Nurse Researcher
An action research approach to practice, service and legislative change Cite this article as: Sullivan E, Hegney D, Francis K (2013) An action research approach to practice, service and legislative change. Nurse Researcher. 21, 2, 8-13. Date of submission: April 30 2012. Date of acceptance: April 12 2013. Correspondence to Elise Sullivan [email protected] Elise Sullivan RN, master of public policy, is a PhD student at the Department of Nursing and Midwifery, Monash University, Melbourne, Australia and a director of DPAR Consulting, Myrniong, Victoria, Australia Desley G Hegney RN, PhD is a professor of nursing at the School of Nursing and Midwifery, Sir Charles Gairdner Hospital, Perth, Australia and a programme co-ordinator, vulnerable populations at the Curtin Health Innovation Research Institute (CHIRI), Curtin University, Perth, Australia Karen Francis RN, PhD is a professor and head of the School of Nursing Midwifery and Indigenous Health at Charles Sturt University, New South Wales, Australia Peer review This article has been subject to double-blind review and checked using antiplagiarism software Author guidelines http://nr.rcnpublishing.com
8
Abstract Aims To describe the action research approach taken to engage a multidisciplinary group of health professionals and managers from five rural health services with government officers in redesigning their emergency care services and informing legislative change. Background The diminishing size of the medical workforce across rural Victoria in Australia captured the Victorian state government’s attention when this threatened the sustainability of emergency care services in rural and remote hospitals in 2006. The government funded the collaborative practice model pilot between 2006 and 2008 to develop and test an alternative model of emergency care service in which nurses practised at a more advanced and autonomous level. Data sources Data were sourced from a combination of interviews, focus groups and patient records. Review methods Qualitative data were analysed using convergent interview and thematic analysis. Quantitative data were analysed using frequencies and cross tabulations.
Introduction THE GAP between demand for and supply of health professionals is growing at an accelerated rate in Australia, as the population ages and the available workforce contracts (Rural Workforce Agency, Victoria 2006). Rural hospitals across the state of Victoria have experienced this trend most acutely, struggling to maintain 24-hour emergency care services as the size of the medical workforce has reduced (Fowles 2006a, 2006b, 2006c, Schmeiszl 2006, Scopelianos 2006).
November 2013 | Volume 21 | Number 2
08-13 NR Nov 2013.indd 8
Discussion The three critical success factors owing to action research are presented. It provided a politically safe approach to service, policy and legislative change, ensured collaboration permeated the endeavour and helped to shift the focus from a technical to an emancipatory approach to action research. Conclusion Action research was key to the success achieved by the participants in changing clinical practice, service delivery and the Victorian Drugs Poisons and Controlled Substances Act (1981) to authorise registered nurses to supply medicines. Implications for practice This paper offers an approach that nurses in practice, management and government can take to drive changes at practice, service and legislative levels in advanced nursing practice. Keywords Action research, change, emergency care, legislative change, rural nursing,
To address the problem, the Victorian government funded the ‘collaborative practice model pilot’. This aimed to engage a multidisciplinary group of health professionals from four rural health services and one bush nursing centre to develop and test an alternative model of service provision for emergency care that would overcome the reliance on medical practitioners. A bush nursing centre is a community based service, usually staffed only by registered nurses who manage an array of patient conditions, providing the first response to some local emergencies; such © RCN PUBLISHING / NURSE RESEARCHER
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Service redesign centres do not have beds. The alternative model was based on Queensland Health’s rural and isolated practice endorsed registered nursing (RIPRN) model (Timmings 2006). RIPRN is an advanced nursing practice model in which registered nurses are enabled through further education and legislation to operate relatively autonomously using a collaborative practice framework (Queensland Government and the Royal Flying Doctor Service 2011). The primary aim of the pilot was to enable nurses to practise more autonomously, decreasing the need to call in local doctors to attend to emergency presentations. Using an action research approach, the pilot achieved its outcome of reducing the calls on local GPs and increasing the nursing participants’ autonomy, and therefore the number of patients seen only by nurses. The strategies used to achieve this end-point included advanced nursing training, organisational policy change and legislative amendments. The details of the pilot are reported elsewhere (Sullivan 2013), but this paper presents the action research approach used. It will briefly discuss the action research methodology underpinning the pilot, and then describe the various methods used to engage participants in understanding their situations, and developing and reviewing actions to create a more sustainable emergency care service model. Finally, the success factors of the action research approach to changing practice, health services and legislation are discussed.
