An action-research study exploring midwives’ support needs and the affect of group clinical supervision

ARTICLE IN PRESS Midwifery (2005) 21, 161–176

www.elsevier.com/locate/midw

An action-research study exploring midwives’ support needs and the affect of group clinical supervision Ruth Deery, BSc (Hons), PhD, RM, ADM (Senior Lecturer in Midwifery) Division of Midwifery, University of Huddersfield, Queensgate, Huddersfield HD1 3DH, UK Received 22 June 2004; received in revised form 4 October 2004; accepted 8 October 2004

KEYWORDS Action research; Clinical supervision; Support; Organisational change; Participation

Summary Objective: to explore community midwives’ views and experiences of their support needs in clinical practice, and then to identify how they would wish to receive such support. Further objectives were to redress the imbalance identified by planning and facilitating a model of clinical supervision devised by the participating midwives. Design: a qualitative study using an action-research approach based on collaboration and participation. Action research has the potential to facilitate understanding of, and is able to adapt to, changing situations within clinical practice. Data were collected in three phases using in-depth interviews and focus groups. Setting: a large maternity unit in the north of England, UK. Participants: eight National Health Service (NHS) community midwives working in the same team. Findings: recent and ongoing organisational change and increased demands placed on the midwives by their managers were found to be detrimental to the process of clinical supervision and working relationships with their peers and clients. These pressures also inhibited the process of change. The midwives’ behaviour and coping strategies revealed an apparent lack of understanding on their part, and that of their midwifery managers, of the regulation of emotion and the amount of energy this generated. Pseudo-cohesion and resistance to change were key defence mechanisms used by the participating midwives. Key conclusions: a large amount of published literature supported the existence of stress and burnout in midwifery, but no research addressed ways of alleviating this situation. Effective facilitation of midwifery support is needed, which can be met through support mechanisms such as clinical supervision. During the process of

E-mail address: [email protected]. 0266-6138/$ - see front matter & 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.midw.2004.10.006

ARTICLE IN PRESS 162

R. Deery clinical supervision, strong messages emerged about the necessity to ensure that midwives are prepared educationally for the difficult situations that are brought about through collaborative working. There are also messages about the cultural legacy of NHS midwifery and how this can inhibit autonomous behaviour by midwives. Implications for practice: developing and increasing self-awareness is still not viewed as being intrinsic to the work of the midwife, and midwives are being asked to undertake a level of work for which they have not been adequately prepared. The bureaucratic pressures of working in a large maternity unit exaggerate this further. In this situation, the system is seen as more important than the midwives. & 2005 Elsevier Ltd. All rights reserved.

Introduction The past decade has seen the maternity services in the UK trying to respond to strategic planning and policy-making directives (DoH, 1993a, 1999) without giving due consideration to a well-established cultural legacy within midwifery (Kirkham, 1999; Hughes et al., 2002). This contradiction has resulted in a midwifery workforce that is struggling with the pace and nature of large-scale change, and recruitment and retention difficulties are egregious (Ball et al., 2002). Midwives have reported feeling stressed (Sandall, 1998) and unsupported (Kirkham and Stapleton, 2000) as they have juggled competing organisational and client demands within a culture that is resistant to change. The increased support that midwives are expected to offer their clients in order to provide a ‘with woman’ approach sharply contrasts with their own impoverished support. In this paper, some of the findings from a study that sought to explore midwives’ support needs, and the ways in which they would wish to receive such support, are reported. An action-research approach fostered the initiation and development of the midwives’ own framework for clinical supervision, highlighting a need for the effective facilitation of midwifery support in a bureaucratic context, such as the National Health Service (NHS).

Culture and change in midwifery The culture of midwifery has remained mostly unchanged and unchallenged, despite the NHS as an organisation and provider of health care being subject to continuous, unprecedented change (Kirkham, 1999). Only in recent years have researchers dared to challenge some of the deeply entrenched cultural codes and routinised practices (Davies, 1995) within the NHS, and suggest how changes in practice and working relationships can

create real benefits for NHS hospitals and midwives (Stapleton et al., 1998; Kirkham, 1999; Kirkham and Stapleton, 2000; Hughes et al., 2002; Brooks and Brown, 2002). At national level, the Government has set some difficult challenges in the NHS Plan (DoH, 2000) in order to achieve change. Some of these challenges have provided midwives with the opportunity to extend their role to ‘Modern Matron’ and ‘Consultant Midwife’, whereas others have taken on more responsibility for various aspects of the junior doctor’s role with the reduction in their working hours (DoH, 2000). Midwives have seen clients given new powers and more influence over the way the NHS works. At the same time, they have been asked to increase and improve care given to clients in deprived areas, to introduce antenatal screening programmes for women and provide smoking-cessation services (DoH, 2000). The priority that is now given to delivering highquality services means that midwives are often working in complex and difficult circumstances. The climate of continual change often brought about by various policy directives has become a potential health hazard for midwives, leading to stress-related disease (Sandall, 1997, 1998, 1999; Mackin and Sinclair, 1998). This is further complicated by the fact that midwifery as an occupation involves direct contact with women and their families, which in turn makes midwives susceptible to a particular type of occupational stress known as ‘burnout syndrome’ (Sandall, 1995, 1997). New working patterns have meant that a large number of midwives are now working longer hours than previously, which can result in emotional exhaustion (RCM, 1997; Sandall, 1999). There is also evidence to suggest widespread dissatisfaction with the maternity services among those midwives who leave the profession (Ball et al., 2002). Strategies that provide effective support were cited by the ex-midwives in this study as being important to effective re-recruitment and retention of midwives. Reported here is the first

ARTICLE IN PRESS Support needs and the affect of group clinical supervision action-research study of its kind in midwifery to move beyond acknowledging the existence of stress and burnout by devising and mobilising a support mechanism for midwives in practice. The organisation of work in the NHS is such that midwives mainly work in a medically and managerially dominated culture, in which clear hierarchical divisions still persist between all levels of staff (Kirkham, 1999). Routinised practice promotes and preserves (Brooks and Brown, 2002) an existing task-based approach to work within the NHS, and has been likened to an industrial model of childbirth (Kirkham, 2003). From a sociological perspective, Lipsky (1980) provides important insights into the occupational culture described above that help to contextualise the way midwives cope when working within an industrial model. ‘Street-level bureaucracies’ (Lipsky, 1980) are created where workers are unable to meet their full potential because of increasing demands within the organisational setting: Ideally, and by training, street-level bureaucrats respond to the individual needs or characteristics of the people they serve or confront. In practice, they must deal with clients on a mass basis, since work requirements prohibit individualized servicey.At best, street-level bureaucrats invent benign modes of mass processing that more or less permit them to deal with the public fairly, appropriately, and successfully. At worst, they give in to favouritism, stereotyping, and routinizing – all of which serve private or agency purposes (p. xi). Working within an industrial model of childbirth (Kirkham, 2003), therefore, means the organisation is seen as more important, power and control lie with managers and relationships become disconnected and undervalued. Yet in the rhetoric of policy documents (DoH, 1993a, 1999, 2000), the traditional relationship of dominant expert midwife and passive client continues to be discouraged in favour of a relationship in which midwives can engage in equal, empowering relationships with clients. In practice, this approach to midwifery demands a relational style that offers support and practical expertise for the client, encouraging the woman to make her own decisions in order to be in control of the childbearing experience (Leap, 2000; Stapleton et al., 2002). Working in partnership with women (Fleming, 2000; Pairman, 2000) suggests a less paternalistic approach to the midwife–mother relationship. Instead, there is a sharing of knowledge and power, and an understanding of the dynamics involved within this relationship. Current government policy

