How participatory action research supported improvements to the postnatal discharge summary system used for remote dwelling Aboriginal mothers and infants in the Top End of Australia

Abstract

This article describes and analyses the participatory action research (PAR) process used by researchers working with policy leaders and clinicians to facilitate health service improvement around postnatal discharge planning processes for remote-dwelling Aboriginal women and their newborn infants. The research presented here was a sub-study of a larger health system improvement investigation. Stringer’s (2007) working principles of PAR – participation, relationships, inclusion and communication – were used by researchers as a framework to describe a process of engagement for the research. Application of these principles prompted changes to the discharge planning process within the health system. The improvements included: redesign of discharge paperwork; nomination of a designated health professional at the remote community to receive summaries; training for hospital staff in the computer application that generated the discharge summaries; and the development of a length of stay and discharge policy. Hospital service providers have formed a working group that meets to discuss issues around the discharge process and work continues to review and strengthen the discharge process. Hospital management reported that the changes resulted in improvements, with more likelihood that correct information is being received by the appropriate health professional in a timely manner.

Keywords

Aboriginal health system improvement Indigenous participatory action research

Introduction

In the Northern Territory (NT) of Australia, 79 percent of the Aboriginal population lives in remote areas (ABS, 2010) and 39 percent of total births are to Aboriginal women (Tew & Zhang, 2010). The provision of maternity services in a remote environment is challenging. This is exacerbated by the poorer health status of Aboriginal Australians, which contributes to higher rates of maternal and perinatal morbidity and mortality (Laws, Li, & Sullivan, 2010; Tew & Zhang, 2010). There is no written government policy in the NT on where birth should take place although the clinical protocols of the NT Department of Health (DH) – formerly the Department of Health and Families (DHF) – strongly encourage remote-dwelling Aboriginal women to give birth in regional hospitals (Department of Health and Families, 2009). Arrangements are made for these women to transfer into regional centres prior to the anticipated onset of labour (Department of Health and Families, 2009).

Despite poorer outcomes, remote-dwelling women have significantly less access to trained service providers locally. Remote communities experience high levels of staff turnover and staff shortages can influence service provision. Many communities do not have resident midwives or doctors and rely on nurses, Aboriginal Health Workers (AHWs) and outreach services from the regional centre to provide care. Remote-dwelling women generally remain in hospital for three to five days following birth and are then either discharged home directly to their community or to a hostel to await a flight home. The regional hospital provides early postnatal care and should facilitate the safe return of a woman and infant to their home community. It is the responsibility of the regional hospital to ensure a discharge summary for mother and infant is sent to the community health centre (Department of Health and Families, 2009) where subsequent postnatal care is provided. Discharge information is relayed between health service providers via telephone, fax and email.

Postnatal care beyond the brief period of hospitalization is very important for Aboriginal women and infants because their health outcomes are considerably worse than their non-Aboriginal counterparts.

Any patient moving through the health system is susceptible to ‘periods of enhanced risk’ (Maslove et al., 2009). Nowhere perhaps is this risk more evident than at the point of discharge from hospital where poor, delayed or absent transfer of information can result in poor outcomes for patients (Callen, Alderton, & McIntosh, 2008; Kripalani et al., 2007; O Leary et al., 2009). Multiple handovers of information between care providers, particularly between hospital-based services and primary health care facilities, have been highlighted as points of increased risk for patients (Moore, Wisnivesky, Williams, & McGinn, 2003).

A recent study that investigated the journey of Aboriginal people living in remote areas of the NT who travelled to a major metropolitan hospital and back again showed that travel details, discharge information and prescriptions were not relayed to travel and care providers (Lawrence et al., 2009). This resulted in inappropriate follow up of Aboriginal patients returning to their communities. AHWs and Remote Area Nurses (RANs) often came upon these patients by chance. The tragic death of an elderly Aboriginal man at an airstrip in the NT following his discharge from hospital is an example of miscommunication between services leading to an avoidable fatality (Chalmers, 2007).

The aim of this article is to describe how PAR principles as described by Stringer were enacted to support health systems improvement and a redesign of the discharge process. This work involved a broader team of researchers, policy-makers, clinicians and managers sharing the results of baseline data analysis and identifying and solving problems. It also reports the experience of the research team in applying these principles.

The health system investigation (the ‘larger study’) was initiated by Aboriginal women working with our research team. We do not present the process initiated by Aboriginal women here or our continued PAR engagement with them. This will be reported fully elsewhere.

Background to the sub-study

The larger study, named the 1 + 1 = A Healthy Start to Life project, was developed in response to recommendations made over a number of years by Aboriginal women for improvement to maternity services (Kildea, 1999; NT DHCS, 1992; Watson, Hodson, Johnson, & Kemp, 2002; Watson, Hodson, Johnson, Kemp, & May, 2002). Three senior researchers and four student researchers worked on the Healthy Start to Life project. Students were responsible for some of the sub-studies with senior researcher guidance. A description of the Healthy Start project is provided in Figure 1.