Action research methodology Action research is a methodology designed to engage people meaningfully in change processes that affect them and to empower them to shape the changes that are made (Roberts and Taylor 2002). The participants define the problem and methods of data collection and interpret the results (Holter and Schwartz-Barcott 1993, Popay et al 1998, Kenny and Duckett 2004). Action research involves repeated cycles of assessing the situation, planning and implementing actions to change it, and reassessing the situation to measure the effects of the actions and refine subsequent actions (Stringer 1999) (Figure 1). This cyclic process can permeate every aspect of the research, from overall research design to the critical self-reflection undertaken by the researcher and participants with respect to their individual actions and interactions (Dick 2005b).
Action research methods for the pilot Ethics clearance The Victoria Department of Health’s human research ethics committee and the University of Queensland behaviour and social sciences ethical review committee granted ethics approval. © RCN PUBLISHING / NURSE RESEARCHER
08-13 NR Nov 2013.indd 9
Figure 1 The cyclic process of action research Action planning Considering alternative courses of action for solving the problem Diagnosing Identifying or defining a problem
Specifying learning Identifying general findings
System and practice development through action research
Action taking Selecting a course of action
Evaluating Studying the consequences of action
(Adapted from Susman and Evered 1978)
Participant selection strategy The pilot used convenience sampling at two levels to select participants: health services and staff including nurses, doctors and pharmacists working in the selected health services and in a bush nursing centre. Pilot sites The five pilot sites were selected purposively (Roberts and Taylor 2002) on the basis that they were struggling to maintain a medical presence in emergency care, had viable access to medical practitioners (ranging from doctors on staff to fortnightly visits from a local GP) and had viable emergency patient volumes ranging from 867 to more than 5,000 patient presentations a year. The health services also varied in terms of whether their nurses were designated to the emergency department (ED) in accordance with the then Victorian public nurses’ enterprise bargaining agreement. Health services with more than 5,000 patients a year were required to have nursing staff designated to their EDs, whereas EDs with fewer than 5,000 patient presentations relied on the ward nurses attending to patients who presented in the emergency department (Table 1, page 10). Participant selection Action research participant selection aims to include those who can provide ‘the most relevant and richest forms of information’ (Popay et al 1998, Kenny and Duckett 2004), as well as the broadest range of perspectives possible (Dick 2005a). This aim was achieved by targeting health November 2013 | Volume 21 | Number 2
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
9
24/10/2013 14:13
Nurse Researcher service executives and managers, doctors, nurses and pharmacists in participating health services that had direct interests in the provision of emergency care. The project plan In this study, there were four action cycles, which started with the engagement of the health service executives in late 2006. The remaining participants were engaged between March and May 2007. Figure 2 illustrates the planning forum and action learning sets (ALSs), data collection and analysis across the project timeline. An ALS is an integral element of action research that provides the forum for study participants to deliberate on and develop strategies to change their current situation. The action cycles are bounded by the action planning forum and subsequent ALSs. All participants were invited to attend a two-day planning forum at the start of the project and the three one-day ALSs held at regular intervals during the project. Planning forum and ALSs The research facilitator aided the planning forum and the ALSs. The aims of the planning forum and the ALSs are listed in Boxes 1 and 2, page 12 At the end of the planning forum and the ALSs, participants completed evaluation surveys. The detailed results are reported elsewhere (Sullivan 2013), but participants agreed that the aims of the planning forum and ALSs were achieved. Additionally, the participants generated the agenda of each ALS, ensuring maximum participation and ownership of the process. Data collection and analysis Action research produces knowledge relating to the diagnosis of the situation Table 1
before and after actions, as well as the actions taken and their effects on their situation (Lewin 1947). Data collection provides participants with profiles of their situations so they can identify what changes are needed and monitor if the desired changes occur. To this end, participants determined the approach taken to data collection, and were involved in developing the data collection tools and interpreting the data in the ALSs. Figure 2 shows the points in the study where data were collected. Quantitative data collection and analysis The quantitative data collected and analysed in this study were drawn from the patient records. To collect data from each participating site, a minimum dataset (MDS) was developed in Excel. The data fields and definitions were based on the outcomes the participants aimed to achieve in this project and on the ease of data capture by clinicians and the existing data systems. The MDS was then piloted in three of the study sites and amended according to the feedback provided. Data were collected on emergency patients for four to six weeks at the start of the project and for the same time before the final ALS. The first data collection and analysis were presented as part of the pilot sites’ organisational diagnostic profiles using frequencies (for example, ‘proportion of patients seen by nurse’) and cross-tabulations (for example, ‘proportion of patients seen by a nurse by triage category’). The second data collection before the third ALS enabled comparative analysis to measure the effect of actions on key indicators, such as the ‘number of presentations managed by nurses without doctors’.
