What professionals say and do: the tension between egalitarianism and hierarchy in interprofessional teamwork

ABSTRACT This study aimed to explore the power dynamics in interprofessional teamwork by conducting an ethnographic study of three interprofessional teams working in mental health and substance use services in Norway. Data were collected through 14 observation sessions and 18 in-depth interviews with health and social work professionals. Given the potential difference between “what people say and what people do,” we explored how ideas of power were articulated by health and social care professionals and how such structures were observed to be played out in practice. The findings suggest a presence of contrasting egalitarian and hierarchical structures, and that professionals were aware of the resulting tension and operated within it. This study contributes to the literature on interprofessional health and social care through providing an analysis of the power dynamics of teamwork interaction and how professionals relate to such structures. The results are relevant to a broad context of interprofessionalism as they provide valuable insight into how power should be understood as a continuum of changeable positions and motivations.


Introduction
Interprofessional teamwork is an active and ongoing partnership between professionals with various backgrounds and distinctive professional competence working together to solve problems and provide services.In the literature on interprofessionalism, the terms "teamwork" and "collaboration" are often associated with ideals of joint action, equal participation, shared decision-making, and interdependence (D'Amour et al., 2005;Reeves et al., 2010;Sangaleti et al., 2017).However, when the relationships between health and social care professionals have been examined, it is evident that an unequal distribution of power informs their interactions (Freidson, 1970;Konrad et al., 2019;Lokatt et al., 2019).These two narratives, one that we call egalitarian and the other hierarchical, indicate potential tension between two qualities of interprofessionalism. Taking this tension as its point of departure, this study aimed to examine: (1) how professionals articulate ideas of egalitarianism and hierarchy concerning their experience of teamwork, (2) how such articulated ideas are observed to be played out in actual interprofessional interactions, and (3) how power structures in interprofessional teamwork can be conceptualized.
We conducted an ethnographic study of three interprofessional teams working with outpatients in mental health and substance use services in Norway.In the field of mental health and substance use, patients' problems are often compounded by and intertwined with other health and social problems.Offering comprehensive care and treatment for such patients demands close interaction and collaboration between different professionals with various, and sometimes contrasting, perspectives (Hannigan & Coffey, 2011).Since interprofessional teamwork necessitates that professionals operate interdependently, it is important to understand the power dynamic in such interaction.

Literature review
Numerous studies have contributed to the literature on power involved in interprofessional interactions.While some studies have claimed that interprofessional work supports a shift away from hierarchies, such as medical dominance and the exclusive authority of some professions (Franklin et al., 2015), others contend that the notion of an optimal team comprising members with equal status is a potentially misleading rhetoric, as power relations are often sustained (Gibbon, 1999;Keshet et al., 2013;Nancarrow & Borthwick, 2005).
Regarding power structures portrayed in the existing literature, an array of concepts and perspectives can be found.Exploring the differences in professional status as a challenge to interprofessional work, some studies have argued that the power imbalance among healthcare professionals negatively influences the effectiveness of interprofessional teamwork (Atwal & Caldwell, 2005;Okpala, 2020).Consequently, the presence of egalitarian elements has been emphasized.For example, to address power imbalances, Karam et al. (2018) highlighted trust, mutual acquaintanceship, and respect among team members as key elements.Other studies have reached similar conclusions, emphasizing interdependence regarding competence and roles as core components of efficient interprofessional work (Baker et al., 2011;Bronstein, 2003;Sangaleti et al., 2017).Such egalitarian analyses have an intuitive appeal, as they invoke the ideal of inclusion, highlighting that all professionals have an equal right to speak, be listened to, and take part in decision-making.
In contrast to the literature emphasizing egalitarian elements, other studies have focused on how professionals negotiate their cultural, social, or symbolic capital to maintain or improve their own positions (Schot et al., 2020).For instance, Goldman et al. (2016) found that professionals, such as nurses and social workers, developed informal alliances to transform their positions into more dominant ones.In comparison, Lokatt et al. (2019) claimed that nurses' entry into interprofessional work came at the expense of their opportunities to speak and act freely as nurses' spaces of action were constructed in relation to the traditional power balance between doctors and nurses.Similar findings have been reported in other studies.Finn et al. (2010) found that lower-status professions have fewer opportunities to challenge organizational structures than higher-status professions have.Pichonnaz et al. (2021) observed that professionals in lower hierarchical positions avoided formulating direct requests to professionals in higher hierarchical positions, while Thylefors (2012) found that decision-making processes were associated with verbal dominance and professional status.
This brief review of the relevant literature suggests that, although ideals of egalitarianism have been highlighted as key components of successful interprofessional work, hierarchical structures are strongly present in the interactions between professionals.This conclusion is supported by Lingard et al. (2012), who found that although members of interprofessional teams emphasized nonhierarchical values, they practiced within a more hierarchical reality of medical dominance.Nevertheless, power relations are not static, but negotiable and situational.Furthermore, Nugus et al. (2010) observed how health and social care professionals articulated and exercised both "competitive power" involving domination of one profession over another and "collaborative power" involving interdependent participation evident through information sharing and common decision-making.
In summary, although many studies have sought to investigate egalitarian or hierarchical elements of interprofessional teamwork, less research has been conducted on how team members relate to both such elements.While building on the relevant literature in terms of its focus on power as a larger field of tension that shifts according to changing situations, this study explored how both hierarchical and egalitarian structures can co-exist and the degree to which professionals involved are aware of this.However, when contrasting egalitarian and hierarchical elements and structures in interprofessional team settings, egalitarianism might be assumed to be an ideal worth striving for while hierarchical structures might be considered negative.Regardless of such viewpoints, this study presupposed that both elements were functionally operational and influential.

