The (im)possibilities of clinical democracy

HEALTH SOCIOLOGY REVIEW

The (im)possibilities of clinical democracy observations, including six months of video observation, were undertaken by a medical anthropologist. Clinic sessions, team meetings, case conferences, teleconferences, and staff inservice training sessions were observed for four months, and then videoed. In total, approximately sixty hours of video data were collected. Themes explored in the data analysis were developed by both researchers and clinicians, and included temporal-spatial analysis, exploration of formal, informal and nonformal modes of communication (with a resultant focus on corridor communication), infection control practices and organisational analysis (Iedema et al. 2005a;Iedema et al. 2006a;Lee et al. 2005;Long 2005;Long et al. 2006;Long et al. in press, a;Long et al. in press, b;Pontivivo and Long 2006). Findings were discussed with the clinical team in regular research feedback sessions, which took place on average every two months throughout the project. These research feedback sessions were also videoed. Quotes from clinicians in this article are excerpts from video or audio transcripts from this project, in many cases from reflections expressed in the research feedback sessions.

All clinician names used are pseudonyms. The term, 'clinicians', refers to all health care workers on the team: medical, allied health, nursing and surgical. The project was undertaken with ethics approval from the authors' university and from the hospital's health departmental ethics committee.

The (im)possibilities of clinical democracy

Challenges to medical dominance have now been extensively documented (e.g., Boyce 2001;Braithwaite and Westbrook 2005;De Voe and Short 2003;Flynn 2002;Germov 2002;McKinlay and Marceau 2002). McKinlay and Marceau make a claim for the 'decline of the golden age of doctoring ' (2002:379) in the USA on the basis of, among other things, increasing corporatisation and market control over doctors' practices (McKinlay and Arches 1985;McKinlay and Marceau 2002;McKinlay and Stoekle 1988). In the UK, changes in government regulation and health system governance are credited with diminishing clinical autonomy and increasing the visibility of what doctors do (Britten 2001:492).

In Australia, challenges to medical dominance have arisen through patient safety concerns (Germov 2002:293;Iedema et al. 2006bIedema et al. :1605 and increased health consumerism (Germov 2002:294;Iedema et al. 2006bIedema et al. :1606. According to Germov, it is: '[m]anagerialism in its general and clinical modes', as formulated in the introduction of Australia's clinical governance policies, which 'may represent the most effective challenge to medical dominance to date' (Germov 2002:300). Clinicians are increasingly expected to undertake administrative-managerial and organisational roles, and to engage in practice measurement and improvement and clinical error investigations (Boyce 2001;Braithwaite and Westbrook 2005;Iedema et al. 2006b). These organizational reforms have resulted in a 'reduced dominance of the previously universal medical model' (Boyce 2001:22) and shifted doctors' 'power vis-a-vis nurses and allied health professions' (Braithwaite and Westbrook 2005:11). This shift has not been even: although medical and nursing voices are now often well represented in management structures within Australian hospitals, allied health is less well represented at key decision making levels (Braithwaite and Westbrook 2005;Rowe et al. 2004:17).

This unequal representation has particular implications for the dynamics of multi-disciplinary health care teams, which are increasingly needed to provide complex, multi-factorial care to chronically-ill patients (Colombo et al. 2003;Latham et al. 2000;Long et al. in press, b;Sherer et al. 2002;Wagner 2000). Clinicians in these teams are required to be experts in their own area of 'content knowledge', and are also expected to be able to effectively communicate in other clinicians' areas of content expertise. In order to effectively communicate across disciplinary boundaries, clinicians must negotiate the delicate task of destabilising established hierarchies (Cott 1997;Gair and Hartery 2001;Long et al. in press, b).

Although much macro-and meso-level analysis has been undertaken on health professions and medical dominance, Elston (1991) and Carmel (2006) note that on a micro-level practices and strategies of medical dominance remain relatively unexplored. Based on ethnographic research undertaken within a small multidisciplinary clinical team, this article offers a micro-level analysis of practices of, and resistances to, medical dominance. This article illustrates the complexity and embeddedness of values of medical dominance in health workplace culture. While confirming that organisational and policy change on their own cannot shift deeply enculturated behaviours and norms, this article also suggests that emerging ways of working appear to be disturbing professional hierarchies among hospital clinicians.