163

(DoH, 1993a) encourages a ‘power-with’ approach to relationships. This approach depends on personal responsibility, on creativity and a willingness to collaborate (Starhawk, 1987, p. 11). However, in reality, the patriarchal ideology of power associated with the medicalisation of childbirth and organisational demands far outweigh the enablement of partnership with clients (Levy, 1999). Therefore, even though midwives might be willing to share their power with clients, they could find themselves buried in bureaucratic and administrative barriers that prevent this power shift (Lipsky, 1980). The emphasis placed on the way midwifery work is organised today, and increasing pressure from managers to conform to organisational demands (Ball et al., 2002), seem to have neglected the development and potential benefits of relationships. At the same time, the support that is required for midwives working within these increasingly complex situations has been ignored. In this study, clinical supervision was identified by the participating midwives as an approach that might help them cope with the uncertainty and challenge of change within their practice.

Clinical supervision as a potential source of support Clinical supervision has been described as a formal process of professional support and learning (DoH, 1993b) that has a ‘process-orientated, interrelational’ approach (Lindahl and Norberg, 2002, p. 809). Clinical supervision focuses on reflection and clinical practice, and how practitioners engage with clients for the benefit of clients. The emphasis is also on the personal and professional development of the supervisee; the value of supporting the supervisee is emphasised, in addition to learning from the experience of clinical supervision. The process seems to be supervisee-led, and does not suggest compulsory attendance and can be terminated at any time by the supervisee (Bond and Holland, 1998; Butterworth, 1998). Nursing has generated an interest in clinical supervision as a framework that could contribute to quality service provision (Bishop, 1998) in the same way that statutory supervision of midwives claims to be a quality strategy (Winship, 1996). When clinical supervision and statutory supervision of midwives become incorporated into comprehensive programmes of quality improvement activities, clinical audit, evidence-based practice and risk management (DoH, 1998) become important. Initiatives such as critical incident reporting and complaints procedures are then put into place by

ARTICLE IN PRESS 164 managers, so that they can monitor and act on what is reported as poor performance. In some hospitals, midwifery managers are also supervisors of midwives, probably leading to statutory supervision of midwives being viewed by some midwives as an imposition on their practice and a policing mechanism (ARM, 1995; Deery and Corby, 1996; Stapleton et al., 1998). Overseeing midwifery work in this way can create a culture of fear and commitment to the organisation; the needs of workers become secondary to the needs of the organisation. Similarly, clinical supervision is also resisted in some areas of nursing because it is perceived as another management monitoring tool (Rolfe et al., 2001). However, clinical supervision can act as a supportive framework for health practitioners, which does not have as its focus monitoring or investigation of clinical practice (Faugier, 1998; Wilkins, 1998). As a process, it has been described as supportive and enabling (Faugier, 1998), concerned with professional development (Hawkins and Shohet, 1989; Cutcliffe and Epling, 1997; Faugier, 1998), client-centred (Morris, 1995) and an investment in staff (Hallberg and Norberg, 1993; Dudley and Butterworth, 1994; Butterworth et al., 1997). Deery and Corby (1996) put forward a case for clinical supervision in midwifery, but there seems to be no reported research within midwifery on clinical supervision.

The focus of clinical supervision The nursing literature address a number of different approaches to clinical supervision. However, these varying approaches also highlight a lack of consensus around definitions and models, and present a confusing picture when trying to decide which model to use in clinical practice. Yegdich and Cushing (1998) argue that alternative approaches to clinical supervision have arisen according to the differing needs of nurses and health visitors. This would seem appropriate where a diversity of clinical needs is found in nursing (Butterworth et al., 1997). However, this same argument cannot be applied to midwifery, as agreement has not yet been reached on a shared philosophy of care according to the needs of midwives and clients. Debate would be useful within the midwifery profession to articulate and demonstrate how midwives have been influenced by varying approaches to clinical supervision. Clinical supervision is not a singular concept, and the approach initially taken by the supervisor will depend on the school of counselling or psychotherapy that the supervisor subscribes to (Farrington,

R. Deery 1995), and the way in which this is applied to clinical supervision. Within midwifery, the theoretical concepts underlying these approaches would have to be understood in more depth by midwives before clinical supervision could be developed. One of the aims of this research was to facilitate such exploration and help midwives to develop their own model of clinical supervision. Clinical supervision can facilitate supervisees to participate in a process that is likely to challenge, stimulate and encourage exploration of contemporary practice (Wilkins, 1998). As a process, it provides the opportunity for midwives to step back and become more self-aware in their interactions with clients and other midwives. As midwifery is essentially about human relationships, the development of interpersonal skills and ways of managing emotions at work are essential to help midwives cope with the, often, stressful nature of their work.

Methods An action-research approach, which incorporated in-depth interviews, focus groups and workshop sessions, was used to obtain data from the participating midwives. Meyer and Batehup (1997) claim that action research is not easily defined, because it is more of an approach to research rather than a specific method, meaning that it is fluid and may involve many methods rather than a pre-determined methodological menu. Waterman et al. (2001) also highlight how action research is now used in a variety of health-care settings, although its scope and purpose are often unclear, and it is used under a variety of names. However, at the heart of action research lies a democratic, reflective and collaborative approach between researcher and participants, who aim to change or better understand a social situation (Waterman et al., 2001; Winter and Munn-Giddings, 2001). Action research is appropriate in rapidly changing maternity settings, because it accommodates the unpredictability of midwifery work and complex clinical-practice situations through collaboration. Its cyclical nature, with fact-finding, action and evaluation within each cycle, is highly appropriate for researching clinical work in times of change (Deery and Kirkham, 2000; Waterman et al., 2001). Uncertainties and tensions within midwifery can be reflected upon through a collaborative, democratic and empowering approach to change, and through a research approach that reflects the complex, messy nature of clinical practice. This process also mirrors the cyclical framework of care planning within midwifery.

