Consensus or contention: an exploration of multidisciplinary team functioning in an Irish mental health context

Multidisciplinary teams (MDTs) are conventionally recommended in mental health care literature as an important way to offer holistic treatment provision to patients. This study aims to explore multidisciplinary teamwork in contemporary mental health settings, particularly what aids and hinders the process of multidisciplinary teamworking, and the social work contribution in such teams. In order to attain an in-depth exploration of these phenomena, a single case study design was employed. Within this design, data were generated through semi-structured interviews and structured observation of a mental health MDT in Ireland. These data were analysed using interpretative phenomenological analysis. The research highlights how the concept of mental illness is contested within this MDT, with the medical model dominant within an environment of fluid working arrangements. Professional role blurring and stereotyping were found to impact the division of labour on this team, with role negotiation found to be an integral part of retaining a professional's practice identity. This research raises concerns for social work's capacity to function within mental health MDTs in Ireland, and highlights ways in which social work educators might respond in order to empower Irish social workers to meet the challenges of mental health multidisciplinary teamworking.


Introduction
Irish mental health services in line with the general trend throughout the EU-25 is shifting from an institutional model towards a comprehensive, integrated, community-based model of care (Mental Health Commission, 2010). The Irish government's current policy document on mental health Vision for Change (Department of Health and Children, 2006)which broadly echoes academic literature and policy within the EU-25 (Knapp, McDaid, Mossialos, & Thornicroft, 2007)-and its two policy predecessors Commission of Inquiry on Mental Illness (Department of Health, 1966) and The Psychiatric Services: Planning for the Future (Department of Health, 1984), have recommended a recovery oriented, bio-psycho-social model of mental health care provided by different professional disciplines, such as psychiatrists, social workers (SWs) and psychologists within the context of an egalitarian community-based, multidisciplinary mental health teams. In this provision structure, Vision for Change (Department of Health and Children, 2006) emphasises full team participation in order to facilitate holistic bio-psycho-social discussions on patient assessment and care planning, with the consultant psychiatrist (CP) acting as a nominal team lead in order to coordinate the work of the team. Academic literature generally argues that multidisciplinary teams (MDTs) of this nature are the most effective means of organising mental health services and offer numerous benefits to both mental health patients and professionals working on the team such as an increased capacity to: solve problems together through an open exchange of ideas (Colombo, Bendelow, Fulford, & Williams, 2003), carry out comprehensive, holistic needs assessments (Mental Health Commission, 2006) and meet identified needs with the availability of a broader range of skills (Bailey, 2012). Empirical research data substantiating these claims are sparse, however (Armstrong, de Burca, Flood, & MacCurtain, 2007), due to the dearth of observational studies on how teams of this nature function available (Shaw, Heyman, Reynolds, Davies, & Godin, 2007). The aims of this study are: (1) to explore in-depth how a single mental health multidisciplinary team functions, (2) to examine what factors aid and hinder the process of effective multidisciplinary teamworking, with a specific emphasis on examining the factors which influence the social work contribution on such a team and (3) if applicable suggest relevant developmental responses which may be developed to improve mental health MDT functioning in Ireland and the social work contribution to such teams.

Literature
Contested nature of mental health Mental health is a highly contested and acrimonious phenomenon (Herron & Mortimer, 1999), with many different perspectives and frameworks which can be used when speaking about mental normality and abnormality in society (Rogers & Pilgrim, 2005). Medical, psychological and social approaches to mental illness are the three dominant ideologies held by differing professional groups which have historically permeated mental health practice (Rogers & Pilgrim, 2005). Each perspective is based upon differing understandings of the etiology, nature and treatment of mental illness (Strauss, Schatzman, Bucher, Ehrlich, & Sabshin, 1981), with the level of individual professional support for each model inextricably linked with their different professional culture and value systems, developed in training and nurtured further by professional socialisation and experience of working with mental disorder (Colombo et al., 2003).
Historically, community psychiatric nurses (CPN) and psychiatrists have been found to clearly favour the medical approach, traditionally the dominant framework used in mental health services (Kirk, 2005), while SWs supported the social model, and psychologists the psychological model of mental health (Mental Health Commission, 2006;Strauss et al., 1981). The medical perspective has retained its hegemonic status in mental health practice, despite the fact that there is minimal conclusive scientific evidence confirming a biological basis for abnormal behaviour (Pollard, 2010). This fact highlights how medicine is not necessarily justified as having a higher status when compared to other mental health perspectives. As a consequence, the decision-making process regarding the management, care and treatment of patients can become highly contested, allowing the social and psychological models of mental health, and a combination of both of these models in the form of a psycho-social perspective to compete for authority and recognition alongside the traditional medical approach (Colombo et al., 2003).