Figure 1.

Description of the Healthy Start to Life Project.

The larger study began with initial scoping work undertaken by our team discussing the issues with Aboriginal women. The women were flown subsequently from two large remote communities to the regional centre to join a group of clinicians and senior policy-makers to clarify joint goals (Barclay & Kildea, 2006). This scoping work confirmed the need for improvement to maternal and infant health services and a shared desire by all participants to change services. The researchers then organized workshops in the remote centres to confirm the research was welcome in the participants’ communities. The funding application for this research, prepared subsequent to the scoping activity, outlined how research would support and investigate health system improvement. The design used mixed methods and a PAR approach with Aboriginal women and health providers.

We were awarded National Health and Medical Research Council (NHMRC) funding in 2007. As part of baseline studies, the third author led investigations into the quality of maternity services and maternal and infant health outcomes using record audit (2004–2006), observation, interviews and analysis of standard data. Services investigated included primary health care centres in the remote communities, the tertiary referral centre approximately one hour’s flight away, transport, hostel accommodation and community services (Bar-Zeev, Barclay, Kruske, Bar-Zeev, Gao, & Kildea, forthcoming).

The baseline analysis highlighted numerous discontinuities in the women’s journey through maternity services. Miscommunication across services, specifically at the point of discharge from hospital to the community, was identified as a particular problem by the researchers, policy makers and clinicians who made up our Health Advisory Group (HAG).

In the sub-study reported here a senior researcher, Lesley Barclay (Lesley), and a student researcher, Cath Josif (Cath), worked with Maureen Brittin (Maureen), a health services manager, and her colleagues at the regional hospital to improve discharge processes from the regional centre to remote communities.

Reflexivity and researcher identity

It has been argued that reflexivity is as vital to a publication reporting action research as the methodological and literature review statements (Bradbury-Huang, 2010). Reflexivity is described as an ‘active, ongoing process that saturates every stage of the research process’ (Guillemin & Gillam, 2004, p. 274). It reminds us that the researcher is inextricably linked to their past and life experience. This in turn, colours the way researchers perceive the world (Grbich, 2004). The reflexive researcher needs to ask him- or herself ‘who am I?’ and ‘what part do I play?’ in the research process and constantly review choices of research method, data collection tools and how data are analysed. Seale (1999) suggests that preoccupation with ‘self’ is little more than a smokescreen for poor research; however, others disagree by arguing that reflexivity encourages rigour in research because it locates the researcher socially and politically in the research process (Liamputtong & Ezzy, 2005). This is far from ‘autobiographical self-indulgence’ but, rather, aims to ‘contextualize the claims, create transparency and also to anchor ownership of expression that can otherwise masquerade as worryingly disembodied and neutral’ (Bradbury-Huang, 2010, p. 95).

The personal experiences, team reflection and the leadership style of the team contributed towards ameliorating unequal power relations that may have arisen during the course of the project. The first author, Cath, is a research student with a background in remote area nursing, child health and midwifery. Her interest lies in promoting partnerships between professionals, women and their families – a key element in providing effective maternity services and family healthcare. Cath was first introduced to PAR when she worked on a playgroup development program in Aboriginal communities in Central Australia. The experience of working with Aboriginal families to strengthen family functioning in response to agreed community needs, and the potential of the methodology, inspired an enduring interest in PAR. During her time in Central Australia Cath also met Ernest Stringer. Ernest Stringer has collaborated extensively with Aboriginal people on community development programs and authored a number of action research texts. It was this meeting that prompted Cath to read the author’s work on action research and, as a novice PAR researcher, to apply the working principles of PAR as described by him. Cath has been able to pursue and consolidate her understanding of PAR through involvement with the Healthy Start to Life project and the sub-study, which she led, focusing on improvements to the discharge summary process.

Cath is guided in her academic career by mentors and supervisors with experience and commitment to PAR as a means of supporting health systems improvement. The second and fourth authors, Lesley and Sue Kildea (Sue) are two of these mentors and supervisors. The originator of the Healthy Start to Life funding proposal and team leader is Lesley. She is a senior health researcher and reformer able to link academic researchers, non-government health service partners and remote community partners with senior levels of the DH to effect system change. Lesley is well skilled in PAR and has taught participatory processes in community development and education. Sue, a chief investigator on the project, is a nurse and midwife who has worked extensively with Aboriginal communities including the two communities that are the focus of the larger project. She has developed strong relationships with both of these communities, considered vital when undertaking work with Aboriginal communities. She is passionate about improving the quality of maternity care for Aboriginal women and her own doctoral studies used PAR in four case studies in remote Australia with the overarching aim of strengthening remote maternity services (Kildea, Barclay, Wardaguga, & Dawumal, 2009).