Theoretical framework
The concept of power used in this study has various sources.First, following a Weberian understanding, power is considered as one agent's capacity to impose their will on other participants in a social relationship, even in the face of resistance (Weber, 1978).According to this view, power involves clear and visible dominance in which some agents seek to make others do what they would otherwise not do (Scott, 2001).However, the exercise of power is not always visible or conscious.According to Lukes (2005), power can also be exercised in more hidden ways, taking the form of ideological influence and false consciousness, which can lead individuals to want things that are opposed to their own self-interest.Finally, power can be understood according to what Goffman (1974) characterized as "framing," that is, the process of influence and culturally determined definitions of reality that allow individuals to make sense of objects and events.In such cases, power emerges through interactions between participants and is ordered by the effect of symbolic power in defining the situation (Hallett, 2003).
That said, power is one of the most contested concepts in social theory, and it is important to note that the essence of power can never be fully captured.Analytical concepts of power will only be able to grasp certain aspects of the social interaction and must always be adapted to the study context (Avelino, 2021).For this study, we believe all the three aforementioned concepts of power to be highly relevant, as the presence of power can be found in (1) the practice of "jurisdiction," that is one profession's claim to have legitimate control over a domain of work through its expertise (Abbott, 1988), (2) more subtle ways in workplace relations and the broader system of professional status determination, as well as (3) discourse related to professional interdependence and teamwork.

Setting
In this study, the three teams were selected based on two criteria: (1) teams had to comprise both social workers and health care professionals, and (2) teams were organized according to an interprofessional team model which implies specialized professional roles, with a high level of communication, mutual planning, interdependency, collective decisions, co-ordination and shared responsibilities among team members (Thylefors et al., 2005).The teams were recruited through the first author's professional network.However, the author had no direct affiliation with the teams prior to the study, nor had the author worked in this field of practice.
Each team comprised 8-14 employees with professional backgrounds in nursing, occupational therapy, medicine (psychiatry), psychology (clinical), social education (in Norway, this is a bachelor's degree), and social work.The patient groups included individuals aged above 18, with problems related to mental health, substance use, or both.Many patients had additional challenges related to housing, finances, somatic health, and social life.As shown in Table 1, one team worked at a daytime clinic.In this team, the patients' challenges and follow-up care were less extensive than those of the other two teams.The other two teams worked with comprehensive patient challenges, involving outreach and extended follow-up care.
The inclusion of three different teams ensured varied settings to collect data from.However, we considered that the data were comparable, as there were several important similarities between the teams.The goals of all three teams were treatment, rehabilitation, and social support for the patients.Each professional had a list of patients for whom they were responsible, either individually or with the help of another team member.In team meetings, tasks were distributed, information was shared, patient cases were discussed, and input from colleagues on complex problems was sought and provided.As such, although they worked closely together and had continual communication and interaction, each professional worked autonomously.