Clinical democracy: an imagined possibility

Sociological critiques of medical dominance frequently focus on the resulting empowerment of some health care professionals and the associated disempowerment of others (e.g., Wearing 1999). Situated within the subtext of a social justice framework, dominance of any kind is regarded as morally problematic (Dzur 2002;Germov 2002;Wearing 1999;Willis 1983). Within the clinic we will discuss, however, the dominance of any single professional voice is not rejected (solely) on the basis of interprofessional power politics. Medical dominance is rejected by clinicians, the doctor included, primarily on the basis of clinical outcome. Muting of non-medical clinical voice (nursing, allied health and community health) is seen to compromise We use the term clinical democracy to describe the aims of this multidisciplinary hospital-based health care team in which communication and decision-making structures were deliberately designed to be 'flat', that is, non-hierarchical (Lee et al. 2005;Senge 1990), and in which team members actively worked both to subdue the dominance of the medical voice and to enhance nursing, allied health and community health voices (Long et al. in press, b). We discuss how, even within a clinical team where medical dominance is consciously recognised and actively opposed, challenges arose to implementing modes of communication and decision making that were clinically democratic. We show that while the medical professional who established the team aims for a participatory democratic structure, in which everyone has equal say, he at best achieves a representative democratic structure in which the team supports and has faith in their (s)elected (medical) representative (Long et al. in press, b).

The clinic

The clinic in question provides care for people with spinal cord injury who have pressure ulcers. The pressure area clinic is held twice per month as a hospital outpatients' clinic. The clinical team on hand for each outpatients' clinic is comprised of a spinal medical specialist, a spinal occupational therapist, a spinal physiotherapist, a wound care clinical nurse consultant, a social worker, a dietitian, and a spinal peer support worker. In addition, a number of specialists are on call for the clinic, including an orthopaedic surgeon, a plastic surgeon and an infectious diseases staff specialist.

The multidisciplinary clinic is a relatively new innovation in pressure area care. Previously, patients with pressure areas were referred to clinicians for separate appointments. This had disadvantages for both patients and clinicians. Clinicians worked in greater isolation to treat their narrow aspect of what was inevitably a multifactorial issue, and patients had to arrange multiple visits to see these practitioners, with the associated time and transport complications involved for people with spinal cord injury. Although it results in complex, crowded and often chaotic clinic sessions, and although these sessions are tiring for both clinicians and patients, the multidisciplinary approach is generally preferred by both patients and clinicians, and shows markedly improved clinical outcomes. For patients with areas requiring surgery, the average time spent in hospital was reduced from 264 to 54 days. 1 The cost benefits to the hospital system are significant: in this hospital, the cost of treating a pressure area surgically, for someone who had not come for pre-operative consultation in the clinic, was on average $198,000 per patient. The average cost for patients who came through the clinic was less than a quarter of that, $42,000 (Iedema et al. 2006a). 2 The clinician who established the clinic model, a doctor, is passionately committed to multidisciplinary practice, as the following comment illustrates:

When you look at the medical problems, or the problems that occur after someone is discharged from hospital, most of those problems are actually preventable, and most of those problems are social, environmental issues. Usually [patients are] clear to go on a very narrow medical filter, whether they're going to live or die, whether or not their blood pressure's stable or whatever. But things that make them fail [that is, have pressure areas] are these other issues, and they just haven't been given voice.

(Kim, Doctor, research feedback session)

As one measure of evaluating the organisation of the clinic, Kim facilitated the involvement of the clinic in an ARC (Australian Research Council) funded research project utilising video ethnography to explore clinician identity in multidisciplinary teams.