Participants in sample design A Designated nurses > 5,000 presentations
B No of designated nurses with more than 2,500 presentations
C No of designated nurses with fewer than 2,500 presentations
D No of designated nurses with fewer than 1,000 presentations
E Remote area nurse (867 presentations)
Total no of participants
Registered nurse, division one
4
3
4
2
1
14
Visiting medical officer
0
1
1
1
0
3
Pharmacist
1
1
1
0
0
3
CEO
1
1
1
1
0
4
Director of nursing
1
1
1
1
0
4
Paramedic
0
0
0
1
0
1
Total
7
7
8
6
1
29
Pilot site
10 November 2013 | Volume 21 | Number 2
08-13 NR Nov 2013.indd 10
© RCN PUBLISHING / NURSE RESEARCHER
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Service redesign Figure 2 Mar 2007 Engage
Study action cycles, data collection and analysis Apr
May
Jun
Jul
Aug
Sep
Diagnostic phase
Pre-data collection ■■ Emergency presentations profile ■■ Semi-structured interviews Data analysis ■■ Service diagnostic ■■ Profile – service activity, issues and potential solutions Qualitative data collection and analysis Qualitative data were gathered from semi-structured interviews, the planning forum, ALSs and focus groups. Convergent interviewing To simultaneously gather and analyse data, the research facilitator engaged participants in semi-structured interviews of approximately one hour’s length using the convergent interviewing technique described by Dick (2005b). Convergent interviewing is said to suit action research because it allows the interview ‘process to be driven by the (participants) and the data they provide’ (Dick 2005b). Convergent interviewing also enables the researcher to interpret the data while interviewing participants, and ensures that the interpretations and theory that evolve from the interviews reflects the participants’ reality (McDowell et al 1996). The results of the data analysis are continually validated and challenged by subsequent interviews. Convergent interviewing also follows the action research cycles by allowing interviewers to adjust their interviewing style, the questions and even the participants following reflection about previous interviews (Dick 2005b). For example, a paramedic was invited to participate in the project after participants identified her role as important in the emergency service. Each participant was asked about his or her background, role and perspective on how the emergency service operated, identifying areas of
08-13 NR Nov 2013.indd 11
Nov
Dec
Jan 2008
Feb
Implementation phase
Planning forum
© RCN PUBLISHING / NURSE RESEARCHER
Oct
Action learning set
Mar
Apr
Withdraw Action learning set
Data collection ■■ Emergency presentations profile Data analysis ■■ Comparative analysis of process/ output indicators
Action learning set
Data collection Focus groups Data analysis ■■ Nominal group technique
strength and opportunities for improvement. This gave the participants an opportunity to present their views on the subject. The researcher identified issues that emerged following each interview. Dick (2005b) suggested that where there is agreement between interviewees on issues, a probe question for the following interviews should be devised ‘to find the exception (Dick 2005b). Questioning became more specific in subsequent interviews, focusing on the themes that emerged in previous interviews in an attempt to understand the different perspectives (Dick 2005b). Data and analysis from the planning forum and ALSs Participants generated and analysed a significant amount of data in the ALS. ALSs represent ‘communities of inquiry’ (Dewar and Sharp 2006) or ‘self-critical communities’ (McTaggart 1991). The data gathered from these forums were: project outcomes agreed by the participants; action plans developed and refined in these sessions; progress assessments of strategies; factors that enabled or obstructed progress; and participant evaluations of the planning forum and ALS. Focus groups – nominal group technique A focus group was held at each of the four health services and the remote area nurse was re-interviewed before the last ALS. The purpose of these final focus groups and the interview was to identify emerging issues and areas for November 2013 | Volume 21 | Number 2 11
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Nurse Researcher Box 1 Planning forum aims
Action research success factors
■■ To provide a detailed understanding of the pilot, how it fits into the state picture and what commitment is required for it to succeed. ■■ To provide an opportunity to network and form working relationships in and across pilot sites. ■■ To provide opportunities to learn from other participants and challenge assumptions that may get in the way of improvement. ■■ To provide an understanding of the boundaries to improving practice, including the medicolegal boundaries and enablers, as well as the practice boundaries relating to the nurses’ training programme. ■■ To develop achievable plans of action to improve the emergency care service and practice. ■■ To identify the synergies and actions best pursued across the pilot sites or by the government.