The third author, Sarah, is a fellow research student working on the larger project and is also a nurse midwife; her research provided the baseline data for this PAR sub-study. Sue, Cath and Sarah have all worked as RANs and appreciate the challenges of health service delivery in remote communities. The fifth author, Maureen, is a nurse midwife and senior health service manager who worked with the research team and led much of the health system improvement around the discharge process in the hospital.

How we approached the sub-study

PAR is a ‘participatory, democratic process concerned with developing practical knowing in the pursuit of worthwhile human purposes, grounded in a participatory worldview’ (Reason & Bradbury, 2001, p. 201). In action research there are no ‘subjects’, rather ‘participants’ and it is the participants who by working in partnership with researchers, identify problems, determine the research agenda and ultimately benefit from the research.

Much has been written about PAR and the cycles, spirals and helixes that depict the different stages of the PAR process (McNiff & Whitehead, 2006; Reason & Bradbury, 2001; Stringer, 2007; Stringer & Genat, 2004; Wicks & Reason, 2009). There is less literature, however, around the challenges faced by researchers as they attempt to establish partnerships and open a ‘communicative space’ (Wicks & Reason, 2009). Communicative space has been described as ‘rather like tidal wetlands where salt and freshwater mix, they are not restful places but continually changing and offering new possibilities and challenges’ (Wicks & Reason, 2009, p. 258). This is a particularly appropriate metaphor for this study given its location in the ‘Top End’ of tropical northern Australia. It also resonates because it elicits a sense of the fluid exchange of ideas that is possible in such a space and the notion that anything can happen, and often does. In fact, similar metaphors are used by the Yolngu¹ people of East Arnhem Land. ‘Ganma’ has been described as a metaphor; an Indigenous theory; a social theory explaining how Yolngu society works; and a place where new knowledge is recreated (Hughes, 2000; Marika, Ngurruwutthun, & White, 1992; Pryrch & Castillo, 2002; Yunupingu & ABC, 1994). Mandawuy Yunupingu² used the Ganma metaphor to describe this constant state of flux, and its dependence on balance, relationships and harmony (Yunupingu & ABC, 1994).

Ganma is taken as describing the situation where a river of water from the sea (in this case Balanda [non-Aboriginal] knowledge) and a river of water from the land (Yolngu knowledge) mutually engulf each other on flowing into a common lagoon and become one. In coming together the streams of water mix across the interface of the two currents and foam is created at the surface so that the process of Ganma is marked by lines of foam along the interface of the two currents. In the terms of the metaphor, then the line of foam that is formed by the interaction of the two currents marks the interface between the current of Yolngu life and the current of Balanda life. Both Yolngu and Balanda can benefit from theorising over the interaction between the two streams of life. (Marika et al., 1992, p.26)

Researchers were confronted with a number of challenges in the conduct of the overarching Healthy Start to Life project. These included working cross-culturally, difficult weather conditions and being geographically isolated. These aspects of the larger project will be reported elsewhere. However, being town-based and regional hospital focused the sub-study was less affected by these challenges and there were fewer difficulties establishing and maintaining communicative space between researchers and health providers. The working principles of PAR – participation, relationships, inclusion and communication, as described by Stringer (2007) – were used by researchers in the small sub-study. The researchers worked with health service providers and policy-makers to design strategies that supported improvement to the discharge process for women and infants returning from the regional hospital to their remote communities.

Participation

‘Active’ participation is vital to the sense of ownership that encourages people to spend time moulding the nature and quality of activities and behaviours in which they participate (Stringer, 2007). The researchers sought to link research, health policy and service interventions through the representation and participation of relevant institutions. Participation and governance were encouraged and supported through the establishment of advisory groups.

Local reference groups (LRGs) were established at the two field sites as part of the larger project. The LRGs comprised Aboriginal women, remote health centre and Aboriginal health board staff and members of the Community Councils. In the regional setting a HAG was established and membership included health management staff, policy advisers, researchers and clinicians. Some of the researchers and clinicians were members of both the HAG and the LRGs.

The research team used the data collected by Sarah as part of the baseline phase to prompt discussion regarding problems in health service delivery. Efforts to improve the health system were facilitated through the HAG acknowledging that the key stakeholders in the research process are those best placed to identify and solve their own problems.

Twenty-nine people attended the HAG meeting held in October 2008. At this meeting the group decided that making improvements to the discharge process was a major priority. This became the focus of the sub-study reported on in this article. Significant issues with the discharge process included:

The electronic data collection system defaulted to normal vaginal birth if the field was left blank.

Significant delays in discharge summaries arriving at the health centre.

Missing discharge summaries (Graduate School of Health Practice, 2008).

As a result of the above the remote health centre staff either failed to receive discharge information or received incorrect information. They were frequently unaware when a woman and her baby returned to the community, resulting in their missing out on important care (Bar-Zeev, Barclay, Farrington, & Kildea, 2011). Numerous examples of missed medication and treatment were observed, and information such as newborn admissions to the Special Care Nursery (SCN) and incidences of postpartum haemorrhage that had required surgical intervention were not communicated to the remote health centre staff (Bar-Zeev et al., 2011).