Method

The data that this article is based upon was collected as part of the above-mentioned project. From August 2004 to May 2005, ten months of

Vocal dominance and muting

As we have shown elsewhere (Long et al., in press, b), in spite of the best intentions of the team, there were a number of internal and external factors which challenged working in a clinically democratic structure. While openly encouraging multivocality, the doctor who established the team was vocally dominant in many of the team's interactions. The medical voice dominated both in care settings where the patient was present (in procedure and consultation rooms) and in case conferences and team meetings, where only clinicians were present. In one particular team meeting involving nine clinicians, including allied health, nursing and medical staff, the only doctor in the room spoke for 43 out of the 61 minutes (70% of the time). This was not atypical. When shown the 'talk time' figures in a research feedback session, the doctor in question held his head in his hands. He later commented to the researcher: 70% of the time [...] being medical is possibly going to compromise patient safety … that density of medical talk time is a problem... [we need] to allow some of the other light to shine on the subject.… it's going to take some self-monitoring on my part ... [otherwise] we're going to lose information. I think we're going to potentially make worse clinical decisions because the different co-factors that lead to failure are not going to be properly expressed.

(Kim, Doctor, interview)

Non-medical team members were often vocally subordinate in meetings. Claire, the Clinical Nurse Consultant, commented on the weekly case conference on the ward, a meeting regularly attended by more than twenty people, including nursing, physiotherapy, occupational therapy, social work and medical staff. This centrality of the medical voice in formal meetings was evident when clinicians viewed a section of video that showed a team meeting. One segment showed a situation where Kim's mobile phone rang. As he answered the call, all the other team members shifted their body language, 'in', towards the centre of the table, and started talking in a totally different way. Although the topic of the talk was still work related, the tone was different, with lots of giggly laughter, in contrast to the more formal professional 'meeting-speak' used just seconds before. The researcher asked the team to comment on the video footage.

Nicky Mobile phone usage also illustrates professional asymmetry within the team. All team members leave either their mobile phones or pagers on in meetings, and all team members will respond to phones or pagers when they ring. Kim's phone rings most often, he will talk longer when he answers calls, and the meeting will 'stop' and resume when he finishes. When other team members take calls, they either move outside, or take them quietly, and the meeting continues without them.

As with any form of social dominance, medical dominance exhibits aspects of Hegelian master/ slave dialectics (Hegel 1977:191), in that people in subordinate positions in hospital hierarchies will always know more about some things than people occupying dominant positions in the social/ professional hierarchy. The team reviewed video segments that showed interactions between the patient and clinical staff when there was and was not a medical presence in the room. The nonmedical clinicians in the research feedback session had all experienced both types of interactions before. For Kim, however, this was the first time he had seen the types of interactions that occurred in the clinic when neither he nor the surgical members of the team were in the room. This led to a discussion about the types of things that medical staff may not get to know about, which may hinder optimal patient outcomes.

Kim Lupton claims that despite consumer-driven challenges that have been made to the 'ideal figure' of the paternalistic doctor operating from a framework of medical dominance, 'the position of doctors is still powerful in terms of patients being unable to react assertively ' (1997:376). Britten also notes passivity of patients, stating that in interactions with GPs 'patients rarely articulate their whole agenda … and do not ask questions, even when encouraged to do so ' (2001:488). Similarly, although they were overtly encouraged to speak up, as we have shown, a number of non-medical team members struggled with Unlearning Not To Speak (Piercy 1973), and maintained 'muted group' behaviour (Ardener 1975;Morrison and Milliken 2003) in team meetings, case conferences and other formal communication situations. In formal communication, the doctor in the team found it difficult, despite his best intentions, to break away from established patterns of vocal dominance, and the other members of the team found it equally challenging to break away from patterns of vocal subordination.

Complexity and informality of communication

One of the most striking aspects of the data gathered in the project was the quantity and complexity of the informal and non-formal (workrelevant but unstructured) communication in the clinic. The team communicated with each other in team meetings, case conferences, by email, text and telephone, and 'bedside' -while they were consulting with the patient. In addition, significant decisions regarding the management of patients' cases were often undertaken in nonformal, fluid, dynamic corridor conversations that took place during the clinic. The corridor provided space for what Hardt and Negri term 'immaterial labour' (2000), the communicative, emotive and affective work increasingly required of clinicians involved in complex care (Iedema et al. 2005b;Long 2005). Although all the clinicians described the corridor conversations and movement during the clinic as chaotic, they were in agreement about the value of the informality and nonformality.