While action research has been used to help bring about change in education (Wang et al 2010), health care (Elsey and Lathlean 2006), geriatric care (Lindeman et al 2003, Dewar and Sharp 2006), and child and family support services (Crane and Richardson 2000), this study is probably the only published account of action research being used as a collaborative strategy by government to drive service delivery and legislative change. We believe that the success of this study in achieving change in practice, service and legislation can be attributed to the action research approach, which: ■■ Provides a politically safe approach to service, policy and legislative change. ■■ Ensures collaboration permeates the endeavour. ■■ Shifts the focus from a technical to an emancipatory approach to action research. Each of these success factors will now be discussed.
Box 2 Action learning set aims ■■ To provide opportunities for critical appraisal and discussion about issues that get in the way of delivering effective emergency care, tapping into the wide variety of perspectives and experiences present at the ALSs. ■■ To develop strategies, processes and attitudes that enable effective change management. ■■ To create opportunities to network and form working relationships in and across pilot sites. ■■ To provide opportunities to learn from others and challenge assumptions that may get in the way of improvement. ■■ To review achievements so far, and what blocked and enabled progress. ■■ To refine action plans and decide on the next steps for the pilot. improvement, strategies to address these, and processes to sustain this continuous improvement process. Nominal group technique (NGT), developed by Delbecq and Van de Ven (1971), was used with the focus groups. NGT is an approach to solving complex problems that involve a wide range of stakeholders with different perspectives of the solution (Van de Ven and Delbecq 1974). NGT ensures that the greatest number of ideas is generated, by creating a safe environment in which individuals can share their ideas and thoughts. Initially, group members work alongside each other individually, considering and generating their ideas, rather than interacting. These are then ranked and discussed by the group to reach a collective decision (Delbecq and Van de Ven 1971, Van de Ven and Delbecq 1974). 12 November 2013 | Volume 21 | Number 2
08-13 NR Nov 2013.indd 12
Provides a politically safe approach to service system and legislative change Previously, attempts to advance nursing and increase nurses’ ability to practise more autonomously have been met with significant resistance from the medical profession (Willis 1989, Ball and Cox 2004, Bryant-Lukosius et al 2004, Lindeke and Jukkala 2005). Given that this project focused on potentially contested ground between medicine and nursing, action research was considered the most politically safe approach because it actively engaged people from medicine and nursing, tested their tolerance for the new model, and actively involved doctors in implementing the new model. Furthermore, the project brought about change incrementally, which assured participants that changes could be reversed if the change was ineffective or caused stakeholders to have negative reactions. The evidence and support from stakeholders generated by the study contributed to a successful case for changing the Victorian Drugs, Poisons and Controlled Substances Act (1981), enabling nurses to supply medicines without a doctor’s order under certain circumstances. Ensures collaboration permeates the endeavour True to its name, the ‘collaborative practice model pilot’ used the action research principles and processes to establish genuine collaboration between participants at every stage of the project, from design to delivery. This meant that the forums and ALSs encouraged collaboration between members of the health disciplines, clinicians and managers, clinicians and government, and health services. There is an abundance of evidence that stakeholder engagement and collaboration are key to successful change (for example, Kotter and Schlesinger 1979). © RCN PUBLISHING / NURSE RESEARCHER