A small group led by Lesley and comprising researchers and the midwifery services manager at the regional hospital formed to focus on this sub-study. Twenty-four face-to-face meetings and telephone calls took place between stakeholders and researchers over a six-month period. An informal agenda or topic was usually the focus and meetings generally lasted for 30 minutes. The research team used these meetings to provide support and track the progress of improvements to document information, and to remind stakeholders of actions they had planned to conduct. Four separate meetings took place between researchers to provide updates on progress towards system improvement.

Six months after the group convened the hospital representatives in the study reported the following changes:

Discharge summaries were in the process of being redesigned by Information Technology staff so that the default option to vaginal birth was removed and birthing information had to be entered manually.

Maternity ward staff had been educated regarding despatch of discharge summaries so that a designated person in each community was nominated to receive them.

Training in the computer application responsible for the generation of discharge summaries was initiated weekly and as part of the induction of new staff to the maternity services unit.

Business rules were developed for the software and made available for all computers in midwifery services. The business rules provided instruction on the generation and completion of discharge summaries. This reduced the likelihood of inaccurate or incomplete summaries being forwarded to the remote health centre.

A length-of-stay and discharge policy with a focus on streamlining information systems was developed.

Case conferencing and discharge planning was identified as a priority for women with infants discharged from SCN.

Hospital representatives reported that they were receiving fewer requests from community health centres for ‘missing’ discharge summaries and that the content of the discharge summaries had improved (C. Josif, field notes, 2009). Hospital representatives were also able to identify instances where discharge summaries had been despatched but had not been downloaded in communities. The research team was able to feed this information back to members of the HAG who in turn followed up with the health centres.

The researchers and other colleagues confirmed that, together, progress had been made in resolving the problems identified in the baseline data regarding the discharge summaries, and acknowledged that more work was required to fully embed the changed and improved system.

An example of this was when one hospital representative who was interviewed observed that a backlog in despatching discharge summaries could be due to lack of orientation for Resident Medical Officers (RMOs) around the importance of completing discharge summaries accurately and in a timely fashion (C. Josif, field notes, 2009). There was also a high turnover of RMOs and midwifery staff. Because the discharge summary had to be completed by an RMO there was often a lag in completion and despatch of information. As a way of further streamlining the discharge process hospital representatives were considering a process where women who had a normal antenatal, birth and postnatal period could have their discharge summary completed and despatched by a midwife (C. Josif, field notes, 2009).

Relationships

Much emphasis was placed from the outset of the larger study on establishing relationships between participants and researchers. As described by Stringer, the attributes of relationships directly influence both the quality of peoples’ experience and the outcomes that can be expected (Stringer, 2007).

At the time of the larger study Lesley was a Professor of Health Services Development and the Director of a postgraduate school at the university based in the regional centre where women transferred for birth. This school itself was a joint endeavour between the DH and the university that facilitated a closer alignment between policy, practice and research (Graduate School Health Practice, 2009). It had been established by Lesley in 2005 to promote and facilitate links between academic researchers, non-government organizations (NGOs), remote communities and the DHF through postgraduate education and research to effect system change. Organizational relationships therefore were favourable for the sub-study, as much effort had already been invested in these as part of the larger study.

Regular research team meetings allowed for an update of progress and different insights and perspectives to be shared between team members involved in different parts of the larger project. This broader group developed cohesiveness and provided a forum where project goals and methods could be regularly revisited to allow for the consistent application of the principles of PAR both in the Healthy Start project and the small sub-study.

Senior researchers with a background in PAR processes actively supervised and mentored students in using these techniques. Our shared vision and praxis was achieved through relationships within and outside the research team. The respectful and collegial relationships established between project team members modelled relationships that developed between the research team and other participants in the larger research project.

Maintaining close and effective relationships outside this small research group required considerable effort. An overarching PAR plan for the Healthy Start to Life project was developed early to guide the research process and researchers included relevant extracts from their field notes in this plan. Cath and Lesley regularly followed up with service providers by telephone, email and face-to-face contact to review how improvement to the discharge process was progressing.

Good relationships in the sub-study were maintained as Cath and Lesley met and spoke with service providers in the health system, particularly those working in maternity services and in the remote health centres, to follow up on progress in regard to discharge planning and the information systems underpinning this. Meetings were held in venues nominated by the health service staff and at times selected by them. Emails were sent to various stakeholders and members of the research team to seek clarification, update and follow up on different aspects of the process. The teleconferences and face-to-face meetings between Cath and Lesley specifically relating to the discharge process allowed for feedback, discussion and planning in how best to support improvements to the discharge process and mentoring in PAR.