ARTICLE IN PRESS Support needs and the affect of group clinical supervision The nature and degree of democratic participation in action research are also important factors (Hart and Bond, 1995). As action research is context specific, that is, focusing on a local or discreet situation, location or group (Morrison and Lilford, 2001), the amount and nature of participation involved is diverse. Participation involves interaction between researcher and participants, with an emphasis on the importance of valuing knowledge that is created from clinical practice. Developing this knowledge further through collaboration with others and critical personal reflection is emphasised, which is seen as crucial to the process of action research (Waterman et al., 2001). Important parallels can be made here with clinical supervision (i.e. complexity of interpersonal relationships, reflection, change and the nature of support). The metaphor of ‘voice’ has also been highlighted as an important concept in action research, facilitating the inarticulation of viewpoints from participants that previously had not been heard or had been ignored (Maguire, 2001). The importance of understanding and evoking silence was crucial in this study, and the participating midwives were facilitated to reflect on deeply entrenched and internalised ways of working, and then voice their concerns.

Phase one of the study The study was conducted in three phases between March 1997 and November 2000 (Fig. 1). In Phase 1, preparatory work was carried out with the participating midwives, and meetings were held with various midwifery managers, informing them of the nature of the study and gaining informed consent. This phase also enabled the participants to ask questions about the proposed study. During this time, the participants also agreed to their data being used to disseminate the findings of the study. Meetings usually took place in the Health Centre from late July to November 1997, often following an antenatal clinic. Individual informed consent was not sought from each midwife, as the amorphous quality of action research does not allow for the ‘prediction’ of what the participants are ‘consenting to’ (Williamson and Prosser, 2002); however, at each interview, focus group and workshop, the voluntary nature of the research was reemphasised. As such, informed consent was not gained from the participants on the basis of their ability to process information and come to a rational decision unaffected by the influences of others (Holloway and Jefferson, 2002), but as an ongoing process where a safe environment was created in order that the participants could openly

165

express their views within a non-judgmental environment. Eight NHS community midwives working in the same team chose to take part in the study. The researcher initially approached community midwives believing that the practicalities of participation might be more flexible in this group of midwives. Delivery-suite midwives, for example, could not attend interviews or focus groups at agreed times as they may have been supporting birthing women. However, the nature of community midwives’ work, and their extending role, has changed dramatically in the past decade, and the researcher discovered that the practicalities of participation were just as limited.

Listening to midwives’ voices Ribbens (1989) has stated that researchers should be concerned with what they are trying to achieve within their studies and the types of relationships that they want to develop with their participants when choosing their methods of data collection. This stance is particularly pertinent to an actionresearch approach, in which collaboration and participation are seen as essential components (Winter and Munn-Giddings, 2001). In-depth interviews, therefore, seemed to provide the opportunity for the participants to tell their stories (Riessman, 1993), be flexible when gathering information (Arksey and Knight, 1999), interactive (Cotterill, 1992) and reciprocal (Oakley, 1981). During the interview process, the researcher also tried to move the midwives into more reflective thinking by asking questions (Box 1) that focused more directly on aspects of their clinical practice. Close parallels can also be made here with clinical supervision and reflective practice (Bond and Holland, 1998; Wilkins, 1998; Rolfe et al., 2001). Pilot interviews were also undertaken with three community midwives who were not going to be involved in the study. This proved to be a useful exercise, as the researcher was able to check out proposed questions and practice interviewing skills. These three midwives worked in different teams but within the same maternity service in which the research was undertaken. The midwives provided honest, insightful accounts that supported the words of the participating midwives, thus providing evidence that community midwifery, as depicted by the midwives participating in the study, was not peculiar to their team. The researcher sought permission from these three midwives to include their words with those of the participating midwives during data analysis and dissemination of the

ARTICLE IN PRESS 166

R. Deery

Phase one: gaining access and recruitment to the study

Meeting with Head of Midwifery manager

March 1997: ethical approval granted

September 1997 to March 1998: preliminary individual interviews held

Meeting with supervisors of midwives

April 1997: meeting with community midwifery manager

March to September 1997: several meetings with midwives in Health Centre

Pilot interviews held

Phase two: preparing for, and negotiating, clinical supervision

Meeting with midwifery

November 1998: second focus group

Educational input about clinical supervision

April 1998: first focus group held

Meeting with new Head of Midwifery

January 1999: meeting with prospective clinical supervisor

Funding awarded from West Yorkshire Workforce Development

April 1999: further meeting between researcher and clinical supervisor

Phase three: undertaking clinical supervision and evaluation Contract setting meeting with midwives and clinical supervisor

May to October 1999: midwives participated in clinical supervision

December to April 2000: final individual interviews

February 2000: individual interview

Data analysis

Fig. 1 Flow chart sign-posting phases and progress of the study.

research. The midwives consented to their data being shared. A decision was made to interview the participating midwives twice, initially to capture life as a community midwife and identify a need for change, and also at the end of the study in order to evaluate their experience. The midwives decided during the

initial meetings held in the Health Centre that individual interviews would be a good starting point followed by a sharing of information and decisionmaking in follow-up focus groups. However, as this was an action-research study, the researcher had to bear in mind that the participants shaped the progress of the study, and that a need may arise

ARTICLE IN PRESS Support needs and the affect of group clinical supervision

167

Box 1 Interview schedule: preliminary interviews.

Box 2

1.

2.

2. 3. 4. 5. 6.

Tell me about life as a community midwife What makes you feel positive about the service you offer to women? What makes it difficult working for the NHS? How would you like to see support for midwives being offered? What changes would you like to see? What sort of education have you had to prepare you for change?

during the course of the research to alter the initial ‘plan’. As Waterman et al. (2001) state: (y) the identification and clarification of issues followed by a process of implementation, adaptation and evaluation is ongoingyThis means that the design of [action] research projects is evolutionary (y). All the interviews were audio-taped, transcribed in full and then anonymised. The typed transcripts were returned to the midwives within 1 month of the interview, and before the researcher’s next meeting with the midwives, to check for accuracy and to invite further comments. All the interviews were conducted at a time convenient to the midwives. The opportunity to engage in data analysis was initially broached with the participants in the early planning stages of the study, but the midwives seemed reluctant to engage with this process, viewing it as more commitment and extra work.

Phase two of the study Phase two of the study comprised two focus groups and two workshops. The midwives had decided in the early planning stages of the study that the preferred nature of their support mechanism could not be decided in isolation, and they agreed to group together after the individual interviews in order to share ideas. Therefore, the first focus group (Box 2) was held to reflect on the content of the individual interviews, and then to plan a means of gaining support through mutual collaboration. This focus group also brought the participating midwives together so that they could identify important issues that had arisen for them during their individual interviews. The second focus group

1. 3. 4. 5. 6. 7. 8. 9.

Interview schedule: focus group 1.

Reflections on their individual interview What’s expected of you in the service? What’s good about working for the NHS? What’s hard about working for the NHS? What does support mean for them? How do they take care of their own needs? Of all the needs that were discussed, which one is most important for you? Prioritise those issues presented. How far are they prepared to go to get their needs met?

Box 3

Interview schedule: focus group 2.

1.

Where are we now? They’ve had time to reflect on research to date and workshop sessions. What now? How are they feeling? What are their priorities and needs? 2. Models of clinical supervision, discuss differing approaches. 3. Consider the advantages and disadvantages of one-to-one clinical supervision versus group supervision. Should the potential clinical supervisors be part of this planning? Outside facilitator or peer group facilitator?