Multidisciplinary teamworking issues in practice
There is a limited amount of empirical research evidence of effective mental health MDT implementation in practice (Armstrong et al., 2007), with few observational studies on how teams of this nature operate available (Shaw et al., 2007) to verify claims, highlighted in the Irish context by Vision for Change (Department of Health and Children, 2006), that multidisciplinary mental health teams are the most effective means by which to organise patient service provisions. A review of the literature suggests that the organisation of contemporary mental health MDTs around the world differs significantly in practice from policy rhetoric, and is frequently characterised by fragmentation (Molyneux, 2001), with practitioners in Ireland finding the structure, function and practice of MDTs changing within and between teams from day-to-day depending upon the personnel involved (Deady, 2012). Interprofessional collaboration on such teams has been found to be inhibited by role blurring (Onyett, 2003), role rivalry, power struggle, perceived imbalances of power where professionals have felt disempowered due to the perception that their capacity to negotiate new ways of working is inhibited by medical dominance (Shaw et al., 2007), poor interprofessional communication (Bailey, 2012) and implicit adherence to differing models of mental health (Colombo et al., 2003) which can generate tensions, conflicting assumptions and misunderstandings between multidisciplinary groups about treatment provisions (Bailey, 2012). Professional allegiances (Onyett, 2003), coalitions and culture clashes (Shaw et al., 2007) based on stereotypical and ambiguous views of collaborators and their contributions have also been found to be other barriers to effective collaboration (Karben, 2011).
Working in teams of this nature generally tended to elicit two professional reactions. One response was an attempt to establish inflexible role demarcation based on 'jurisdictional claims', with a division of labour based on knowledge-based appeals of professional expertise (Peck & Norman, 1999), allowing some professionals to control tasks and activities against the claims of others (Hannigan & Allen, 2011) while also ensuring that differing professional interests do not overlap, making their position virtually impenetrable (Butler, 2005;Hannigan & Allen, 2011). Donnison, Thompson, and Turpin (2009) found that this phenomenon was particularly pronounced when clinicians found themselves 'feeling threatened' from other professions. The second response was to allow role blurring to occur-the tendency for professional roles to overlap when there is a shared body of knowledge amongst disciplines (Onyett, 2003)resulting in muddled role boundaries and overlapping responsibilities between professionals (Bailey, 2012).

Division of labour
As Pilgrim (2009) argues, the question of which mental health specialist shall treat whom, how and under whose aegis is of crucial importance. Strauss et al. (1981) and Pilgrim (2009) highlight how textbook perspectives tend to provide standard answers to this question, with treatment for mental health problems implying a traditional division of labour centred on medicine, for example, drug therapy flows from a medical diagnosis and prescribed treatment controlled by psychiatrists, with other professional tasks distributed amongst the team based on professional expertise.
However due to a lack of clear professional role and task allocation, the division of labour on mental health MDTs has been found to be a fluid process (Byrne & Onyett, 2010), with the jurisdictional areas of each profession requiring constant adjudication and negotiation (Strauss et al., 1981). Though differences in authority, power and status amongst team members may limit a professional's capacity to influence the team (Byrne & Onyett, 2010), it has been found that in order to retain their professional identity and attain individual freedoms over practice team members need to subtly negotiate a number of inter-and intraprofessional sensitivities on a daily basis (Scholes & Vaughan, 2002), with the personality of the incumbent a key variable in successful role negotiation (Deady, 2012). Reeves, Lewin, Espin, and Zwarenstein (2010) likened this process to a political one, involving negotiation, bargaining and the formation and dissolution of alliances, with Byrne and Onyett (2010) highlighting how the power to determine policy is not clearly located in specific positions and often shifts in response to different issues with the continuing dominance of any one profession not assured and the division of labour always open to conflict, negotiation and change (Strauss et al., 1981).