PAR promotes relationships that do away with the cold and controlling business-like relations characteristic of many bureaucracies (Stringer, 2007), which can appear hostile to outsiders. Our correspondence, while seeking information, was chatty, familiar and actively personal rather than bureaucratic, as the following example shows:

Hi X

Belated New Year greetings! I hope all is well? I am just following up on a few things we spoke about at interview in regard to discharge processes and information transfer

I know you are busy X … so whatever is easiest for you in regard to answering questions … over the phone or by email or I can pop in … what is the most convenient for you?

This elicited the following response:

Hi Y

I hope your year has started well. Hard to believe Feb already nearing its end!

(Lengthy response to questions)

Don’t hesitate to let me know if you need any more info Y

Cheers

(X, personal communication 2010)

Inclusion

A PAR approach involves all participants in the research process. Stringer (2007) describes how PAR fosters surroundings that allow different groups to work through their agendas in a trusting and supportive environment and to work towards resolving issues that concern them. The formation of a HAG and LRGs created an inclusive governance structure and an environment for people to come together to resolve issues they identified and to work through solutions facilitated and supported by the research team.

The logistics and expense of bringing the LRGs and HAG together after an initial workshop were prohibitive. The two communities participating in the larger project are approximately 500 km by road from the regional centre and the roads are impassable during the monsoon season. Air travel is expensive with flights costing approximately AU$400 return. Therefore researchers acted as links between the two communities. The extract below from research meeting minutes illustrates how the research team sought to work inclusively:

Show articles in draft so community can see what is being written and make comment.

Discussion re potential conflict with reviewers and community members over naming of community.

X to provide copy of bound honours thesis to local museum/library.

Feedback responses from presentation of material at conferences.

Take PowerPoint slides out to show them around.

Bring comments and concerns back from community to research group (NHMRC & ARC Research Meeting Minutes, 2009).

The HAG’s mix of researchers, professionals and high-level policy-makers had an authority and commitment considered vital to effect system change at an operational level. This group met twice a year and provided guidance and critique of the research. The logistics of convening the HAGs were complex with researchers needing to factor in senior HAG members’ timelines and availability. All HAG meetings were chaired by Lesley. The principle of ‘inclusion’ within this group is highlighted in research updates published in an email distributed newsletter from HAG meetings. The first example references how the project sought to involve health staff from the outset:

In August 2007 we invited key DHF staff to learn about the program and will hold another update in early 2008. Our plan is to do this every six months so that Department staff and key practitioners are involved every step of the way. (Graduate School of Health Practice, 2007)

In the second example, the research update describes how the HAG and researchers worked together effectively:

[We] identified a number of issues through group discussion that could significantly improve care. (Graduate School of Health Practice, 2008)

The LRGs and HAG sought to dissolve the hierarchical structures often associated with traditional research where control is firmly in the hands of the researchers. The inclusive nature of both the larger project and the sub-study maximized the involvement of – and gave voice to – all participants involved in the research.

Communication

From the outset communication strategies were designed for discrete audiences: clinicians, remote community members and health service staff. This was considered important to make sure that feedback and participation incorporated the perspectives of all participants in both the larger project and the sub-study.

Effective communication in PAR means the researcher is an attentive listener who accepts and understands others viewpoints, is easily understood, is honest, behaves in a socially and culturally appropriate manner and regularly updates others about what is going on (Stringer, 2007). Consistent and ongoing communication is an important part of the PAR process. For the sub-study, Cath and Lesley met and spoke with our clinical and policy colleagues six times in the six months after issues with the discharge summary process were identified. Cath went on to make contact at three-monthly intervals, either in person or by email or telephone, to review ongoing progress. She also made contact with a number of other stakeholders in the DH to follow up on computerized data systems and experiences of stakeholders with the discharge process.

A review of Cath’s field notes identifies 17 contacts overall with stakeholders seeking information or clarification on issues relating to the discharge summary process. Contacts made by researchers reminded people about work still needing to be done, checked progress, encouraged individuals and linked them across roles and departments. Communication was of paramount importance to the researchers and contributed to the success of the sub-study. Considerable efforts were made to ensure effective communication in the PAR process was maintained. The engagement of participants and their enthusiastic and willing participation in the processes we used suggest that this was achieved.

The LRGs met in the field sites and provided a forum for both formal and informal discussions about the larger study. We worked with interpreters when required as there were 11 languages spoken at one of the research sites. The HAG met twice-yearly. The emphasis of these meetings was to describe progress and discuss remaining opportunities for improvement. We worked hard to avoid any perception by stakeholders that we were outsiders criticizing their work practices or telling them what to do. This could have been easily felt given the professional standing of several of the investigators on the study and could have fostered resentment or defensiveness rather than participation. Our style of communication, that was respectful of the challenges of achieving improvement and admiring of their successes helped reinforce a commitment to change systems.

Regular updates on the Healthy Start to Life project and the sub-study were provided via an e-newsletter that was distributed twice a year. The newsletters allowed for a summary of the meetings and key findings so that those who were unable to attend the meetings could be kept informed. In total, seven newsletters were distributed, providing a record of progress and subsequently being used as briefing documents when others wanted to know about the Healthy Start to Life project. The newsletters were circulated initially to 69 recipients. Two years later the official circulation list had grown to 84 recipients. The newsletters were also made available on the university website.