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Service redesign Shifts the focus from a technical to emancipatory approach to action research The degree to which participants ‘participate’ is said to depend on the facilitator’s orientation to the research or practice development, and whether it is technical or emancipatory (Grundy 1982, Manley and McCormack 2003). The main difference between technical and emancipatory orientations to facilitation is that the former places the facilitator as the expert authority on the ‘technique’, providing participants with a topic to research, direction and expertise. The facilitator who adopts an emancipatory orientation creates the conditions for open, critical and reflective discussion by all participants, and the development of ideas and motivation for action. The participants retain the power and responsibility for change (Grundy 1982, Manley and McCormack 2003). It was found in this study that a facilitator’s orientation can shift over the course of a project as participants engage and develop their understanding of the project methods and aims. This study started as technical action research and progressed to emancipatory action research. Initially, the research was driven by the researchers’ critical intent to create a ‘negotiated division of labour’ between nurses and doctors, as defined by Freidson (1976) and interpreted
by Sullivan et al (2008). However, to test this theory, the health services and clinicians involved had to be willing and enabled to question their practice and systems. The research facilitator provided the ‘technical expertise’ regarding the public policy process, the rural and remote nursing model that was piloted, and the action research methodology. From the point that the participants were recruited, the project became emancipatory action research, as it was the participants’ critical intent as much as the researchers’ that drove the project.
Conclusion The Victoria Department of Health funded an action research project to engage a selection of rural health services and a bush nursing centre in developing and testing a new, more sustainable approach to the provision of emergency care. Action research was central to the success achieved by the participants in changing clinical practice, service delivery and the Drugs Poisons and Controlled Substances Act (1981). The factors that make action research ideally suited to driving service, policy and legislative reform are its incremental, cyclic nature, the engagement of stakeholders and the empowerment of participants to drive change.
Online archive For related information, visit our online archive and search using the keywords
Conflict of interest None declared Acknowledgements The authors would like to thank the Victoria Department of Health, the former Victorian Ministerial Health Services Management Innovation Council and the Australian Royal College of Nursing, which funded the project, as well as the research participants and the Victoria Department of Health colleagues for their support
References Ball C, Cox CL (2004) Part two: the core components of legitimate influence and the conditions that constrain or facilitate advanced nursing practice in adult critical care. International Journal of Nursing Practice. 10, 1, 10-20. Bryant-Lukosius D, Dicenso A, Browne G et al (2004) Advanced practice nursing roles: development, implementation and evaluation. Journal of Advanced Nursing. 48, 5, 519-529. Crane P, Richardson L (2000) Reconnect Action Research Kit. tinyurl.com/mycghaa (Last accessed: October 10 2013.) Delbecq AL, Van de Ven AH (1971) A group process model for problem identification and program planning. The Journal of Applied Behavioral Science. 7, 4, 466-492. Dewar B, Sharp C (2006) Using evidence: how action learning can support individual and organisational learning through action research. Educational Action Research. 14, 2, 219-237. Dick B (2005a) Areol: Action Research And Evaluation On Line – Paper 38. www.aral.com.au/ DLitt/DLitt_P38areol.pdf (Last accessed: August 19 2013.) Dick B (2005b) Convergent Interviewing. www.aral.com.au/areol/areol-session08.html (Last accessed: October 22 2013.) Elsey H, Lathlean J (2006) Using action research to stimulate organisational change within health services: experiences from two community-based
studies. Educational Action Research. 14, 2, 171-186. Fowles S (2006a) A&E crisis meeting. Portland Observer. May 15. Fowles S (2006b) Mayor takes A&E crisis to Canberra. Portland Observer. July 23. Fowles S (2006c) No A&E doctor tomorrow. Portland Observer. May 31. Freidson E (1976) The division of labour as social interaction. Social Problems. February 23, 304-313. Grundy S (1982) Three modes of action research. Curriculum Perspectives. 2, 3, 23-34. Holter IM, Schwartz-Barcott D (1993) Action research: what is it? How has it been used and how can it be used in nursing? Journal of Advanced Nursing. 18, 2, 298-304. Kenny A, Duckett S (2004) A question of place: medical power in rural Australia. Social Science and Medicine. 58, 6, 1059-1073. Kotter JP, Schlesinger LA (1979) Choosing strategies for change. Harvard Business Review. 57, 2, 106-114. Lewin K (1947) Frontiers in group dynamics. II: Channels of group life; social planning and action research. Human Relations. 1, 2, 143-153. Lindeke LL, Jukkala AJ (2005) Rural NP practice barriers and strategies for success: one state’s story. American Journal for Nurse Practitioners. 9, 3, 11-18.