Effective collaborative practice
Understanding and appreciating professional roles and effective interprofessional communication have been found to be the two core competencies for patient-centred collaborative practice (Suter et al., 2009). In order to ensure effective mental health multidisciplinary teamworking, it is critical that teams contain professionals who have these two core competencies, and can collaborate and coordinate client and family care effectively (Mental Health Commission, 2010). However, while there is a widely recognised need to train health professionals in these two core competencies, the promotion of effective multidisciplinary practice within educational training and health care settings is rare in Ireland (Mental Health Commission, 2010). This lack of planned collaborative learning opportunities can lead to profession-centric practitioners experiencing very limited training opportunities to explore the roles of their teammates (Pecukonis, Doyle, & Bliss, 2008), and without this mutual interprofessional understanding, meaningful communication and relationships have been found to be more difficult to develop (Hall, 2005).
From a social work perspective, although clear guidelines and considerable knowledge exist about communication and management of professional-client relationships in social work education (Diggins, 2004), little attention has been given by social work educators and practice literature to the management of interaction and effective communication with other professionals (Specht, 1985). The Mental Health Commission (2010) also highlighted this as being the case in relation to social work education and practice literature in Ireland. Communication skills taught to social work students in Ireland usually focus on interactions with clients and their families from the perspective of their profession (Hall, 2005), with relatively little focus on interprofessional communication and collaboration skills (Mental Health Commission, 2010). Thus, SWs in Ireland may begin their careers with interprofessional barriers (Mental Health Commission, 2010), such as a lack of interprofessional communication, role negotiation and collaboration skills (Specht, 1985).
Interprofessional education (IPE) has been found to transcend uni-professional approaches to health professions education (Pecukonis et al., 2008). IPE emphasises interactive learning with and from members of other professions to improve interprofessional practice and care delivery (Barr, 2005). Offering IPE early at university level is considered key to the cultivation of an ethos of cooperation (Mental Health Commission, 2010), and to developing the core competencies outlined by Suter et al. (2009) of understanding professional roles and effective interprofessional communication.

Research design
In order to explore mental health MDT functioning the author employed a single case study approach, utilising interpretative phenomenological analysis (IPA). IPA is a strongly idiographic approach concerned with detailed analysis of single cases (Biggerstaff & Thompson, 2008). IPA tries to understand how participants themselves make sense of their experiences and explores individual personal perceptions of events (Smith, Flowers, & Larkin, 2009). This research design was chosen due to its ability to generate a detailed and in-depth understanding of a single multidisciplinary mental health team in the Irish mental health service (Gerring, 2007) through the professional perceptions and interpretations of those working on such a team (Krysik & Finn, 2010). In this case study, data were derived from open-ended semi-structured interviews with professionals from a single multidisciplinary mental health team and through nonparticipant structured observation of these professionals in four consecutive weekly MDT meetings.

Participants
Sampling for this study occurred at two levels. First, in line with the IPA and case study methodology, a small sample of five research participants were purposively selected (Berg, 2009). The criterion for selection was being a CP, SW, CPN, psychologist or occupational therapist (OT) on the same mental health MDT. This sampling technique was chosen in order to ensure that those interviewed were particularly knowledgeable about the research questions being explored, which would lead to a higher quality of data being obtained (Grinnell & Unrau, 2011).
Second, in order to gain access to a mental health MDT the author used convenience sampling by initially approaching the principal mental health SW in a general hospital in Ireland to discuss the research proposal. The principal SW then approached each mental health MDT's CP (the clinical lead on each team) to discuss the proposal. Preliminary access to a team was granted by one of the CPs on the condition that all of the team members consented to taking part. After discussing the research proposal with this team, all five staff members on the team agreed to participate in the study, the CP, SW, CPN, psychologist and OT. The mean number of years since qualification was 7 years, with the CPN being an outlier, with 18 years' experience.