Data
Ethnography was chosen as the methodology for this study as it allowed us to study both the "saying" and "doing" of interprofessional teamwork.While the interviews supplied us information about the professionals' experience and reflections, the observations gave us insight into the social interaction between professionals.Data were collected between April and December 2019 by the first author, through 14 observation sessions and 18 in-depth interviews.Observations were conducted through attending team meetings once a week.Each team meeting lasted two to three hours.During observations, keywords and near-verbatim quotes were jotted down in a notebook and, later the same day, rewritten as more complete field notes on a computer.The focus in the observations was on "how the professionals discussed patient cases", that is who participated in the discussions, what was said, what types of questions were asked, who posed the questions, and how they were posed.In terms of patient confidentiality, the ethical approval obtained for this study only allowed observations of two of the three teams.For the third team, only interviews were conducted as patient information was not anonymized in the team meetings.Although this represents as a potential limitation of the study, it had little effect on the results as the purpose of compiling the three teams was to obtain richer and more extensive data material, not to perform a comparative study.In other words, the data from the three teams were treated collectively as one case (Table 2).
For the observations, all team members of the two included teams agreed to participate.Additionally, all members of the three teams were contacted by e-mail, and those who replied were included for semi-structured interviews to understand their experiences of interprofessional work.As the interviews were conducted after the observations of each team, many questions were developed from the observational data.Examples of topics addressed in the interviews included professional boundaries and roles, power, conflict, and collaboration.The interviews were recorded and later transcribed by the first author.To secure the informants' anonymity, information concerning how many from each profession were interviewed has not been provided in this study.However, all professions in each team were covered, and all professions were represented by one to three informants.When presenting the findings, the provided quotes and observations represents a spread across the three teams.

Ethics
All participants gave either verbal or written consent to participate after they had received both verbal and written information about the project by the first author.As this study fell into the category of medical and health research, it was approved by both the Norwegian Regional Committees for Medical and Health Research Ethics (approval reference 2019/809) and the Norwegian Centre for Research Data (approval reference 237,074).All methods were performed in accordance with relevant guidelines and regulations.

Analysis
This study involved a qualitative content and thematic analysis (Vaismoradi et al., 2015).Inspired by Tavory and Timmermans (2014) method for abductive analysis, where a parallel and equal engagement with empirical data and extant theoretical understanding is central, all data were coded through two rounds using NVivo 12 software (released March 20, 2018).The first author coded the data, but both Table 1.Summary of the teams.

Team members
Professions Description Team 1 8 Psychology, social work, nursing, and medicine -Daytime clinic -Short-term treatment -Patients' challenges less extensive -Team meetings once a week Team 2 10 Social work, social education, psychology, and medicine -Outreach and extended follow-up care -Long-term treatment -Patients' challenges extensive -Team meetings once a week Team 3 14 Nursing, medicine, psychology, social work, social education, and occupational therapy -Outreach and extended follow-up care -Long-term treatment -Patients' challenges extensive -Team meetings every morning Table 2. Data collection.

Number of team members
Team members interviewed Observation sessions Team 1 8 4 7 (14 hours) Team 2 10 7 7 (21 hours) Team 3 14 7 0 authors contributed to the conceptualization and interpretation of the material.The data from the three teams were analyzed collectively, without focusing on team differences.
The first round was open-ended and focused on identifying prominent themes.After identifying professional power as a distinct theme, a second round of coding was conducted.In this round, all data as considered relevant were categorized, resulting in two thematic concepts: 1) egalitarianism, understood as equality in professional status and in the right to exercise influence, and 2) hierarchy, understood as asymmetric power relations and structures of domination.
This study used two datasets: interviews and observations of interprofessional team meetings.While observations were used to describe the professionals' behavior and the context in which such behavior occurred, interviews were used to obtain the professionals' subjective explanations of interprofessional teamwork and the meanings they attached to it.In both the interviews and observations, we identified examples of egalitarian and hierarchical elements.Inspired by Margaret Mead, who observed that, "What people say, what people do, and what they say they do are entirely different things" (cited in Fitzpatrick, 2011, p. 74), we initially aimed to explore whether there was a tension between the "saying" and "doing" of teamwork when it came to expressions of power as represented in Figure 1.For example, whether there were articulations of egalitarian teamwork ideals, but in practice a more hierarchical approach was involved.
This hypothesis, we used as a starting point for the analyses, but soon discovered that it did not reflect in our material.In contrast to the previously mentioned study of Lingard et al. (2012), we found no specific tension between the informants' articulations of teamwork and how teamwork was played out.This finding led us to conclude that the professionals involved had an awareness of both types of power structure and operated within them accordingly.