I think the informal ad hoc system seems to work really well, there are problems with it, I agree, but in general, because I mean we're all adult, or most of the time we're adult.

[laughter] I think we feel that we can all say what we feel is important.

(Don, Social Worker, research feedback session)

In the corridor the proportion of talk time was much more evenly distributed between the various clinicians. Unlike in meetings and with patients, in the corridor the medical specialist in the team did (almost) as much listening as talking. Analysis of the communication structures of this team have emphasised the potential of non-formal, fluid communication structures to facilitate multivocal interdisciplinary communication (Iedema et al 2005a;Long et. al. in press, a).

Leadership

In a research feedback session, the researcher asked the team how they felt about the corridor conversations in the clinic (Long et al. in press, a). This question led to a comment about leadership, and the ensuing discussion showed that while all of the clinicians were in agreement about the importance of a clinically democratic structure, they were differentially engaged in terms of how to achieve it (Long et al. in press, b Kim's concer n over the long-ter m sustainability of the clinic, and the necessity to establish communication structures that worked to cross professional boundaries irrespective of the individual clinicians occupying those roles, came up repeatedly during the research. Furthermore, this concern over the sustainability of the clinic was also reflected by the team's collective doubt that other doctors could operate with the team in the way Kim did. The non-medical team members' concerns have so far proven to be valid: two other doctors were trialled in the clinic, neither of whom proved to be viable long term alternatives to Kim.

In addition to the internal challenges discussed above, the team experienced external challenges to enacting a clinically democratic communicative structure. Discussed below are inequitable valuation of specialist knowledge; privileging of medical time; and the authoritativeness of the medical voice within broader hospital communications.

Specialisation

Not only Kim, but also the other team members expressed concer ns around the clinic's sustainability, given the key roles played by both Nicky [physiotherapist] and Liz [occupational therapist]. These were core members of the team who had established the clinic with Kim, and were seen as irreplaceable both because of their specialist spinal knowledge, and their long familiarity with the clinic's patient population. Nancarrow and Borthwick (2005) note that specialisation is well recognised and legitimated in medicine, and that specialisation for doctors is associated with greater autonomy and social prestige as well as increased financial rewards. In contrast, they argue, there is little evidence to suggest that these benefits arise for clinicians from other health-related professions who specialise (2005:907). These inequities were glaring in the case of this team. All three clinicians who were highly specialised in spinal care -the doctor, the physiotherapist and the occupational therapistworked far longer hours than they received credit. They were all regarded as having unreasonable workloads, even within a system (the Australian public hospital system) where unreasonable workloads are regarded as the norm. However, the doctor carried the title, staff specialist, and earned a salary far greater than that of the occupational therapist or physio-therapist. Neither of the allied health specialists was given additional 'rank' because of their expertise, and it was generally agreed by colleagues that their level of expertise was not reflected in their salaries. While the doctor's specialisation allowed him additional professional autonomy, the allied health professionals' additional expertise was often a site of contestation within their units. The occupational therapist was required to undertake delicate juggling acts to gain permission to follow up on her insights and expertise outside of the (often narrow) job description her department held for her, and the physiotherapist at one stage took annual leave specifically to catch up on paperwork, as the immense workload generated by her specialist expertise was not recognised by her department.