Data collection
Observations and semi-structured interviews are the most commonly used data collection methods in an IPA approach (Smith et al., 2009). The author utilised structured observation, by sitting in on and observing four consecutive weekly inpatient MDT meetings. The inpatient team meeting was selected as it would provide the best opportunity to observe an environment in which each professional had the opportunity to contribute to the assessment and development of care plans for each patient. During this non-participant structured observation, professional involvements in patient cases and types of interactions between professionals during the team meeting were noted quantitatively on a pre-prepared observation schedule. Using the research questions and literature review as the coding basis for the observation schedule (Berg, 2009), behaviours and interactions to be observed were categorised according to: (a) the profession of the speaker, (b) which bio-psycho-social aspect of the patient's life was being discussed during assessment, (c) which model(s) of mental health were used in the patient assessment process and in the development of patient care plans, (d) incidents of interprofessional collaboration, (e) incidents of interprofessional disagreement, (f) incidents of interprofessional role negotiation, and (g) incidents of hierarchical working arrangements during the team meeting. The number of incidents in each category was formatted proportionally and reported in the findings section as percentages. Field notes of the researcher's observations were also recorded. In this context, the medical aspect to be observed comprised of statements about the patient's medical or pathological condition and possible medical treatment interventions such as medication (Blomqvist & Engstrom, 2012). The psychological aspect comprised of statements about the patient's psychological condition such as personality, cognitive and affective functioning and possible psychological interventions such as psychotherapy (Blomqvist & Engstrom, 2012). The social aspect comprised statements about the patient's living environment and concerned information such as employment, housing, family and social networks and possible interventions such as family therapy (Blomqvist & Engstrom, 2012). Professional involvement in patient cases was coded by how often a professional's formulation on the patient's assessment and suggested treatment intervention was noted in the patient's care plan. These observational categories were pre-selected in order to ascertain to what extent: (1) each professional was involved in patient assessment and care planning, (2) there were differing levels of professional input into these processes, (3) each model of mental health was used in the formulation of patient assessments and care plans, (4) any model(s) of mental health dominated in these processes, (5) professionals exhibited dominance of, or submission to, other professionals on the team, (6) team members disagreed on patient assessment and care planning decisions, (7) team members collaborated and/or negotiated their professional roles in order to ensure that patient assessments and care plan actions were completed.
The semi-structured interview contained 21 open-ended questions about the experience of mental health multidisciplinary teamworking, exploring four of the main themes derived from the literature: (1) mental health as a contested phenomenon, (2) multidisciplinary teamworking issues in mental health, (3) division of labour on MDTs, and (4) effective collaborative practice. The questions asked were open-ended in order to allow the participant to set the parameters of the topic (Smith & Osborn, 2003). Interviews, lasting between 60 and 90 minutes, continued until such a point that all areas of interest had been explored. As a part of these interviews, data and the general themes generated by the observation schedule were also verified with each participant, along with their interpretations and clarifications on some of their input during the team meetings.

Data analysis
The analysis of the semi-structured interviews was conducted according to IPA methodology. Transcripts of each interview were read several times and preliminary summary notes were made. Transcripts were then reread and preliminary themes were identified. A master list of themes was produced for each transcript. Master lists were compared and a final integrative set of themes emerged (Smith & Osborn, 2003). All of the semi-structured interviews were then analysed using the final integrative set of themes as a coding frame. In order to analyse the observational data collected from the structured observations, a content analysis was employed. This process initially involved the categorisation of data from the field notes into the same categories as already outlined in the observation schedule. The frequency and duration of the various verbal and nonverbal behaviours observed and noted using the observation schedules were then coded in the analysis (Grinnell & Unrau, 2011). The coded field notes and coded data from the observation schedule were then examined and interpreted in order to identify patterns, themes and meanings (Berg, 2009). Six master themes, discussed below, emerged from the data. These were 'team meeting structure', 'contested nature of mental health', 'multidisciplinary teamworking issues in practice', 'division of labour', 'effective collaborative practice' and 'factors specifically impacting the social worker'.