(Box 3) was held to construct the midwives’ chosen support framework for use in clinical practice. To emphasise the collaborative nature of the study, it was important for each midwife to be invited to contribute their views on the progress of the study and how they imagined support for themselves to develop as a result of the research. As one of the aims of the study was to explore midwives’ working relationships, it was important that the researcher was able to observe the team together, albeit out of the work context. This is reminiscent of ethnography, in which the researcher is concerned with: (y) capturing, interpreting, and explaining the way in which people in a group, organization, community, or society live, experience, and

ARTICLE IN PRESS 168 make sense out of their lives, their world, and their society or group (Bentz and Shapiro, 1998, p. 117). Kitzinger (1994) believes that focus groups should be used to encourage people to engage with one another, to express themselves and to articulate previously unheard views. Focus groups are, therefore, different to interviews in that participants engage with each other as well as the researcher (Agar and MacDonald, 1995). Individual narratives, as well as the interactions taking place between the participants and the researcher, are an important source of data, and play a crucial part in the analysis of focus groups (Webb and Kevern, 2001). However, the dynamic of focus groups does not always meet the needs of all group members, and might, in some circumstances, serve to silence or mute some of the participants. Baker and Hinton (1999) have identified this dynamic as the participants and the researcher bringing their own expectations and personal agenda to the focusgroup setting. In this study, the dynamics within the day-to-day functioning of the team setting became exaggerated, with some of the midwives appearing to become muted while others tended to dominate the focus group. Thus, the researcher’s eagerness to facilitate a non-hierarchical participatory approach was not fully realised during the focus groups. Eventually, this turned out to be a reflection of how the work team functioned and communicated to the researcher the group dynamics within the team. After the first focus group, and at the midwives’ request, the researcher met with a colleague from mental-health nursing who had agreed to facilitate some workshop sessions on clinical supervision with the midwives. Without disclosing confidential information, the researcher explained that the midwives had identified that they needed some sort of structured support in practice, and that their initial deliberations resembled clinical supervision. The midwives had also identified a need to work through some sensitive interpersonal issues, particularly in relation to managing conflict and confrontation. This colleague was therefore asked to address the nature of clinical supervision, the midwives’ role in this process, the role of the clinical supervisor and the benefits of clinical supervision. During the same month, a second workshop session was held by the same colleague for the participating midwives. This session reviewed the first workshop, invited questions and then addressed the essential requirements and qualities

R. Deery of a clinical supervision relationship. This was based mainly around the work of Heron (1991), and focused particularly on ‘confrontation’ and ‘support’ as interventions in relationships. The midwives were given a reading list and directed to key texts on clinical supervision. A revision session to take place at a later date was offered to the midwives in case they wanted to build on existing interpersonal skills to prepare for their role in the clinical supervision relationship. Interestingly, they declined this offer even though they had previously identified a need to develop this aspect of their role. A midwifery model of clinical supervision The second focus group formed a further cycle of the action-research process. Here, the midwives began to devise their model of clinical supervision after the workshop sessions. Significantly, at this meeting, they set time limits on the duration of the focus group, because they had clients to visit and an antenatal clinic in the afternoon. The researcher had begun to sense that the midwives’ participation and time commitment to the study was not being fully realised by midwifery managers, and there was a sense of exasperation from the midwives. Participation has been reported as being important at every stage of the action-research process because it can lead to greater understanding and contextualisation of the issues, within the study (Waterman et al., 2001). However, as was the case in this study, participation can also lead to shifting relationships that the participants, researcher and managers can find difficult. At this second focus group, several of the midwives interacted angrily with the researcher, and a distinct reluctance to participate any further in the study was perceived by the researcher. On reflection, the researcher realised that this was displaced anger (Fineman, 2003), and probably a result of feeling frustrated and unsupported by midwifery managers. Waterman et al. (2001) found that those people who did not support or respond to action-research projects were often members of staff in positions of authority. In this study, such a situation led the midwives to view themselves as being in a ‘them’ and ‘us’ situation. After heated discussion, the researcher agreed to meet with the Head of Midwifery and other managers in order to negotiate extra cover and support for the midwives. External funding was also sought and granted, although the money injected into the midwifery budget was used to pay the researcher to undertake bank shifts to enable the participating midwives to finish work early so that they could spend time with the researcher.

ARTICLE IN PRESS Support needs and the affect of group clinical supervision During the latter part of the second focus group, the midwives decided on their proposed model of clinical supervision. They had already identified a potential clinical supervisor during workshop sessions, and now stated that they wished to undertake clinical supervision fortnightly in a group situation offsite. The researcher agreed to arrange an initial meeting with their potential clinical supervisor so that a date could be set to discuss their proposed model and contract setting for clinical supervision. Clinical supervision was undertaken over a 6-month period. The researcher did not take part, as participation on this level might have jeopardised the relationship between the researcher and the participants and affected the dynamic of the clinical supervision group.

Phase three of the study Phase three of the study comprised individual interviews (Box 4) with the five remaining midwives and the clinical supervisor. These interviews aimed to evaluate the process of clinical supervision, and took place at a time when the maternity unit where the midwives worked was undergoing a recruitment and retention crisis. Time as a scarce commodity had become pronounced during clinical supervision, with time costs being experienced differently by the midwives. The clinical supervisor had reinforced the conditions necessary for an effective clinical supervision relationship, and a ‘contract setting’ meeting had taken place, in which the midwives discussed frequency of sessions, continuity of group membership (i.e. attendance) and confidentiality. However, attendance was poor at the sessions, and the midwives stated that they had

Box 4 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

169

no time in which to undertake clinical supervision. Neither were the midwives able to keep to their initial contract that was set with the clinical supervisor.

Ethics Ethical approval had been sought in the early planning stages of the study, and permission was granted by the Local Research Ethics Committee. The Chairperson of the committee required reassurance that the midwives were not being put under any undue pressure by taking part in the study. Attendance at the Local Research Ethics Committee was not required, but communication by letter and telephone about the midwives’ participation meant that the researcher was able to provide reassurance that the midwives were not being pressurised. Reassurance was given to the midwives that data would be treated confidentially and anonymously, and that no one would be able to trace information back to them, although midwifery managers knew that they were participating in the research. The nature of confidentiality and anonymity were constantly reinforced to the managers during meetings with the researcher. The midwives were also informed that they could withdraw from the study at any time and without prejudice, although discussion did take place about how their withdrawal might affect the dynamics of their work team, clinical supervision and the action-research process. No one withdrew, although, during the course of the study, one midwife left to work in a different Trust, one retired and another went on sick leave.

Interview schedule: final interviews.