The 'saying' of teamwork
With the term "saying of teamwork," we refer to how the informants articulated their experiences and thoughts about teamwork in the interviews.When asked about the benefits of interprofessional teamwork, several informants referred to ideas, such as equality in participation and influence, shared decision-making, and professional interdependence.These three elements correspond to findings from previous research (Karam et al., 2018;Sangaleti et al., 2017).For example, in an interview, one social educator stated: A square is made up of four separate lines, right?Together, the lines form a square.They make up something more than each individual part.I think that is a good metaphor for interprofessional teamwork.
Comparable statements were also made by other professionals.They all described interprofessional teamwork in terms of interdependence and "joint work," where the contribution of each profession played a significant role to the greater whole, that is, in solving multi-faceted patient problems.Furthermore, such statements of professional complementarity were found across professional status.When talking about medical responsibility and interactions with other team members, one psychiatrist stated: [The other professionals] are present in arenas other than me, and their observations can complement mine.Their observations are an important part of my decision making.
In saying that the contributions of other professionals contributed to personal decision-making, the psychiatrist's statement suggests some form of "equality in influence," as it emphasizes the need for complementarity.In comparison, when talking about participation in team meeting discussions, one nurse highlighted freedom of speech, indicating that there were no topics on which their comments or opinions were not allowed.The nurse stated: I often have an opinion on a topic that may originally belong to other professions.[e.g.housing].In such cases, I do not feel I have to "hold back" just because it is not my field of expertise.
The above-mentioned articulations can be interpreted as examples of decentralization of power between professionals, and as representative of an egalitarian framework in which professionals make sense to their experiences.It does not follow, however, that there was an effective balance of power between the professionals in all situations.We also found views expressed on how power was distributed unequally in relation to the informants' experiences of interprofessional work.Especially, our findings suggested that hierarchical differences between the professionals were related to differences in professional jurisdictions and authority.
In the three teams, because of varying levels of authority provided in professional accreditations, only specialists in psychology and psychiatry were authorized to conduct psychological diagnostic assessments of the patients.Furthermore, psychiatrists had overall responsibility for the patients' medical treatment, and other professions (nurses, social workers, social educators, and occupational therapists) had less formal authority.Linking such varying professional authority to differences in power, one psychologist stated: When some professions have an independent legal responsibility for the patients, but others do not have, then of course, it does something with the power balance between us.
Medical expertise, in particular, was associated with a high degree of authority and independence.In an interview, when asked how disagreements among team members were handled,

Egalitarianism
The 'saying' of teamwork

Hierarchy
The 'doing' of teamwork one social educator said, "It does not happen often, but if the team disagrees on a matter, then in the end, it is the psychiatrist who decides".Similarly, a psychologist noted, "Of course if the psychiatrist says something, we listen to it".These statements differ from the more egalitarian articulations of power as the informants articulated their experiences in terms of hierarchical differences, professional authority, and medical dominance, which accords with previous findings (Pichonnaz et al., 2021;Thylefors, 2012).However, it is notable that this discourse was often linked with constraints in relation to the professional "space of action" (Lokatt et al., 2019).For example, while expounding on a tension between "all voices counting equally" and "the imbalance in authority," one social worker stated: When it comes to issues such as coercion, for example, the voices of psychiatrists count more because they are the only ones who have the authority to make such decisions, but in the team discussion (before the decision is made), everyone's perspective counts just as much.[The psychiatrist], however, may have arguments based on more knowledge than the rest of us.For example, there may be medical reasons that the rest of us do not see or understand.So, if I am concerned about a patient, I must ask [the psychiatrist] for an assessment.I cannot do such an assessment myself.
Here, it can be noted that, although the team discussions were based on the principle of equality (everyone's perspective counts just as much), the psychiatrist had the final say when it came to certain topics.Although this viewpoint was reasonable as expertise gave the psychiatrist more power to make decisions, it also indicated there was an asymmetric relationship between the psychiatrist and the rest of the team.
In comparison, one psychiatrist talked about deliberately restraining himself so as not to overrule other professionals in the team.
Including the experience and perspectives of all professions in the team is a resource and to succeed with that, it is important that I do not act instructively or as "the one who has the last word."It is not always easy, but I try to sit back, be quiet, and listen.Thus, the psychiatrist articulated an awareness of hierarchical differences while also expressing the need for more egalitarian teamwork.In contrast with the social worker's statement, which noted limitations in the form of a lack of competence ("I have to ask the psychiatrist.I cannot do such an assessment myself"), the psychiatrist described self-imposed restraints involving a kind of internal responsibility and motivation, meaning that it was up to the psychiatrists themselves to decide whether to exercise such restraints.In other words, based on these two statements, egalitarian and hierarchical elements would appear to be in tension in terms of how professionals of different authority articulated teamwork interaction and that this tension affected the professional space for action in different ways.