Valuation of time

In clinic, there were distinct waiting hierarchies: surgical waits for no-one, medical waits for surgical, but not for allied health or nursing, physiotherapist and occupational therapist wait for medical and surgical, but not for nursing, social work, peer support or dietician. Nursing waits for surgical, medical, physiotherapist and occupational therapist and occasionally for social work and dietician, but not for peer support. Social work, dietician and peer support most frequently do their work when everyone else is finished, or when the team are waiting for people further up the hierarchy to arrive. The surgeons involved in the team did not stay for the whole clinic, but were only on call when needed. When the research project began, all team members identified the physiotherapist, Nicky, as having the largest patient workload of the team. Yet Nicky frequently waited until medical and surgical clinicians were finished before she could do her work with the patient. One of the changes that was instigated as a result of the research project was that Nicky started working on call, like the surgeons. That change was logical when looked at in organisational terms, but the underlying logic of clinical hierarchies appears to have prevented the team from envisaging it prior to their involvement in the research project. Additionally, filling in time sheets was a often a problem for all of the team's allied health clinicians, but especially for social work, occupational therapy and peer support, as much of their work is done in ways which make it difficult to quantify. As we have stated elsewhere:

While the clinical team values this lack of structure, they are still under pressure to record their activities within formal hospitaldefined categories. Justifying people's time expenditure is a major challenge in gaining funding for multidisciplinary clinics, and funding for such a large complex team is an on-going vulnerability to the sustainability of the team. Facilitating nonformal channels means, ironically, structuring in unstructured time. Accounting for every minute of every clinician's time in the clinic is an impossibility, but it is an impossibility required of the funding structures of the Australian health system. (Long et al. in press, b) In terms of hospital accounting practices, many of the tasks which allied health and nursing staff undertake are valued differently, or are not provided with a descriptive vocabulary so that they can be measured, counted, and accounted for in hospital budgets. Allied health and nursing time is treated as vastly subservient to medical time, and there appears to be a direct relationship between waiting hierarchies and billing hierarchies: the further down the waiting hierarchy a health professional is, the more difficulty they may have in accounting for their time in ways that 'count' in budgetary terms. In a team such as this, where relatively informal communication is crucial to deciding on patient management plans, this makes funding these positions even more difficult than they already are, and adds to the potential for unrecognised workloads such as were found in this team.

Authority, responsibility and accountability

However democratic the decision making structures within the team may be, the members of the team still have to operate, as a group and as individuals, within the broader organisational structure of the hospital. This is where the representative aspect of Kim's role is most apparent. Even if he accepts that the occupational therapist, for example, may have the expertise to recommend a certain type of treatment, many aspects of patients' management, for example pharmaceuticals and equipment expenditure, have to be authorised by the doctor.

Kim is the team's link between the hospital and the clinic. Only a doctor has admitting rights, or can book someone for surgery. I can't do that. I can't go to a doctor and just say 'admit this patient'. (Liz, Occupational Therapist, informal interview) This means, in effect, that while Kim has the autonomy to operate independently of other team members, non-medical team members are reliant on Kim to implement treatment decisions they may make. At the same time, they do not have the level of responsibility or accountability that Kim has: Kim is directly liable to litigation, whereas other team members are covered by hospital insurance and are not vulnerable to individual litigation to the same extent. This differential level of autonomy and concomitant differential level of responsibility was not discussed in the team. Nevertheless, it was always present as sub-textual knowledge. It was one of the aspects that contributed to the conflation of 'team' knowledge with 'medical' knowledge:

Without Don and Nicky [and Liz], I wouldn't actually get information on compliance or any of this other stuff. I mean Nicky's actually touching the patients, and examining their shoulders and the rest of it, and actually takes a lot of time, 20 minutes plus, maybe an hour to do that, and they tell her things that they never tell me, and they are so relevant. I think if I didn't have that information, there's no way that we would know what was happening.

(Kim, Doctor, informal interview) Kim: Yeah, probably true, because he knows me, and we talk anyway.

Kim's response implies that any of the team members should be able to go to the medical superintendent about an issue that concerns them. However, outside of the team environment, proper channels have to be observed, and the medical specialist is the appropriate person to approach the medical superintendent. Kim did not just have access to the medical superintendent because they 'knew each other and talk anyway'. He knew him and talked to him because of his position as staff specialist, access which was not, in practice, available to other team members (Long et. al. in press, b).

The complexity of these issues is well illustrated by the fact that Kim feels that for the team to work well, they have to develop a culture different to that of the hospital. I think we need to operate differently from the ward ... we have to be more flexible than the ward situation.