Limitations
Not all mental health MDTs will have exactly the same characteristics and will contain their own idiosyncrasies. Therefore, using purposive and convenience sampling (which are non-probability sampling approaches) and a single unit of analysis case study design means that the findings from this research cannot be generalised to a larger population of mental health MDTs (Grinnell & Unrau, 2011).

Ethics review
Ethical approval was sought and granted from the ethics committee of the School of Social Work and Social Policy in Trinity College Dublin. In order to attain informed consent, an information sheet about the purpose of the study was provided to each participant to sign. The information sheet explained how the data would be used, what participation would require of each potential participant, gave assurances on confidentiality and anonymity (which was attained by omitting the names of the research participants and the mental health setting when reporting data and keeping records of any interactions) with the proviso that the participant could withdraw from the study at any time, or responses to particular questions could be omitted in the semi-structured interview.

Findings and discussion
Team meeting structure The MDT meeting had the same format each week. Prior to the patient coming into the team meeting, the patient's case was presented to the team by the CP. The team then discussed the patient's case history and tried to identify priorities in the patient's care. The patient was then brought to the team meeting and questions were asked by the CP of the patient based on the prior professional discussion and the priorities identified. When the patient left the team meeting, the team then discussed the patient's responses and how the patient appeared in the meeting and from there an assessment was formulated and a care plan was created. In total, 36 patients were seen during the team meetings observed. The CP saw every patient. The SW was present for the duration of all of the team meetings, but did not see two patients because the CP saw one in the corridor outside on their own, and on another occasion the CP saw the patient with the CPN before the team meeting. The OT and the CPN saw 28 and 21 out of the 36 patients, respectively. This was due to the fact that each professional had competing oneto-one patient appointments at the same time as the team meeting. Due to resource constraints the psychologist, who was working across two MDTs and only attended team meetings every 2 weeks, saw 18 out of the 36 patients. The input levels of each professional were formatted proportionally when reporting findings, for example, the psychologist's input level was multiplied by a two as they attended half of the meetings.

Contested nature of mental health
The interview and observation data highlighted a consensus amongst team memberssupporting Herron and Mortimer (1999) and Rogers and Pilgrim (2005)-that the concept of mental health was contested on this team with different team members utilising differing models of mental health. The CP and CPN both utilised a medical model of mental health. The SW and OT utilised either a social or psycho-social model depending on the patient's needs. The psychologist utilised a psychological model. There was also a consensus amongst the team that the medical model was the most dominant framework used in team meetings, supporting Kirk (2005), with the SW outlining how 'oftentimes medication can be the sole type of treatment' (SW). The dominance of the medical model was supported by data derived from the observations. During assessment and care planning phases of the team meetings, the medical model was utilised 65.02% of the time, the social, psychological and combined psycho-social models were used 34.98%. These findings support Colombo et al. (2003) who highlighted how the decision-making process regarding the management, care and treatment of patients is contested, based on the individual's practice perspective of mental health, with the social, psychological and psycho-social models of mental health competing for authority and recognition alongside the traditional medical approach.