Discussion

The aim of this article is to describe how the principles of PAR as described by Stringer – participation, relationships, inclusion and communication – were enacted to support health systems improvement. Certainly the application of the principles in this instance contributed to the participants’ ability to identify a problem, seize the opportunity to make improvement, make progress towards rectifying the identified issue and identify areas requiring work to improve the discharge process. The problems inherent in establishing communicative space and building relationships were ameliorated by the application of these principles.

Establishing a communicative space and building participative, inclusive relationships is essential for successful PAR (Arieli, Friedman, & Agbaria, 2009; Wicks & Reason, 2009). This communicative space is often difficult to achieve, especially with research involving vast geographical distances and such a variety of stakeholders. There is a dearth of studies however dealing with how researchers can do this successfully (Arieli et al., 2009). It is important therefore to report on PAR processes as this provides useful examples for other researchers in how to go about achieving changes in their service (Courtney, 2005).

The objectives of the larger Healthy Start to Life project were to improve health system delivery, and thus maternal and infant outcomes, for remote-dwelling Aboriginal families. The work around improving discharge planning processes was a smaller study nested in the larger study but sharing the same overall objectives. Many of the processes described here were used with other small studies also nested in the larger research project. It was considered crucial to the successful implementation of this project that the research was developed in response to DH priorities and an expressed community need. The project and this sub-study met these joint objectives.

Authors emphasize the importance of ‘authentic commitment’ (Liamputtong & Ezzy, 2005) in the conduct of PAR. Nowhere is this authenticity more important than in the establishment of relationships between researchers and participants. Relationships established in PAR cloud the bounds of usual roles of researchers and the researched (Arieli et al., 2009). The partnerships established between researchers and participants and the efforts of participants to rectify issues identified with the discharge process demonstrate the extent to which the research team established relationships and shared authentic commitment.

The development of partnerships between researchers and participants was crucial to the facilitation and management of change (Khresheh & Barclay, 2007; Reason & Bradbury, 2001). The researchers worked hard to build relationships and engage policy-makers and clinicians and continued this through formal and informal feedback and a range of activities. This article reports how this worked in the sub-study, and it succeeded as part of the overall system improvement agenda.

While it is acknowledged that establishing relationships is important in PAR this is difficult to do, time-consuming and not always successful. Our approach allowed for a climate where the researcher was a ‘facilitator and creative investigator’ (Stringer, 2007), rather than a distant uninvolved overseer.

University-based researchers write of the personal challenges they face stepping outside ‘the comfort zone of the circumscribed roles for which graduate school prepared us’ (Smith, Bratini, Chambers, Jensen, & Romero, 2010, p. 409) when they become members of a university research team involved in PAR projects. The experience for student researchers working on this project was very different but the research itself created significant challenges.

PAR takes you out of your comfort zone because it is time-consuming, exhausting and confronting. Just as tidal wetlands constantly change with the ebb and flow of water, so too researchers in PAR must accept that any number of things may happen that will turn meticulously formulated research plans upside down and affect the research course. As Cath found in this project, meetings and appointments can be cancelled and require rescheduling and competing work priorities mean that people don’t always get back to you as planned. This in turn means that at a regular team research meeting you don’t always have a great deal to report. As a novice PAR researcher one can become frustrated that the whole process seems to be taking too long and secretly wish to speed things up a little!

PAR is challenging. There is a blurring of roles of researcher and participant that serves to maximize participant involvement but also changes the nature of the relationships. This can cause ‘moral distress’, a phenomenon well recognized in health care where practitioners experience distress because they believe institutional limitations prevent them from behaving in the way they feel is required (Corley, Elswick, Gorman, & Clor, 2001). For instance, a research team member adopts a number of roles when working with partners. In the sub-study described here, partners included the university, the DH (in both the remote and urban setting), the funding body and the two remote communities. A tension exists between satisfying the needs of all these partners and, as a student researcher, the demands of supervisors. One can feel like the servant of many masters at risk of meeting the needs of none. The consequence of this can be burnout and disillusionment with the PAR process (Corley et al., 2001).

Senior researchers were supervisors and mentors for novice research students, thus guiding the next generation of researchers into PAR as a research method or approach of choice. The role of the senior researchers as mentors was a crucial one because it provided students the opportunity to debrief with colleagues who had both prior knowledge of PAR and experience of working in the communities that were part of this sub-study. Students were also encouraged to take the lead in various aspects of the sub-study with the knowledge that they were well supported by their mentors.

By reviewing how the principles of PAR were applied in this instance the authors attempted to ‘unpack’ how PAR can support health system change and describe their roles and experience in relation to this.