Tell me about clinical supervision and how it worked? Did it meet your expectations? Tell me what has been good about it. Tell me about what you have learned from the experience. How do you think it fits to midwifery practice? What sort of forum has it provided for you? What have you been able to do in clinical supervision that you would not normally do? Is there any difference between the way things are now within the work team, to what they were before? What has clinical supervision offered you that is currently not available for you within the NHS? Can you tell me about one significant incident from your clinical supervision which you have found to be helpful or unhelpful or something that sticks in your mind? Do you think that we achieved what we set out to do?

ARTICLE IN PRESS 170 Data analysis During earlier phases of the study, data gathering and data analysis took place concurrently; however, once the study had ended, time was spent retrospectively analysing the same data but using voice-centred relational methodology (Mauthner and Doucet, 1998). Initially, the preliminary analysis of the data was undertaken by first reading the interview, then re-reading and generating themes that corresponded to narrative sections within each of the transcripts. As this was preliminary data analysis, phrases or words that captured meaning of highlighted sections of the data were used. These were then elaborated on, the themes highlighted and their relationships with each other examined more closely within the transcript. By this time, the researcher was able to provide descriptive labels for sections of narrative. At this stage, an urge to ‘clean up’ the data because of its sensitive nature had to be resisted. However, Riessman (1987) provided reassuring and useful insights here stating that ‘narratives are laced with social discourses and power relations’ (p. 65) that must be preserved, otherwise important information will be lost. The voice-centred relational method of data analysis (Mauthner and Doucet, 1998) helped to preserve this important information by focusing on understanding individuals in their social contexts, and the complexities of their relationships with other people. This method of data analysis also complemented a feminist approach brought to the research by enabling women’s voices to be heard, and facilitating connections between the individual life histories of the midwives and their work situation. Within this method of data analysis, there are four or more readings of each interview transcript. During the first reading of the transcript, the researcher listened to occurring main events, main characters, recurrent images, words, metaphors and contradictions in the narratives of the midwives that comprised the plot. The second reading of the transcript focused particularly on listening to how the midwives related their experiences, thoughts and feelings about themselves and clients. This second reading was cathartic for the researcher, as the full extent of the midwives’ stress and lack of support was realised. The third reading provided the opportunity to concentrate on and explore the midwives’ working relationships more closely, and how such relationships affected the midwives. The researcher realised that the organisation of the maternity services was insensitive to their needs as both women and midwives, and that, overall, the midwives experienced working relationships as

R. Deery unsupportive, constraining and intimidating, in which their voices were silenced or rejected. In the fourth reading, the researcher listened to how the midwives experienced their midwifery work in the wider social context of the NHS, concentrating on their narratives and the way in which they experienced midwifery and how well these reflected dominant ideologies at that time. The researcher’s own fifth reading returned to the metaphors used by the midwives in the study, and how these provided further insight into the culture of midwifery work and associated emotional consequences for midwives.

Findings Data analysis suggested that life as a community midwife working within the NHS arouses anxiety for midwives, and that the emotion work associated with the job is not acknowledged or understood by midwives, managers and the wider organisation. The maternity unit where this research was undertaken was being challenged by new and different government-policy initiatives (DoH, 1993a, 1999), which seemed to threaten the midwives’ existing work situation and compromise their well-being. In phase-one interviews, change was viewed as constant by the midwives, and all recently externally imposed change, including participation in this study, was seen as generating extra work and stress. Team midwifery, although no longer in operation, was seen as an imposition, and was criticised for exposing dysfunctional team working and increasing workloads and stress levels. The following data extracts illustrate the midwives’ views: (y) one of the problems has been the intensity of the work since team midwifery (Interview 5) (y) it’s thrown more work onto everybody really but particularly the traditional community midwives. (Interview 8) I just wasn’t happy in the team (y) underlying nastiness (y) dreadful group dynamicsypersonality differences (y) I know we’re all different but there was more going on under the surface than ever came to the top. (Interview 3) (y) it’s [team midwifery] made me more stressed (y) I mean I only did four months of team midwifery and I was just on my knees. (Interview 4)

ARTICLE IN PRESS Support needs and the affect of group clinical supervision Radical changes in working practices had left some of the midwives wanting to practise like the ‘olden days’. The recognition and admittance that professional power was detrimental to working relationships with clients and peers had left some of the midwives feeling vulnerable and exposed, with a fear of complex clinical situations. ‘Pseudocohesion’, where the midwives projected an image of a supportive work team, was used as a defence mechanism in order to mask unsupportive behaviour within the work team. There seemed to be a fear in the work team of hurting a colleague through clumsy communication and, in their efforts to save each other from emotional discomfort, the midwives reported dealing with work-related issues superficially, sometimes manipulatively, often destructively, and in a manner that often sabotaged their good intentions. The clinical supervisor identified this as behaviour akin to ladylike saboteurs; the midwives preferred to avoid the consequences of what they perceived as bumbling and inept communication. The following data illustrate the midwives’ views: (y) we bury a lot because we don’t want to fall out as a team (y) we all recognise the value of having this gelled team and we all swallow bits and pieces that we are maybe not happy with and then we don’t act on things that we think should be acted on because we don’t want to destroy (y). (Interview 5) I find it very hard to deal with someone who is not acknowledging that there is a problem (y). (Interview 8) (y) but then I was in a position where I thought well has she offered because she thinks I don’t want to do it (y) you know I twisted it round (y) I hope she’s not thinking that lazy bitch she never does anything. (Interview 1) The data from phase-one interviews also provided insight into midwives’ emotional well-being, and the ways in which they had to regulate their emotions as a way of ‘getting the work done’. Data analysis revealed how midwives were expected to relate to, and develop partnerships with, clients when they themselves were inadequately prepared for this aspect of their role. The midwives articulated feelings of being overwhelmed by the organisational demands of the maternity service and their increasing workloads. Flexible working practices that were encouraged by managers meant that midwifery work still had to be done and slotted into the working day without additional help. Their words also suggested that they had

171

been overwhelmed by their relationships with each other. (y) and it was stressful (y) very stressful because we were (y) we had two people’s work to do in a day basically and with the best will in the world you can only be in one place at once (y) and then the mobile phones (y) meetings to attend (y) study leave (y) we were getting (y) arguing amongst ourselves (y) niggly (y) bickering (y) one person felt another wasn’t doing enough (y) the situation put us under such pressure. (Interview 8) (y) people think that you can cope and think that you are all right (y) this is something you often perpetuate because you wouldn’t have them know anything else (y) I think there are times inside when I’ve thought ‘god if my colleagues knew how I was feeling right now’ (y) you’ve got to keep going (y). (Interview 6) I had to psych myself up to go into someone particularly if their circumstances were sensitive or there was a language barrier (y). (Interview 1) I went home and I sat in the chair for about an hour and a half (y) just like zombified (y) thinking about what had gone on (y) and if I had done everything. (Interview 7) Data from both focus groups suggested that the midwives’ support needs were not being met. An inherent contradiction became apparent when midwives suggested that they felt unsupported, yet they were constantly reiterating that they were a supportive team and appeared keen to project a ‘united front’ (Goffman, 1990). Pseudo-cohesion as a defence mechanism became apparent again, as well as an avoidance of issues, which meant some of the team could not work seriously with their anxieties and hopes for the future: (y) we’re lucky because not every group is like us (y) supportive I mean (y).(Interview 6) (y) there are those things that we need to sort out but a lot of the time we don’t confront them because we don’t want to break down that team spirit thing because it’s really important to us all to keep it going. (Interview 5) A lack of consistent management before commencing the study had left the midwives feeling unsupported overall. The collaboration and participation necessary within action research was unfamiliar to them, and something that they had not previously experienced. Their sense of isolation became pronounced in a culture in which midwives