© RCN PUBLISHING / NURSE RESEARCHER
08-13 NR Nov 2013.indd 13
Lindeman MA, Black K, Smith R et al (2003) Changing practice in residential aged care using participatory methods. Education for Health. 16, 1, 22-31. McDowell C, Hine D, Bakker P (1996) Convergent Interviewing: An Application To Small Business Research. Small Business Advancement National Center, Conway AR. McTaggart R (1991) Principles for participatory action research. Adult Education Quarterly. 41, 3, 168-187. Manley K, McCormack B (2003) Practice development: purpose, methodology, facilitation and evaluation. Nursing in Critical Care. 8, 1, 22-29. Popay J, Rogers A, Williams G (1998) Rationale and standards for the systematic review of qualitative literature in health services research. Qualitative Health Research. 8, 3, 341-351. Queensland Government and the Royal Flying Doctor Service (2011) Primary Clinical Care Manual. Seventh edition. Cairns Qld. Roberts K, Taylor B (2002) Nursing Research Processes: An Australian Perspective. Second edition. Thomson Nelson, Southbank Vic. Rural Workforce Agency, Victoria (RWAV) (2006) White Paper – The Viability of Rural And Regional Communities. Resolving Victoria’s Rural Medical Workforce Crisis. RWAV, Melbourne Vic. Schmeiszl R (2006) Phillip Island medical crisis. South Gippsland Sentinel Times. January 10.
Scopelianos S (2006) Hospital screws tighten. Warrnambool Standard. May 13 2006. Stringer ET (1999) Action Research. First edition. Sage Publications, London. Sullivan E (2013) All Hands on Deck: An Autonomous Collaborative Practice Model to Sustain Emergency Care in Rural Victoria. Monash University Faculty of Medicine, Nursing and Health Services, Nursing and Midwifery, Melbourne Vic. Sullivan E, Francis K, Hegney D (2008) Review of small rural health services in Victoria: How does the nursing-medical division of labour affect access to emergency care? Journal of Clinical Nursing. 17, 12, 1543-1552. Susman GI, Evered RD (1978) An assessment of the scientific merits of action research. Administrative Science Quarterly. 23, 4, 582-603. Timmings RW (2006) Rural and isolated practice registered nurse (RIPRN) – Emergency nurses of the Queensland ‘bush’. Australasian Emergency Nursing Journal. 9, 1, 29-34. Van de Ven AH, Delbecq AL (1974) The effectiveness of nominal, Delphi and interacting group decision making processes. Academy of Management Journal. 17, 4, 605-621. Wang Y, Kretschmer RE, Hartman MC (2010) Teacher as researcher: Theory into practice. American Annals of the Deaf. 155, 2, 105-109. Willis E (1989) Medical Dominance. Revised edition. Allen and Unwin Australia, St Leonards.