Conclusion

This article describes a PAR process applied to address fragmented and dangerous discontinuities in a postnatal health care discharge system. A baseline analysis of the patient journey from the remote health centre to hospital for birth and back highlighted a number of discontinuities to care. Risks to client safety were identified because of problems with the discharge process, particularly the accuracy of a patient’s discharge information and timeliness of despatch. PAR provided the participants; health service providers, service users and researchers, with opportunities to collaborate to resolve practical problems and construct action and knowledge in a united manner via a cyclical process (O’Leary, 2005).

The researchers on the sub-study reported here successfully applied PAR processes to support a significant improvement to the discharge process within the health system. Evidence from our research informed health service redesign and contributed to improvement in the continuity of care for women and infants returning to their communities after hospitalization for birth.

Footnotes

Glossary of acronyms

AHW

Aboriginal Health Worker

NT Department of Health

DHCS

NT Department of Health and Community Services

DHF

NT Department of Health and Families

HAG

Health Advisory Group

LRG

Local Reference Group

NGO

Non-government organization

NHMRC

National Health and Medical Research Council

Northern Territory

PAR

Participatory action research

RAN

Remote Area Nurse

RMO

Resident Medical Officer

SCN

Special Care Nursery

Acknowledgements

The authors would like to thank the anonymous reviewers for their review of this article. The first author in particular is grateful for the time and effort the reviewers put in to their critique of this article through many revisions. Their comments have been most helpful in crafting this article. We thank Hilary Bradbury-Huang, for leading the review process for the author of this paper. Should there be any comments/reactions you wish to share, please bring them to the interactive portion (Reader Responses column) of the website: .

Notes

Cath Josif is a registered nurse, child health nurse and midwife. She has extensive experience living and working with Indigenous communities both as a remote area nurse and working on community development programs. She is a PhD candidate at the University of Sydney.

Lesley Barclay is Professor & Director of the University Centre for Rural Health, Lismore NSW. Lesley has worked in international development for over 2 decades providing advice in primary health care, maternal infant/child health and capacity building in health worker education systems. As a researcher she works closely with governments and local health services to assist in health system improvement.

Sarah Bar-Zeev is a registered nurse and midwife with experience working with Indigenous communities in Central Australia. She holds Masters degrees in Public Health and Nursing Science.

Sue Kildea is a registered nurse and midwife with clinical, management, policy, education and research experience across acute and primary health care settings. Sue is currently Professor of Midwifery at the Mater Hospital, South Brisbane and the Australian Catholic University.

Maureen Brittin is a registered nurse and midwife with extensive experience working in Intensive Care. Maureen is currently the co-director of maternal and child health at Royal Darwin Hospital. Maureen was awarded an Order of Australia for her role in caring for victims of the 2002 Bali Bombings.

References

ABS. (2010). Population Characteristics, Aboriginal and Torres Strait Islander Australians, 2006. Retrieved 29 August 2012, from http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/6E6D19F5BB55AD66CA2578DB00283CB2?opendocument#.

Arieli

Friedman

Agbaria

. (2009). The paradox of participation in action research. Action Research, 7(3), 263–290.

Bar-Zeev

Barclay

Farrington

Kildea

. (2011). From hospital to home: The quality and safety of a postnatal discharge system used by remote dwelling Aboriginal women and their infants in the Top End of Australia. Midwifery, 28(3), 366–373.

Bar-Zeev, S., Barclay, L., Kruske, S., Bar-Zeev, N., Gao, Y., & Kildea, S. (forthcoming). Maternal health service use among remote dwelling Aboriginal women in northern Australia and their disease burden. Birth.

Barclay, L. M., & Kildea, S. M. (2006). Improving Primary Health Care Maternity Services for Indigenous Women. Report of Workshop Proceedings, 16–17 May 2006. Prepared for Department of Health and Community Services.

Bradbury-Huang

. (2010). What is good action research?: Why the resurgent interest. Action Research, 8(1), 93–109.

Callen

Alderton

McIntosh

. (2008). Evaluation of electronic discharge summaries: A comparison of documentation in electronic and handwritten discharge summaries. International Journal of Medical Informatics, 77(9), 613–620.

Chalmers, E. (2007). Report of the Investigation into the disappearance and subsequent death of an elderly gentleman transferred from Katherine Hospital to kalkaringi in August 2006. Katherine West Health Board and Department of Health and Community Services.

Corley

M. C.

Elswick

R. K.

Gorman

Clor

. (2001). Development and evaluation of a moral distress scale. Journal of Advanced Nursing, 33(2), 250–256.

10.

Courtney

. (Ed.) (2005) Evidence for nursing practice. Sydney: Elsevier.

11.

Department of Health and Families. (2009, November). Remote Health Atlas. Retrieved 31 March 2010, from http://www.health.nt.gov.au/Remote_Health_Atlas/index.aspx.

12.

Graduate School of Health Practice. (2007). 1+1 A Healthy Start to Life – Research Update. Charles Darwin University, Darwin, NT.

13.