Multidisciplinary teamworking issues in practice
Vision for Change (Department of Health and Children, 2006) recommends that all members of a team should be treated equally, with the CP only acting as a nominal leader who coordinates the work of the team. However data from the interviews highlight how the team studied did not conform to this egalitarian model, but was hierarchical with the CP dominating decision-making on patient management: I don't think everybody's view carries equal weight, multidisciplinary working is very consultant centred, so I have the final say on treatment and on who I feel requires the team's attention. (CP) This finding was further evidenced by data from the observations, when on two occasions the CP unilaterally decided that two patients would not be brought into the team meeting, as he felt that the patients did not need input from the other professionals on the team and only required changes to their medication maintenance. This phenomenon highlights a medical bias in treatment orientation on the team which fails to comply with recommendations in Vision for Change (Department of Health and Children, 2006) by reducing the opportunity for collaborative bio-psycho-social discussions on patient assessment and care planning and reducing the range of professional expertise available to patients.
The hierarchical working arrangements embedded in the structure of the team appear to be exacerbated by the level of collaboration and coalition which exists between the CP and the CPN. This phenomenon, outlined by the psychologist, was found to have a negative impact on the level of professional collaboration amongst the team as a whole and led to the team effectively being divided into medical and psycho-social halves: The CPN and the Psychiatrist are closely linked, and work through medication … it would be fair to say that this has a team-splitting effect with the CPN and consultant on one side and psychology, social work and OT on the other. (Psychologist) The strength of this coalition, coupled with the hierarchical working arrangements, impacted the extent to which other professionals felt that they could get involved and thus reduced the potential for genuine team collaboration. This is highlighted in the data derived from the interviews with the SW and OT: It is such a medical and consultant-CPN led team, it can be very difficult to get involved. (SW) They have a powerful dynamic together … whenever one of them leaves the room then it sometimes gives the rest of the group a space to put things out there. (OT) The dominance of the team by both the CP and the CPN is further evidenced by data derived from the observations, which showed that between them the consultant and CPN had 73% of the entire input into patient treatment discussions, with the OT having 12%, the SW having 10% and the psychologist having 5%. These findings support Shaw et al. (2007), who found that in general non-medical professionals, such as SWs, psychologists and OTs, felt that their capacity to collaborate with colleagues and to negotiate alternative social and psycho-social ways of working was inhibited by medical dominance and hierarchical working arrangements.
The interview data highlight that though the MDT was found to be formally CP and informally CPN led by virtue of the CPN's longevity within the team, the structure of the team meetings was found to be a fluid process. The mechanics of how the team functioned were not set in stone each week but fluid depending on who attended the team meeting, with professionals on occasions not attending the meeting, or attending parts of it due to the differing levels of priority given to the meeting: You don't ever know for sure who will be there each week … one week it was just the consultant and me, and other weeks it might include us all. (SW) The team meeting wouldn't be top priority … if I have a patient appointment I would prioritise that. (CPN) The team dynamics depends on the people that are there … If someone leaves the room, they can change, and offer the team a space to bat something out differently. (OT) This fluid nature of the team's dynamics is backed up by the data from the observations of the team meetings, with the involvement of the psychologist, SW and OT in patient treatment decisions increasing significantly by 60%, 66% and 67%, respectively, when the CPN was not at the meeting. This finding concurs with Strauss et al. (1981) and Scholes and Vaughan (2002) who found that MDTs formed around patients are amongst the most fluid in organisational structure, with the jurisdictional areas of each profession requiring constant adjudication and negotiation. The finding also supports Deady (2012) who highlighted how the working arrangements of such teams are strongly impacted by the differing professional personalities and by which professionals were present at the team meeting.
The CP illustrates how the fluid nature of the team can impact treatment provision: It can impact treatment quality, if you want to get everyone's professional view on a patient and that professional isn't there or they go early, it's a problem. (CP) It is clear from these findings that the Vision for Change (Department of Health and Children, 2006) ideal of full weekly bio-psycho-social discussions on patient assessment and care planning is not being carried out in practice. Practitioners find the structure, function and practice of the team changing from day-to-day depending upon the personnel involved, with different disciplines attending team meetings with more regularity than others, thus impacting the range of expertise available to patients each week.