ARTICLE IN PRESS 172 have previously reported experiencing fear and isolation (Stapleton et al., 1998; Kirkham, 1999; Kirkham and Stapleton, 2000): (y) it’s when you said to us that it’s joint and collaborative (y) I’ll never feel that it’s just as much my study as it is yours (y) how can I (y) you couldn’t ever expect us to believe that (y). (Interview 8) (y) you [another midwife] were right (y) you were right (y) you said we wouldn’t get the support. (Interview 5) (y) they’re [the managers] not there to back you up with help when it’s needed. (Interview 2) Data from the final interviews suggested that the challenges the researcher presented to the midwives, in the form of participating in an actionresearch study, were too much for them to contend with. Although the data suggested that they found clinical supervision to be beneficial, and necessary, for midwifery work, they stated that they had no time in which to undertake this process: (y) it was like well what’s the point of it [clinical supervision], we just haven’t got time for it and yet I suppose if there was ever a time we needed it, it was at that time. (Interview 5) (y) it’s [clinical supervision] something new and it’s something that’s time consuming and you need time to do it. (Interview 7) I’ve learnt that in the hurley burley of everyday work we do need something to bring a group of people together (y) we’re all going in our different directions all the time and we only have five minutes in a morning to see each other and it was a good opportunity to get together and relax and speak our minds and I think to have one person there who is impartial, who can make you see things in a different light (y) I think it’s good. (Interview 6) The midwives’ apparent resistance to this change, and what seemed to be subsequent subversion of clinical supervision, meant that they almost certainly missed the opportunity to improve and change their working lives. Clinical supervision highlighted their pseudo-cohesion even further, but at the same time their vulnerability in an everchanging climate. The demands of the organisation, therefore, took priority over their own needs. The clinical supervisor found that the midwives were not able to keep to the ground rules set in their contract, and attendance levels were sometimes low. The data suggested that the midwives found sensitive interpersonal issues, and the some-

R. Deery times dysfunctional dynamics of the work team, difficult to address. This was not surprising when any further investment into working relationships through clinical supervision just seemed like extra work for the midwives: We were doing something that seemed as if it was going to encroach on our time (y) it was hard at times to see how we could possibly benefit from this [clinical supervision], other than more work, more commitment and more hassle. (Interview 8) (y) feelings were aired and views were offered that were honest, I mean I can’t say how honest any of the others were, I can only say what I perceived the level of honesty to be and I think we all held back slightly because nobody wanted to get hurt and upset. (Interview 3) (y) it made me realise that we weren’t as honest as we could be with each other and that there were issues that needed sorting out and it made me aware of problems that we’d been shoving away (y) but it did benefit me. (Interview 5) I think it [clinical supervision] is a different way of thinking (y) it’s not the usual way of thinking (y) it would be a good thing for morale and stress levels. (Interview 7) The data suggested that the midwives found clinical supervision beneficial and necessary for the midwifery profession. However, they stated that they had no time in which to undertake this process, and rather than use the opportunity of clinical supervision in a safe environment to help themselves, they chose to continue working in the same manner.

Discussion There are limitations to this action-research study, which relate to the site of the research and the sample. The sample was limited to a team of community midwives, and has not considered the support needs of hospital midwives or midwifery managers. Further research is needed to address this deficit. The site of the research had particular problems in that it had a high ethnic-minority population with a high concentration of births in the inner city, high unemployment and associated socioeconomic deprivation. Many of the women presented with high-risk pregnancies. There was also a massive staffing crisis at the time the research was undertaken. The problems

ARTICLE IN PRESS Support needs and the affect of group clinical supervision experienced at this site remain common to NHS hospitals at this time. The midwives participating in this study articulated that their needs were not being met, and also that their needs were not congruent with the needs and interests of the NHS. However, on a more productive level (i.e. through clinical supervision), they were unable to address this deficit in any depth because of various constraints. As a result, the midwives spent most of their working lives in cultural conflict, meeting the needs of others, and not able to work with their own personal philosophies or practices of midwifery. Hunter (2002) found that ‘conflicting ideologies’ impinged on the work of the midwife, and were often experienced as emotionally difficult. The voices of community-based midwives have rarely informed NHS strategic planning (Hughes et al., 2002), and this can result in contradictions and conflicting values becoming apparent among the workforce (i.e. the values of managers clashing with the values of grassroots midwives). Although developing relationships is encouraged within the rhetoric of policy initiatives in the UK (DoH, 1993a, 1999, 2000), the potential benefits of this aspect of midwifery need to be recognised by those midwives and managers working in institutions. In the UK, there is often a culture that encourages midwives to work towards increasing their professional image and status based on their grasp of technological achievement rather than their relationships with clients. As a result, midwives learn to value what they can measure. The midwives in this study articulated that midwifery needs a different way of thinking. However, the midwives worked in a bureaucratic, hierarchical NHS system, in which a technocratic paradigm of healthcare existed, and this has been shown to be intolerant of different ways of thinking (DavisFloyd, 2001). Emotion work in midwifery can mean that midwives become overwhelmed by their relationships with each other. In this study, connectedness and relationship formation with colleagues and clients were made problematic by the organisation of midwifery work. The pressure to meet organisational demands within an industrial model of childbirth meant that the midwives felt threatened and unable to cope with their work. At times, this meant the midwives had to put on a ‘united front’ (Goffman, 1990), which required considerable energy on their part, and meant that they often experienced midwifery negatively. Projecting an image of supportive behaviour, rather than challenging the status quo, was easier in turbulent, stressful times, as this meant they did not have to

173

scrutinise their working relationships and practices more closely. As healthy supportive working relationships between midwives and clients, and midwives themselves, are crucial to positive childbearing experiences for clients and midwives (Flint, 1986; Kirkham, 2000), midwives need to find different ways of dealing with the realities of practice. Clinical supervision offers one such opportunity. Clinical supervision in this study offered the opportunity of support from someone from a different professional background who had no vested interest in the midwifery services. A deficit, however, was highlighted in the midwives’ knowledge of group dynamics. They were not prepared educationally for group functioning, and seemed to have been grouped together with no consideration to the effect this might have on them as individuals and on their working lives as midwives. The midwives in this study stated that they often felt ill-prepared for some of the difficult situations brought about through collaborative working in the NHS. Clinical supervision requires the development of interpersonal skills at a high level (Wilkins, 1998), which will be parallelled in midwifery work. It is crucial then that the skills necessary for effective group working and relationship building are addressed by midwifery, especially educationalists. Psychotherapeutic concepts and group-work theory are neglected areas in the midwife’s repertoire of skills, even though they can help midwives become more psychologically aware of their interactions with others (Raphael-Leff, 2000).