The 'doing' of teamwork
With the term "doing" of teamwork, we refer to observations of interaction in team meeting discussions.In the teams included in this study, the weekly team meetings were the main arena for teamwork and used more as a forum for discussion rather than for decision-making.When discussing complex patient issues, interdependence in the division of labor and complementarity between the professionals were often observed.
One observation illustrating this feature was a team meeting discussion about a young patient living at home with his parents.The psychologist had conducted a home visit the day before and expressed a concern about the patient to the rest of the team.The psychologist suspected that, besides having a serious mental illness, the patient needed medical help and stated: "[The patient] refuses to eat, is up all night, and hardly sleeps.The parents are worried."The team discussed back and forth how to best resolve the case.The psychiatrist suggested that she could perhaps join the psychologist on a home visit for a medical examination, and they could find a solution together.A social educator said that she could also join and talk to the parents while the other two talked to the patient, "I might be able to alleviate some of their frustration."The discussion continued for several more minutes before the team leader concluded that the three should work on the case together.
This example illustrated how tasks and responsibilities were distributed based on each profession's competence and role.Furthermore, as the professionals both complemented and showed a willingness to depend on each other in the way they handled the case, the discussion revealed a process of interdependency.How various professional perspectives and roles intersected was also evident in other observations.
In another similar situation, a social educator wanted to discuss a patient case that the team had been working on for some time but without yet fully understanding the complexity of the patient's problems.The social educator gave details of a meeting conducted with the patient in the previous week.The social educator had assessed the patient's financial situation and considered that the team could help the patient better if the patient had a mental health examination.In response, one of the psychologists asked if the patient had agreed to such.In this situation, the psychologist and the social educator had different information, as they had encountered the patient in different situations and for different reasons.While the social educator provided insight into the patient's social situation (the patient's financial state) and potential substance use, the psychologist expressed an awareness of the patient's mental health.The above-mentioned situations represented typical examples of team meeting discussions in which no single profession alone had a complete understanding or answer to a patient's problems.
These two examples of observation indicated that interdependence, equality in participation, and influence were central elements in interprofessional teamwork, and that the informants articulated their experiences in those terms.Although such egalitarian interaction does not represent a total balance of power, we consider that it indicates a potential shift in power from professionals acting independently to a more shared responsibility.Egalitarian elements, however, comprised only one aspect of the interprofessional framework as sometimes hierarchical elements related to professional status and authority were more visible.
Although most professionals interviewed in all the three teams claimed that there was a culture of "freedom of expression", the observation notes also showed how some professional perspectives were perceived as more dominant than others.One example of this feature involved observations from a team meeting where one psychologist expressed frustrations with the work of another similar team.The team was discussing, at a general level, how the various departments at the clinic handled patient cases.According to the psychologist, there was an excessive focus on medication in one of the other teams: "It goes at the expense of other more social perspectives".The team members continued discussing how the focus of the other team differed from theirs.One nurse stated, "[The other team] have a strong focus on diagnoses.[Our team] is more holistic", while the team leader noted, "Well, who the psychiatrist is has a lot to say in how a team discusses patient issues".
This discussion showed that there was an awareness among the team members of the potentially dominant position of psychiatrists.Another observation from a team meeting illustrated restrictions in the professionals' space of action.The situation involved a team meeting discussion in which one of the social workers requested some medical information about a patient who was hospitalized.The psychiatrist in the meeting asked, "Why don't you call the hospital and talk to them?" Expressing a hesitation to call the hospital herself, the social worker asked whether the psychiatrist could make the call: "[The hospital] will probably listen more to you than to me".Regrettably, our data contained no information on how the case was handled beyond this point, and whether the psychiatrist called the hospital.However, when asked about the episode later in an interview, the social worker replied that the psychiatrist had more authority than her: "The psychiatrist would probably get a better answer from the hospital than what I would get.That I have experienced many times before".Consequently, feeling that calling the hospital and asking for medical information would likely be less effective coming from her, the social worker expressed a more submissive stance of "holding back".
Corresponding restrictions regarding the more dominant professions were less apparent in the data material.Instead, there were several observations from team meeting discussions where the psychiatrists would move beyond their professional boundaries and talk about performing tasks that were not expected of them such as, for example helping a patient to move, cleaning their house, or putting up their window blinds.This aspect was highlighted by the other professionals as positive and beneficial for the team and its patients.This example further underlined how professions with less authority had less space for action and fewer opportunities to challenge organizational structures than professions of higher status (Finn et al., 2010;Lokatt et al., 2019;Pichonnaz et al., 2021).