Division of labour
The division of labour on this team was found in the interview and observation data to be in line with Vision for Change (Department of Health and Children, 2006) and Pilgrim (2009), as being based on professional expertise. When the division of tasks could not be allocated based on professional expertise, for example, when roles were blurred, the team member who was deemed to be best acquainted with a patient carried out necessary tasks. This division of labour strengthened the CPN's position, who being the first point of contact with patients through his direct work within the community was found to act as a gatekeeper, by allowing or restricting other professionals from getting involved in the patient's care: The CPN is a gatekeeper, he may or may not encourage someone to attend psychotherapy, and because of the patient's relationship with the CPN this will influence their decision. Individual personality factors, which were observed during the team meetings, further strengthened the CPN's position, specifically the apparent comfort with which the CPN copes with conflict. There were five strong arguments about patient treatment provision within the team meetings, all of which were initiated and eventually won by the CPN. An example, similar to the ones observed within the team meeting, of how the CPN utilised this type of role protection strategy is offered by the SW in their interview: Once, I disagreed with a client being sent to a hostel that I felt was completely inappropriate … So I brought it to the team meeting and the CPN said 'you need to stop worrying about this, you've got this thing in your head, I don't know what you're trying to do'. I was so floored that I couldn't say anything. (SW) It is clear from the interview and observation data that the CPN, who is the most experienced, has consolidated his powerful position in the team by negotiating a strong working coalition with the CP. Through his greater influence and power, the CPN has been able to control the division of labour on the team and has been able to maintain a position of dominance easily by simply blocking the SW who feels marginalised and alienated. These findings fit neatly with Peck and Norman (1999) and Hannigan and Allen (2011) who found that when professionals perceive a threat or substitution of their work by other professions, some so strongly defend their positions, utilising their personality and experience, that their position becomes virtually impenetrable and this can lead to reduced collaboration and input from other team members within the team.

Effective collaborative practice
It is clear from the interview and observation data that professionals on this team did not have the two core competencies of understanding teammates professional roles and effective team communication which Suter et al. (2009) identified as the two core competencies for patient-centred collaborative practice.
Lack of communication between the team and clarity on each other's roles are the two biggest inhibiting factors in the team's functioning. (SW) The interview data highlight how each professional's understanding of their colleagues' role was quite superficial and stereotypical, and this certainly appears to be the case in relation to social work: Role blurring between the SW and the CPN, and role stereotyping on the team was a consistent finding in the interview and observational data. This role blurring appears to be due to the fact that the social work role is not understood on the team, because there is a lack of communication between the CPN and the SW about their respective roles and the fact that the CPN, unbeknownst to the team, does not define himself as a traditional CPN, but as 'a holistic worker, a mixture of OT, social worker and psychologist' (CPN). These findings also support Onyett (2003) and Karben (2011) who found that role blurring and stereotyping inhibited the potential for professionals, in this case the SW, to offer their professional skill sets and knowledge to patient cases.

Factors specifically impacting the SW
As already outlined the SW's expertise on this team was clearly reduced by a lack of understanding of the social work role in mental health, and by role blurring with the CPN in an environment where the CPN has negotiated a dominant position as a 'holistic worker' in a team which failed to differentiate between the SW's and CPN's skill sets. The strong impact the discourse of 'who knows the patient best' has on the division of labour within the team, also benefits the CPN-who has direct access to patients in the community and is the longest serving member of the team-and further reduces the social work contribution. These factors led to the CPN carrying out what are generally perceived as social work tasks. An example of this can be seen in how during two of the observations the CPN took over the role of assessing suitable post-discharge accommodation within the community for two patients, traditionally a social work task due to their expertise in systemic psycho-social assessment. This reduced the quality of the provision offered to the patient, and meant that the team failed to comply with Vision for Change (Department of Health and Children, 2006), as the expert in the field in this case the SW did not carry out these interventions.
The interview data highlight how the multidisciplinary mental health team did not have formally agreed role or job descriptions. Instead, in line with Scholes and Vaughan (2002), professionals such as the psychologist and OT created their own place and maintained their professional persona within the team through a process of negotiation: I've taken responsibility to consistently negotiate, create my role and give people the language of OT within the team. (OT) I think it's up to each discipline to stake their claim, by interjecting and presenting their thinking. (Psychologist) The interview data highlight how the SW, however, was found to be less inclined to negotiate, supporting Peck and Norman (1999) and Hannigan and Allen (2011), by arguing for a system of professional demarcation in an attempt to bring clarity to the team and to avoid overlapping of roles.
No I haven't negotiated my role. I've never formally said 'this is what I do'. Negotiating my role is not something I have professional experience of or was taught in college … I'd rather that the CPN, consultant [as mediator] and I sit down formally and discussed our separate roles … rather than try to find a negotiated compromise. (SW) The reluctance of the SW to negotiate in a more sustained and consistent way, much like the OT has outlined, is likely to lead to social work continuing to be marginalised within the team. This reluctance to negotiate could be explained by Specht (1985), Diggins (2004), Hall (2005)