Implications for practice It is important that midwifery finds new and different ways of dealing with the realities of practice. The contradictions that exist within current practice need to be acknowledged rather than resisted, so that midwives feel nurtured and valued, and common goals are aspired to. Ways of achieving this need to be considered (e.g. the provision of structures and appropriate training whereby midwives can work with their own personal philosophies). Support mechanisms, such as clinical supervision, highlight a need for the effective facilitation of midwifery support. Clinical supervision also offers the opportunity to further understand how midwives manage their emotions and relationality in midwifery. Incorporating different perspectives into the practice of midwifery has often resulted in contradictions and conflicting values becoming apparent

ARTICLE IN PRESS 174 among the midwifery workforce. Therefore, there is a need to address the bureaucratic, hierarchical nature of the maternity services, and the prevalent medicalised, technocratic paradigm of health care that does not consider different ways of working. Working across professional boundaries, through a process such as clinical supervision, provides the opportunity to think differently. There is a need to ensure that midwives are prepared educationally for the difficult situations that are brought about through collaborative working. The midwives in this study identified a deficit in their midwifery training and education in relation to this aspect. Curriculum design and development need to take this deficit into account and build effective learning and teaching strategies into their programmes of education. Clinically, managers need to place the same emphasis on communication, interpersonal skills building and therapeutic proficiency that they place on mandatory updating, focusing on the medical management of care and interventions in midwifery. Action research is a tool that can help to bring about change in midwifery practice, although it demands the use of interpersonal skills on a level that acknowledges the complexity of real-life situations in the clinical setting. The degree of participation is crucial to the success of action research, and this concept needs to be fully understood by both participants and midwifery managers. Those participating in action research, as well as managers, need to give sufficient consideration to the potential problems that might arise as a result of conflicting goals and aspirations within the clinical setting. Just as the process of clinical supervision is as important as the content, so too, the process of action research is as important as the final outcomes.

Conclusion This study has provided many insights into midwives’ support needs and the ways in which midwives would wish to receive such support. Further sociological insights were also provided into the organisational culture of the NHS and the way midwives manage their emotions. Tensions and difficulties within action research when it is operationalised were highlighted. However, the potential role of action research to incorporate direct participation in strategic planning cannot be ignored, and needs to be embraced by midwifery as a means to integrate education, research and practice development. ‘Thinking differently’ might

R. Deery involve moving beyond acknowledging the existence of stress and burnout, and using clinical supervision as a means of effective midwifery support.

Acknowledgements The author wishes to thank all the midwives who participated in this study as well as Professor Mavis Kirkham and Dr. Sharon Wray for helpful comments during the writing of the paper. The author wishes to thank the anonymous referees for their very helpful comments that improved the paper. The research was also partly funded by a grant from the West Yorkshire Workforce Development Confederation.

References Agar, M., MacDonald, J., 1995. Focus goups and ethnography. Human Organization 54, 78–86. Arksey, H., Knight, P., 1999. Interviewing for social scientists. Sage Publications Ltd., London. Association of Radical Midwives (ARM), 1995. Super-vision: consensus conference proceedings. Books for Midwives Press, Cheshire. Ball, L., Curtis, P., Kirkham, M., 2002. Why do midwives leave? The Royal College of Midwives, London. Baker, R., Hinton, R., 1999. Do focus groups facilitate meaningful participation in social research? In: Barbour, R.S., Kitzinger, J. (Eds.), Developing focus group research, politics, theory and practice. Sage publications, London. Bentz, V.M., Shapiro, J.J., 1998. Mindful inquiry in social research. Sage Publications, London. Bishop, V., 1998. Clinical supervision in practice. Some questions, answers and guidelines. Macmillan Press Ltd, London. Bond, M., Holland, S., 1998. Skills of clinical supervision for nurses: A practical guide for supervisees, clinical supervisors and managers. Open University Press, Buckingham. Brooks, I., Brown, R.B., 2002. The role of ritualistic ceremonial in removing barriers between subcultures in the National Health Service. Journal of Advanced Nursing 38, 341–352. Butterworth, T., Carson, J., White, E., Jeacock, J., Clements, A., Bishop, V., 1997. It is good to talk: an evaluation study in England and Scotland. The University of Manchester, Manchester. Butterworth, T., 1998. Clinical supervision as an emerging idea in nursing. In: Butterworth, T., Faugier, J., Burnard, P. (Eds.), Clinical supervision and mentorship in nursing, 2nd edn. Nelson Thornes Ltd., Cheltenham. Cotterill, P., 1992. Interviewing women: issues of friendship, vulnerability, and power. Women’s Studies International Forum 15, 593–606. Cutcliffe, J.R., Epling, M., 1997. An exploration of John Heron’s confronting interventions within supervision: case studies from practice. Psychiatric Care 4, 174–180. Davies, C., 1995. Gender and the professional predicament in nursing. Open University Press, Buckingham.

ARTICLE IN PRESS Support needs and the affect of group clinical supervision Davis-Floyd, R., 2001. The technocratic, humanistic and holistic paradigms of childbirth. International Journal of Gynecology and Obstetrics 75 (supp 1), 5–23. Deery, R., Corby, D., 1996. A case for clinical supervision in midwifery. In: Kirkham, M. (Ed.), Supervision of midwives. Books for Midwives Press, Hale. Deery, R., Kirkham, M., 2000. Moving from hierarchy to collaboration: The birth of an action research project. The Practising Midwife 3, 25–28. Department of Health, 1993a. Changing childbirth, Part 1 (Report of the Expert Maternity Group). HMSO, London. Department of Health, 1993b. A vision for the future: the nursing, midwifery and health visiting contribution to health and health care. HMSO, London. Department of Health, 1998. A first class service: quality in the new NHS. HMSO, London. Department of Health, 1999. Making a difference: strengthening the nursing, midwifery and health visiting contribution to health and healthcare. HMSO, London. Department of Health, 2000. The NHS plan: a plan for investment, a plan for reform. HMSO, London. Dudley, M., Butterworth, T., 1994. The costs and some benefits of clinical supervision: an initial exploration. International Journal Psychiatric Research 1, 34–40. Farrington, A., 1995. ‘‘Models of clinical supervision’’. British Journal of Nursing 1, 876–878. Faugier, J., 1998. The supervisory relationship. In: Faugier, J., Burnard, P. (Eds.), Clinical supervision and mentorship in nursing, 2nd edn. Butterworth, Nelson Thornes Ltd., Cheltenham. Fineman, S., 2003. Understanding emotion at work. Sage Publications, London. Fleming, V., 2000. The midwifery partnership in New Zealand: past history or a new way forward? In: Kirkham, M. (Ed.), The midwife–mother relationship. Macmillan Press Ltd., London. Flint, C., 1986. Sensitive midwifery. Heinemann, London. Goffman, E., 1990 (first published in 1959). The presentation of self in everyday life, Penguin, Harmondsworth. Hallberg, I.R., Norberg, A., 1993. Strain among nurses and their emotional reactions during 1 year of systematic clinical supervision combined with the implementation of individualised care in dementia nursing. Journal of Advanced Nursing 18, 1860–1875. Hawkins, P., Shohet, R., 1989. Supervision in the helping professions. Open University Press, Buckingham. Heron, J., 1991. Helping the client: a creative practical guide. Sage Publications, London. Holloway, W., Jefferson, T., 2002. Doing qualitative research differently. Free association narrative and the interview method. Sage Publications, London. Hughes, D., Deery, R., Lovatt, A., 2002. A critical ethnographic approach to facilitating cultural shift in midwifery. Midwifery 18, 43–52. Hunter, B., 2002. Emotion work in midwifery: an ethnographic study of the emotional work undertaken by a sample of student and qualified midwives in Wales [unpublished PhD thesis]. University of Wales Swansea, Swansea. Kirkham, M., 1999. The culture of midwifery in the National Health Service in England. Journal of Advanced Nursing 30, 732–739. Kirkham, M., 2000. How can we relate? In: Kirkham, M. (Ed.), The midwife–mother relationship. Macmillan Press Ltd., London. Kirkham, M., 2003. Birth centres. A social model for maternity care, In: Book for Midwives. Elsevier Science Limited, London.