[…] It's my personal view that it's acceptable that if someone is more qualified to take the lead in a discussion, whether it's social issues or whatever, then they should be directing and aligning the team towards what they feel is best practice. I would hope that this evolves. I think it has to.

(Kim, Doctor, research feedback session)

Discussion

Evetts warns against the tendency to homogenise all members of a profession, commenting on 'diversity, restratification and growing hierarchy ' within professions (2003:403). Medical dominance is frequently discussed in homogenised terms that do not do justice to the complex and dynamic relationships between medicine and other health professions (Evetts 2003;Griffiths 1998;Weiss and Fitzpatrick 1997;Williams 2002), and between medicine and society and/or the state (De Voe and Short 2003; Ham and Alberti 2002;McKinlay and Marceau 2002;Wailoo 2004). One effect of this homogenisation is to reify 'the professions' in ways which ascribe agency to particular health professions as undifferentiated groups of people with uniform behaviours, practices and values. Intrinsic to many of these discussions is an assumption that dominance is something that doctors are wanting to have and to hold, and that it is something that members of other health professions, the state, or individual patients are trying to wrest from them. While this may often be the case, and may even apply to members of the clinical team discussed above in other contexts, as we have seen, in the context of the spinal pressure area clinic, clinicians were struggling with medical dominance in quite a different way to what much of the literature may suggest. Reporting on a study of negotiated medical dominance in an ICU team, Carmel (2006) makes the point that medical voice does more than dominate or mute non-medical voice. He argues that in forging an ICU-identified alliance, nursing staff could utilise medical authority to enhance nursing requests of the wider hospital, thereby enhancing both professions' positions within the unit, and the interests of the ICU unit as a whole (Carmel 2006). Similarly, the team of spinal clinicians utilised their access to medical authority to further the interests of the clinic via Kim's advocacy of their needs. The non-medical clinicians' refusal to label Kim as facilitator rather than leader bears out this recognition of Kim's role in representing the needs of the team in a broader hospital environment.

This alliance between the medical and nonmedical clinicians, strikingly similar to that Carmel (2006), is, we suspect, increasingly common given the growth of multidisciplinary team work. It would take comparison of a number of further empirical studies to usefully disentangle personality from professionality in these teams, and we suggest this is an area of research that warrants further attention.

Secondly, autonomy has been central to many of the discussions around medical dominance, whether between state control and professional autonomy (e.g., Flynn 2002:162), medical autonomy and the control that gives medicine over other health professions (e.g., Britten 2001), or the role of autonomy in doctor-patient interactions (e.g., Roter 2000;Zadoroznyj 2001). Parallel to our argument above, we suggest these discussions display an underlying assumption that doctors battle to maintain autonomy, while governance structures, other health professionals, or informed consumers attempt to undermine it. While, again, this may be true for many areas of health care, this does not provide a satisfying analytical framework for the spinal pressure area team's interactions.

Thirdly, we have shown the value of research which facilitates clinician reflexivity, which may have the potential to challenge deeply embedded norms of medical dominance. As a result of the team's engagement in the research project, for example, waiting hierarchies were challenged, and the devaluation of the physiotherapist's time was addressed by changing the structure of her work in the clinic to being on-call. By reflecting on their interactions in the research feedback sessions, clinicians were able to recognise and begin to act upon the medically dominant voice, and to disentangle, at least to some extent, takenas-given behaviour related to leadership and followership (Degeling et al. 2003).

Finally, a number of commentators discuss consumerism as undermining medical dominance, and in the context of the phenomena of the selfsurveilling, expert patient, comment that patients have internalised, appropriated and/or contested medically dominant frameworks to re-orient their own embodiment (Fox et al. 2005;Kippax and Race 2003;Zadoroznyj 2001). As our research shows, these discussions could also be usefully applied to the internalised, self-surveilling nature of professional relationships between medical and non-medical hospital professionals. Claims about the professional positions in the social hierarchy may benefit by being refracted through the complex, in situ enactments of specific bodies and practices. It is here that our research has sought to make its contribution.