Discussion
This study explored how egalitarian and hierarchical elements were articulated in terms of affecting health and social care professionals' experience of teamwork, and how such elements were observed to be played out in actual interprofessional interactions.While hierarchy implies a strict division of power, egalitarianism can be understood as providing a defense against the power imbalance that can result in strict divisions of power.Based on our data, we concluded that interprofessional teamwork took place in a tension between egalitarianism and hierarchy but that team members transcended this tension while operating within it.For example, we found that, although the professionals articulated an awareness of differences in professional authority, an awareness that was also observable in their practice of teamwork, they also talked and acted in terms consistent with a power balance and of egalitarian ideals (e.g.interdependence, and an equal right to participate, speak, and be listened to).
Similar findings have been reported in previous studies, specifically that different power structures provide the conditions under which professionals make choices (Lingard et al., 2012;Nugus et al., 2010).This allows us to believe that a simple application of how power is exercised does not account for the complexities involved when different professions work together; instead, power should be understood as involving a continuum of changeable positions and motivations.For example, as illustrated in the previous sections, we found two different motivations of "holding back".While the psychiatrist explained "holding back" as not acting instructively and allowing more egalitarian interaction in the team, the social worker described "holding back" as motivated by a lack of authority within hierarchical structures.In contrast to previous studies that have mostly focused on hierarchical interaction when explaining different professionals' spaces of action (Finn et al., 2010;Pichonnaz et al., 2021;Thylefors, 2012), this study suggests that an analysis of professionals' spaces of action needs to include different power positions to enable a more comprehensive understanding of how professionals operate within both egalitarian and hierarchical structures.
Mead's claim that "What people say, what people do, and what they say they do are entirely different things" (as cited in Fitzpatrick, 2011, p. 74) implies a discrepancy between discourse and practice when applied to interprofessional teamwork.Such a discrepancy is supported by Lingard et al. (2012) who claimed that there is conflict between how ideas about interprofessional teamwork are conveyed and how they are practiced.The findings of this study, however, did not support that viewpoint.As shown in our analyses, the professionals articulated an awareness of both hierarchy and egalitarianism, and dealt with this in their everyday practice.For example, several professionals stated that even though they experienced the freedom to speak, in the end, if there was any disagreement, the psychiatrists were in an overriding position to make decisions.
In terms of how power can be theoretically conceptualized in interprofessional teamwork, we found little evidence suggesting that there was a superficial interprofessional discourse highlighting egalitarianism that worked to hide a more hierarchical reality.If the data had indicated clear articulations of egalitarian teamwork-ideals but a more hierarchical doing of it, it would have been possible to claim that this type of tension functioned to promote "masked" realities in which, despite the appearance of egalitarianism, hierarchy was preserved and taken for granted.However, as our findings did not support this interpretation, we consider Lukes (2005) concept of power as "false consciousness" less likely to be applicable.
Furthermore, even though many of the professionals articulated a form of medical dominance (Freidson, 1970), our data did not contain any examples of an explicit exercise of power in a "Weberian sense" (Scott, 2001).For example, it was never the case that the psychiatrists defied the opinions of the others and decided what the team should do.However, although the professionals seemed to agree in articulating teamwork as being both egalitarian and hierarchical, our data also highlighted how ideas of professional inclusion and egalitarianism were regulated by the dominant professions (as in the example where the psychiatrist acted to hold back).This last finding has been supported by relevant literature where interprofessional teamwork has been claimed to either recreate or support a shift away from traditional professional hierarchies (Franklin et al., 2015;Gibbon, 1999;Keshet et al., 2013;Nancarrow & Borthwick, 2005).Our study showed that, while the interprofessional discourse in relation to egalitarianism indicated a potential shift from medical dominance, there were also factors present that may help maintain an imbalance of power.The psychiatrist sharing about how he deliberately restrained himself suggests that the most dominant professions determine whether changes toward more egalitarian ideals should take place.In such situations, power accords with "the Goffman definition", that is, power lies in influence, and various actors' opportunities to frame and define the situation (Goffman, 1974;Hallett, 2003).