175

Kirkham, M., Stapleton, H., 2000. Midwives’ support needs as childbirth changes. Journal of Advanced Nursing 32, 465–472. Kitzinger, J., 1994. The methodology of focus groups: The importance of interaction between research participants. Sociology of Health and Illness 16, 103–121. Leap, N., 2000. The less we do, the more we give. In: Kirkham, M. (Ed.), The midwife–mother relationship. Macmillan Press Ltd., London. Levy, V., 1999. Protective steering: a grounded theory study of the processes involved when midwives facilitate informed choice in pregnancy. Journal of Advanced Nursing 29, 104–112. Lindahl, B., Norberg, A., 2002. Clinical group supervision in an intensive care unit: a space for relief, and for sharing emotions and experiences of care. Journal of Clinical Nursing 11, 809–818. Lipsky, M., 1980. Street 1 level bureaucracy: dilemmas of the individual in public services. Russell Sage Foundation, New York. Mackin, P., Sinclair, M., 1998. Labour ward midwives’ perceptions of stress. Journal of Advanced Nursing 27, 986–991. Maguire, P., 2001. Uneven ground: Feminisms and action research. In: Reason, P., Bradbury, H. (Eds.), Handbook of action research, participative inquiry and practice. Sage Publications Ltd., London. Mauthner, N., Doucet, A., 1998. Reflections on a voice-centred relational method. Analysing maternal and domestic voices. In: Ribbens, J., Edwards, R. (Eds.), Feminist dilemmas in qualitative research, public knowledge and private lives. Sage Publications, London. Meyer, J., Batehup, L., 1997. Action research in health-care practice: nature, present concerns and future possibilities. NT Research 2, 175–184. Morris, M., 1995. The role of clinical supervision in mental health practice. British Journal of Nursing 4, 886–888. Morrison, B., Lilford, R., 2001. How can action research apply to health services? Qualitative Health Research 11, 436–449. Oakley, A., 1981. Interviewing women: a contradiction in terms. In: Roberts, H. (Ed.), Doing feminist research. Routledge and Kegan Paul, London. Pairman, S., 2000. Woman-centred midwifery: partnerships or professional friendships? In: Kirkham, M. (Ed.), The midwife– mother relationship. Macmillan Press Ltd., London. Raphael-Leff, J., 2000. Psychological understanding: its use and abuse in midwifery. British Journal of Midwifery 8 (11), 680–686. Riessman, C.K., 1987. When gender is not enough: women interviewing women. Gender and Society 1, 172–207. Riessman, C.K., 1993. Narrative analysis. qualitative research methods series 30. Sage Publications, London. Ribbens, J., 1989. Interviewing: an ‘‘unnatural situation’’? Women’s Studies International Forum 12, 579–592. Rolfe, G., Freshwater, D., Jasper, M., 2001. Critical reflection for nursing and the helping professions: a user’s guide. Palgrave, Basingstoke. Royal College of Midwives (RCM), 1997. Evidence to the review body for nursing staff, midwives, health visitors and professionals allied to medicine for 1997. Royal College of Midwives, London. Sandall, J., 1995. Burnout and midwifery: an occupational hazard? British Journal of Midwifery 3, 246–248. Sandall, J., 1997. Midwives’ burnout and continuity of care. British Journal of Midwifery 5, 106–111. Sandall, J., 1998. Occupational burnout in midwives: new ways of working and the relationship between organisational

ARTICLE IN PRESS 176 factors and psychological health and well being. Risk Decision and Policy 3, 213–232. Sandall, J., 1999. Team midwifery and burnout in midwives in the UK: practical lessons from a national study. MIDIRS Midwifery Digest 9, 147–152. Stapleton, H., Duerden, J., Kirkham, M., 1998. Evaluation of the impact of the supervision of midwives on professional practice and the quality of midwifery care. ENB and UKCC commissioned project. University of Sheffield, Sheffield. Stapleton, H., Kirkham, M., Curtis, P., Thomas, G., 2002. Silence and time in antenatal care. British Journal of Midwifery 10, 393–396. Starhawk, M.S., 1987. Truth or dare, encounters with power, authority, and mystery. Harper and Row, New York. Waterman, H., Tillen, D., Dickson, R., et al., 2001. Action research: a systematic review and guidance for assessment, Health Technology Assessment, 5 (23) NHS R & D HTA Programme.

R. Deery Webb, C., Kevern, J., 2001. Focus groups as a research method: a critique of some aspects of their use in nursing. Journal of Advanced Nursing 33, 798–805. Williamson, G.R., Prosser, S., 2002. Action research: politics, ethics and participation. Journal of Advanced Nursing 40, 587–593. Wilkins, P., 1998. Clinical supervision and community psychiatric nursing. In: Faugier, T.J., Burnard, P. (Eds.), Clinical supervision and mentorship in nursing, 2nd edn. Butterworth Nelson Thornes Ltd., Cheltenham. Winship, J., 1996. The UKCC perspective: The statutory basis for the supervision of midwives today. In: Kirkham, M. (Ed.), Supervision of midwives. Books for Midwives Press, Cheshire. Winter, R., Munn-Giddings, C., 2001. A handbook for action research in health and social care. Routledge in association with Taylor & Francis Group, London and New York. Yegdich, T., Cushing, A., 1998. An historical perspective on clinical supervision in nursing. Australian and New Zealand Journal of Mental Health Nursing 7, 3–24.