Graduate School of Health Practice. (2008). 1+1 A Healthy Start to Life – Research Report, No 5. Charles Darwin University, Darwin, NT.

14.

Graduate School Health Practice. (2009). Annual Report. Charles Darwin University, Darwin, NT.

15.

Grbich

. (2004) New approaches in social research. London: SAGE.

16.

Guillemin

Gillam

. (2004). Ethics, reflexivity, and ‘ethically important moments’ in research. Qualitative Inquiry, 10(2), 261–280.

17.

Hughes, I. (2000). Ganma: Indigenous knowledge for reconciliation and community action. Retrieved 20 December 2004, from www.fhs.usyd.edu.au/arow/arer/014.htm.

18.

Khresheh

Barclay

L. M

. (2007). Practice – research engagement (PRE) Jordanian experience in three Ministry of Health hospitals. Action Research, 5(2), 123–138.

19.

Kildea, S. (1999). And the women said…Report on birthing services for Aboriginal women from remote Top End communities. Territory Health Service, Darwin, NT.

20.

Kildea

Barclay

Wardaguga

Dawumal

. (2009). Participative research in a remote Australian Aboriginal setting. Action Research, 7(2), 143–163.

21.

Kripalani

LeFevre

Phillips

Williams

Basaviah

Baker

. (2007). Deficits in communication and information transfer between hospital based and primary care physicians: Implications for patient safety and continuity of care. The Journal of the American Medical Association, 297(8), 831–841.

22.

Lawrence, M., Dodd, Z., Mohor, S., Dunn, S., de Crespigny, C., Power, C., & Mackean, L. (2009). Improving the patient journey: Achieving positive outcomes for remote Aboriginal cardiac patients. Cooperative Research Centre for Aboriginal Health, Darwin, NT.

23.

Laws

Sullivan

. (2010) Australia’s mothers and babies 2008 (Vol. Perinatal statistics series no. 24). Sydney: AIHW National Perinatal Statistics Unit.

24.

Liamputtong

Ezzy

. (2005) Qualitative research methods (2nd ed.). Melbourne: Oxford University Press.

25.

Marika

Ngurruwutthun

White

. (1992). Always together, yaka gana: Participatory research as part of the development of a Yolngu education. Convergence, 25(1), 23–29.

26.

Maslove

Leiter

Griesman

Arnott

Mourad

Chow

Bell

. (2009). Electronic versus dictated hospital discharge summaries: A randomized control trial. Journal of General Internal Medicine, 24(9), 995–1001.

27.

McNiff

Whitehead

. (2006) All you need to know about action research. London: SAGE.

28.

Moore

Wisnivesky

Williams

McGinn

. (2003). Medical errors related to discontinuity of care from an inpatient to an outpatient setting. Journal of General Internal Medicine, 18, 646–651.

29.

NT DHCS. (1992). Review of Birthing Services in the Northern Territory of Australia. Department of Health and Community Service, AGPS, Darwin, NT.

30.

O’Leary

. (2005) Researching real-world problems: A guide to methods of enquiry. London: SAGE.

31.

O’Leary

Liebovitz

Feinglass

Liss

Evans

Kulkarni

Baker

. (2009). Creating a better discharge summary: Improvement in quality and timeliness using an electronic discharge summary. Journal of Hospital Medicine, 4(4), 219–225.

32.

Pryrch

Castillo

. (2002). The sights and sounds of indigenous knowledge. In: Reason

Bradbury

(Eds.), Handbook of action research. Thousand Oaks, CA: SAGE.

33.

Reason

Bradbury

. (2001) Handbook of action research: Participative inquiry and practice. London: SAGE.

34.

Seale

. (1999). The quality of qualitative research. Qualitative Inquiry, 5(4), 465–478.

35.

Smith

Bratini

Chambers

D.-A.

Jensen

Romero

. (2010). Between idealism and reality: Meeting the challenges of participatory action research. Action Research, 8(4), 407–425.

36.

Stringer

. (2007) Action research (3rd ed.). London: SAGE.

37.

Stringer

Genat

W. J

. (2004) Action research in health. Englewood Cliffs, NJ: Pearson Education Incorporated.

38.

Tew, K., & Zhang, X. (2010). Northern Territory Midwives Collection. Mothers and Babies 2006, Darwin, NT.

39.

Watson

Hodson

Johnson

Kemp

. (2002). The maternity experiences of indigenous women admitted to an acute care setting. Australian Journal of Rural Health, 10(3), 154–160.

40.

Watson

Hodson

Johnson

Kemp

May

. (2002). Opinions of healthcare professionals regarding the maternity experiences of indigenous women in an acute care setting. Australian Journal of Rural Health, 10(3), 161–167.

41.

Wicks

P. G.

Reason

. (2009). Challenges and paradoxes of opening communicative space. Action Research, 7(3), 243–262.

42.

Yunupingu

ABC . (1994). Finding the balance. Voices from the land (pp. 1–11). Sydney: ABC Books.