November 2013 | Volume 21 | Number 2 13
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
An action research approach to practice, service and legislative change Cite this article as: Sullivan E, Hegney D, Francis K (2013) An action research approach to practice, service and legislative change. Nurse Researcher. 21, 2, 8-13. Date of submission: April 30 2012. Date of acceptance: April 12 2013. Correspondence to Elise Sullivan [email protected] Elise Sullivan RN, master of public policy, is a PhD student at the Department of Nursing and Midwifery, Monash University, Melbourne, Australia and a director of DPAR Consulting, Myrniong, Victoria, Australia Desley G Hegney RN, PhD is a professor of nursing at the School of Nursing and Midwifery, Sir Charles Gairdner Hospital, Perth, Australia and a programme co-ordinator, vulnerable populations at the Curtin Health Innovation Research Institute (CHIRI), Curtin University, Perth, Australia Karen Francis RN, PhD is a professor and head of the School of Nursing Midwifery and Indigenous Health at Charles Sturt University, New South Wales, Australia Peer review This article has been subject to double-blind review and checked using antiplagiarism software Author guidelines http://nr.rcnpublishing.com
8
Abstract Aims To describe the action research approach taken to engage a multidisciplinary group of health professionals and managers from five rural health services with government officers in redesigning their emergency care services and informing legislative change. Background The diminishing size of the medical workforce across rural Victoria in Australia captured the Victorian state government’s attention when this threatened the sustainability of emergency care services in rural and remote hospitals in 2006. The government funded the collaborative practice model pilot between 2006 and 2008 to develop and test an alternative model of emergency care service in which nurses practised at a more advanced and autonomous level. Data sources Data were sourced from a combination of interviews, focus groups and patient records. Review methods Qualitative data were analysed using convergent interview and thematic analysis. Quantitative data were analysed using frequencies and cross tabulations.
Introduction THE GAP between demand for and supply of health professionals is growing at an accelerated rate in Australia, as the population ages and the available workforce contracts (Rural Workforce Agency, Victoria 2006). Rural hospitals across the state of Victoria have experienced this trend most acutely, struggling to maintain 24-hour emergency care services as the size of the medical workforce has reduced (Fowles 2006a, 2006b, 2006c, Schmeiszl 2006, Scopelianos 2006).
November 2013 | Volume 21 | Number 2
08-13 NR Nov 2013.indd 8
Discussion The three critical success factors owing to action research are presented. It provided a politically safe approach to service, policy and legislative change, ensured collaboration permeated the endeavour and helped to shift the focus from a technical to an emancipatory approach to action research. Conclusion Action research was key to the success achieved by the participants in changing clinical practice, service delivery and the Victorian Drugs Poisons and Controlled Substances Act (1981) to authorise registered nurses to supply medicines. Implications for practice This paper offers an approach that nurses in practice, management and government can take to drive changes at practice, service and legislative levels in advanced nursing practice. Keywords Action research, change, emergency care, legislative change, rural nursing,
To address the problem, the Victorian government funded the ‘collaborative practice model pilot’. This aimed to engage a multidisciplinary group of health professionals from four rural health services and one bush nursing centre to develop and test an alternative model of service provision for emergency care that would overcome the reliance on medical practitioners. A bush nursing centre is a community based service, usually staffed only by registered nurses who manage an array of patient conditions, providing the first response to some local emergencies; such © RCN PUBLISHING / NURSE RESEARCHER
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Service redesign centres do not have beds. The alternative model was based on Queensland Health’s rural and isolated practice endorsed registered nursing (RIPRN) model (Timmings 2006). RIPRN is an advanced nursing practice model in which registered nurses are enabled through further education and legislation to operate relatively autonomously using a collaborative practice framework (Queensland Government and the Royal Flying Doctor Service 2011). The primary aim of the pilot was to enable nurses to practise more autonomously, decreasing the need to call in local doctors to attend to emergency presentations. Using an action research approach, the pilot achieved its outcome of reducing the calls on local GPs and increasing the nursing participants’ autonomy, and therefore the number of patients seen only by nurses. The strategies used to achieve this end-point included advanced nursing training, organisational policy change and legislative amendments. The details of the pilot are reported elsewhere (Sullivan 2013), but this paper presents the action research approach used. It will briefly discuss the action research methodology underpinning the pilot, and then describe the various methods used to engage participants in understanding their situations, and developing and reviewing actions to create a more sustainable emergency care service model. Finally, the success factors of the action research approach to changing practice, health services and legislation are discussed.