Limitations
This study aimed to illustrate the interactions between professional articulations and played-out realities, and how resultant tensions shaped the way in which professionals see and act within a socially constructed reality.However, as an ethnographic study confined to cases in the field of mental health and substance use, this study may not provide a fully representative view in relation to other interprofessional settings.For example, the relationships between the interprofessional discourse and the observed interactions would most likely be different within a hospital context with other institutional and organizational frameworks and expectations for various professions.Correspondingly, as services of mental health and substance use care is a field with a large degree of overlap between the professions, it enables a certain level of interdependence that may not be found elsewhere.Therefore, it would be interesting to explore professional interdependency in a different context of interprofessional healthcare, such as surgery for example, wherein the various professionals' contributions are much more specialized, and one might presume that a different level dominates.
Furthermore, this study did not consider how structural power (e.g. the legal framework) and other factors, such as age and gender, may influence relationships of power.Nor did this study go into how power relations between professionals affects the holistic patient approach inherent in interprofessional working.Finally, the study participants belonged to professions with different statuses, which most likely influenced how they articulated their experiences and performed their professional roles.A psychiatrist, for example, would likely experience and articulate their understanding of hierarchical structures within a team differently from professionals of lower status and with less formal authority.Regrettably, our data contained insufficient information to allow further consideration of such matters in the analysis.Further research needs to be conducted within different contexts and settings to increase understanding of how professionals manage the tensions between egalitarianism and hierarchy in interprofessional teamwork.

Conclusion
First, the study findings allow for a more informed view concerning how egalitarianism and hierarchy function in terms of articulated ideals and as observed reality.Based on the empirical data, it would appear that egalitarian team discussions helped enable the professionals involved to feel useful and important in their role.Through emphasizing aspects such as inclusion, complementarity, and the equal right to participate, each individual professional role was confirmed as indispensable.Second, although principles of egalitarianism may facilitate greater diversity in viewpoints and a more comprehensive approach to patient issues, hierarchical elements may be necessary and justifiable in various areas of teamwork, given the situations in which the one who has the most authority and competence would be expected to have the final say.One implication for practical use which might be taken from this study is that if team members have an awareness of both egalitarian and hierarchical elements operating in specific contexts, they will be less likely to be locked into fixed positions.In other words, in the context of interprofessional teamwork, professionals do not relate passively to social structures but navigate actively within them.Awareness of this point is likely to be important for a more effective implementation and management of interprofessional health care.Lastly, we would like to emphasize an analytical advantage of using ethnographic method, that is, the combination of observation and interviews creates a multifaceted material that allows to study different "layers" in interprofessional interactions.
The social educator answered, "No, [the patient] became very aggressive.I was thinking that [the patient] might have been intoxicated."The psychologist took a brief look at the patient's records and replied: I have only met [the patient] once in therapy, and then [the patient] was very passive, but yes, there might be something pointing towards substance use.My notes [from the therapy session] indicate anxiety and that [the patient] could have taken some sedative drugs.