,9,10 or emphasize the central role of an overarching supra-ordinated goal as an integrative device that allows reconciliation among identities 31,37 and moulds multiplicity into a coherent entity. 38 Although our results are in line with these propositions, a more precise strategy can be devised. If identification and patient-focused behaviours are higher in members who perceive their organizations as janusian, then an approach to management development stimulating this style of thinking becomes a core activity aimed at improving HCOs' sustainability. Janusian thought happens when somebody notices the concomitant operation of two divergent ideas or concepts 39 and because it generates cognitive dissonance tends not to appear spontaneously in a human mind more prone to linear and non-ambiguous styles of thought. 40 Janusian thought implies a shift to a more paradoxical approach to management, one that requires embracing rather than resisting or refusing opposing demands. More conventional approaches are based on an assumption of alignment and coherence. Greater organizational performance could be achieved if leaders properly assess the environment, align strategies according to key environmental characteristics, and design organizations by arranging a set of coherent elements in line with strategies. 41 In our view, the inherent complexity of HCO requires a very different perspective, namely a paradoxical approach that challenges this assumption of the alignment and coherence of organizational performance by suggesting these organizations need to accommodate both shortand long-term competing goals linked to powerful internal and external stakeholders. In these contexts, organizational success is achieved not by the creation of alignment and coherence, but by hosting inconsistency, tensions, and contradiction. Leaders' challenge becomes the ability to manage alignment and adaptability, stability, and change, establishing routines and incorporating new procedures. 42 The major challenge is then to undertake a distinct leadership development programme aiming to develop the meta-skills that facilitate the emergence of a janusian thought style, namely paradox acceptance and paradox resolution via the synthesis of different domains. 43 Smith et al. 44 provide a detailed description of a programme aimed at promoting the integration of social and business demands that covers the central tenets of janusian thought.
In addition, our results also show that the perception of an identity type is not independent of individuals' occupational group. Doctors have a more anomic view, while nurses have a more janusian or agent view. Although professional differentiation regarding critical organizational features is widely recognized among HCOs, 45,46 the role of leadership in successfully achieving intergroup collaboration has been neglected. A leadership development programme aiming to increase the overlap between professional group identity and janusian identity type could be developed. Following Hogg, van Knippenberg, and Rast's 47 elaboration, managers could be challenged to develop a sense of intergroup relational identity, a self-definition of a group identity that incorporates the group's relationship with another group, in this case the organization as a whole, as a part of the group's identity.
We believe that a combination of approaches to managing the required compatibility of normative and utilitarian identities inherent to the janusian type will have a better impact on the adoption of a corresponding thought style than the use of just one. Irrespective of the approach used, developmental activities should consider recent work on leadership development that highlights the joint growth of both leaders' and organizational identities, as they are intimately related. 48 To engage HCOs' managers in the development of janusian identities implies the development of more differentiated and integrated manager identities.

Practice implications
In line with the previous discussion about the centrality of the meta-skill of paradox acceptance and resolution and the importance of the overlap between professional group identity and janusian identity, we strongly recommend HCO managers to engage in specific developmental activities targeted at these two outcomes. The first, because it implies the development of a complex skill, could take the form of a training programme in which managers are involved in a combination of class and field activities that lead them to accept paradox by embracing paradoxical thinking. The entire programme should facilitate a shift from an either/or approach to management to a both/and one. Some domains requiring changes in behaviours and mind set are: to value paradoxes as vital elements of high performance especially in uncertain and complex contexts; to proactively engage in identification of tensions; to learn how to escape from traps of anxiety and defensiveness; to recurrently communicate a both/and vision; to separate efforts to develop different sides of the paradox; act consistently inconsistently; promote the generation of new resources from existing ones; embrace multiple strategies and identities simultaneously; tolerate uncertainty. 42,49 The second outcome, because it entails a change of communication routines aimed at expanding the group identity, implies top managers working with professional leaders, such as key medical and nursing leaders, and collaborating with them in writing a professional identity statement that incorporates the group's contribution to both utilitarian and normative identities, and routinely using this element in the rhetoric of both top management and professional leaders' communication. For instance, formal internal communication tools, such as newsletters or memos, co-authored by leaders from different professional groups, could circulate this renewed identity statement. Multi-professional meetings could also be co-chaired by leaders from different professional groups and be routinely opened with the statement of the new identity.

Study limitations, implications and further research
This research has important limitations. The nature of the organizations studied does not cover all the variability or organizational forms existing among health services providers, which limits the external validity of our conclusions. Moreover, the main organizational activity of mental health, haemodialysis, and general hospital co-varies with the ownership status of non-profit, private, and state-owned, which makes it impossible to disentangle separate effects of these distinct dimensions on identification and patient-focused behaviours. Furthermore, we studied hybrid organizational identity from just one of the main established traditions in the field, the so-called 'perceived organizational identity' as an antecedent of individual organizational behaviour. 50 Future research could study HCOs' hybrid identity at other levels of analysis and focus on other perspectives, such as how different professions construct different narratives about what the organization is 51 or identify the strategies used by organizations and their members, especially leaders, to deal with apparently irreconcilable identities.

Conclusion
In view of changes in context, such as increased professionalism, higher quality expectations from patients, scarcity of financial resources and pressures for efficiency, HCOs will continue to develop a hybrid self-view including, at least, the combination of utilitarian and normative identities, perceived by their members as janusian, agent, steward, and anomic organizations. When perceived as janusian, organizations have greater potential to generate higher levels of patientfocused behaviours and organizational identification, and thus, become more likely to achieve both short-term performance and improvement capacity, two key elements of organizational sustainability. Finding strategies to conciliate more or less incompatible identities has become a major challenge for HCO managers, and thus, embarking on a development journey that includes the adoption of a more janusian thought style and the crafting of intergroup relational identities can be an important strategy to improve HCOs' sustainability.

Disclaimer statements
Contributors Francisco Guilherme Nunes contributed to the study design, analysis, and interpretation of data and prepared the manuscript. Luis Manuel Martins contributed to the study design, interpretation of data and commented on manuscript drafts.
Funding None.

Conflicts of interest
The authors declare that they have no conflicting interests.
Ethics approval After obtaining the consent of the administration of each site, ethical approval from the university research department (BRU-IUL) was obtained.