Conclusion
In summary, this study has addressed its own aims by exploring several key aspects of mental health multidisciplinary teamworking in a single team in Ireland. The analysis revealed several themes which demonstrate the complexity of multidisciplinary teamworking, along with potential barriers to the effectiveness of this type of service provision and to social work's input on such teams in Ireland. The study was carried out against an Irish policy background where it has been consistently recommended that mental illness should be conceptualised in an integrative way, through the use of a biopsycho-social model of mental illness within mental health MDTs, reflecting the varying dimensions of the phenomenon. Multidisciplinary teamworking has been lauded and recommended as the ideal form of mental health service delivery, achieved within an egalitarian working environment, where patients' needs are met through enhanced professional collaboration and the utilisation of a variety of professional skills and knowledge bases. However, the empirical research reviewed for this research project generally has revealed a much different reality. This study confirmed what the sociological literature has consistently suggested that mental health practice is a contested area of human activity rather than an area characterised by scientific or ideological consensus, with each team member on this team found to subscribe to and advocate for a different model of mental illness based on different educational and professional socialisation experiences in an environment where the medical model was dominant over psychological, social and psycho-social perspectives. Team members were found to be lacking in the core competencies for patient-centred collaborative practice, which include understanding each other's roles and communicating effectively, and differed fundamentally on each other's perceived professional role boundaries and on how cases should be distributed, as well as on modes of mental health treatment and management. The power dynamics and working arrangements on the team were found to be fluid based on who was at the team meeting, which varied from week to week. Professional rivalry, hierarchical working arrangements, a professional coalition between the CPN and the CP, team member non-participation, role protectionism, professional gatekeeping, role blurring and stereotyping on this team ensured that the organisational order within which this team organised and delivered mental health services can at best can be described as fragmented, with the potential for genuine collaboration between professionals on the team reduced due to these factors. The barriers to effective mental health multidisciplinary teamworking identified in the research literature, and further evidenced by this study, are strongly influenced by the lack of opportunities for mental health professionals to learn and work together in meaningful ways as part of their professional training in Ireland (Mental Health Commission, 2010). It is clear that in order to improve interprofessional practice and care delivery that these issues must be addressed through increased opportunities for mental health MDT professionals to be involved in IPE and interactive collaborative learning environments, in which team members can learn with and from each other, and develop the core competencies and skills required for patientcentred collaborative practice (Barr, 2005). In a practice environment where there are no clear job descriptions, the SW on this team had to contend with an additional interprofessional barrier to collaborative practice-not encountered by other team members-namely, a lack of interprofessional role negotiation skills. This additional barrier, which has been a historical issue for Irish mental health SWs (Butler, 2005), appeared to further limit the extent to which the SW on this team could engage in theoretical discussions on patient assessment and treatment and contribute their skill set to the team. This skill deficiency in interprofessional role negotiation skills (Diggins, 2004) along with the lack of the core competencies for patient-centred collaborative practice (Suter et al., 2009) exhibited by the SW on this team have been identified in the research literature as barriers to social work effectiveness within MDTs. The existence of these barriers is clearly influenced by the fact that social work in Ireland has become preoccupied by social work specialisms which are largely dominated by statutory regulation, such as child protection and probation (Butler, 2005). This has been reflected in the priorities of social work educators, with limited mental health social work education and skills training and no IPE in the areas of understanding interprofessional roles, team communication and interprofessional role negotiation currently available to Irish social work students (Mental Health Commission, 2010). It is clear from the research literature, which is further evidenced by this study, that in order for Irish SWs in mental health MDTs to be effective in their collaborative practice, they must be offered the opportunity to develop the competencies and skills required for effective interprofessional collaborative practice. This can be achieved through SWs being allowed increased access to IPE and mental health social work education and skills training at university and practice levels (Mental Health Commission, 2010).

Notes on contributor
Alan Maddock is a recent graduate with distinction (2012) from the Masters of Social Work in Trinity College, Dublin and is a registered member of the Irish Social Work regulatory board CORU.