Action research methodology Action research is a methodology designed to engage people meaningfully in change processes that affect them and to empower them to shape the changes that are made (Roberts and Taylor 2002). The participants define the problem and methods of data collection and interpret the results (Holter and Schwartz-Barcott 1993, Popay et al 1998, Kenny and Duckett 2004). Action research involves repeated cycles of assessing the situation, planning and implementing actions to change it, and reassessing the situation to measure the effects of the actions and refine subsequent actions (Stringer 1999) (Figure 1). This cyclic process can permeate every aspect of the research, from overall research design to the critical self-reflection undertaken by the researcher and participants with respect to their individual actions and interactions (Dick 2005b).
Action research methods for the pilot Ethics clearance The Victoria Department of Health’s human research ethics committee and the University of Queensland behaviour and social sciences ethical review committee granted ethics approval. © RCN PUBLISHING / NURSE RESEARCHER
08-13 NR Nov 2013.indd 9
Figure 1 The cyclic process of action research Action planning Considering alternative courses of action for solving the problem Diagnosing Identifying or defining a problem
Specifying learning Identifying general findings
System and practice development through action research
Action taking Selecting a course of action
Evaluating Studying the consequences of action
(Adapted from Susman and Evered 1978)
Participant selection strategy The pilot used convenience sampling at two levels to select participants: health services and staff including nurses, doctors and pharmacists working in the selected health services and in a bush nursing centre. Pilot sites The five pilot sites were selected purposively (Roberts and Taylor 2002) on the basis that they were struggling to maintain a medical presence in emergency care, had viable access to medical practitioners (ranging from doctors on staff to fortnightly visits from a local GP) and had viable emergency patient volumes ranging from 867 to more than 5,000 patient presentations a year. The health services also varied in terms of whether their nurses were designated to the emergency department (ED) in accordance with the then Victorian public nurses’ enterprise bargaining agreement. Health services with more than 5,000 patients a year were required to have nursing staff designated to their EDs, whereas EDs with fewer than 5,000 patient presentations relied on the ward nurses attending to patients who presented in the emergency department (Table 1, page 10). Participant selection Action research participant selection aims to include those who can provide ‘the most relevant and richest forms of information’ (Popay et al 1998, Kenny and Duckett 2004), as well as the broadest range of perspectives possible (Dick 2005a). This aim was achieved by targeting health November 2013 | Volume 21 | Number 2
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
9
24/10/2013 14:13
Nurse Researcher service executives and managers, doctors, nurses and pharmacists in participating health services that had direct interests in the provision of emergency care. The project plan In this study, there were four action cycles, which started with the engagement of the health service executives in late 2006. The remaining participants were engaged between March and May 2007. Figure 2 illustrates the planning forum and action learning sets (ALSs), data collection and analysis across the project timeline. An ALS is an integral element of action research that provides the forum for study participants to deliberate on and develop strategies to change their current situation. The action cycles are bounded by the action planning forum and subsequent ALSs. All participants were invited to attend a two-day planning forum at the start of the project and the three one-day ALSs held at regular intervals during the project. Planning forum and ALSs The research facilitator aided the planning forum and the ALSs. The aims of the planning forum and the ALSs are listed in Boxes 1 and 2, page 12 At the end of the planning forum and the ALSs, participants completed evaluation surveys. The detailed results are reported elsewhere (Sullivan 2013), but participants agreed that the aims of the planning forum and ALSs were achieved. Additionally, the participants generated the agenda of each ALS, ensuring maximum participation and ownership of the process. Data collection and analysis Action research produces knowledge relating to the diagnosis of the situation Table 1
before and after actions, as well as the actions taken and their effects on their situation (Lewin 1947). Data collection provides participants with profiles of their situations so they can identify what changes are needed and monitor if the desired changes occur. To this end, participants determined the approach taken to data collection, and were involved in developing the data collection tools and interpreting the data in the ALSs. Figure 2 shows the points in the study where data were collected. Quantitative data collection and analysis The quantitative data collected and analysed in this study were drawn from the patient records. To collect data from each participating site, a minimum dataset (MDS) was developed in Excel. The data fields and definitions were based on the outcomes the participants aimed to achieve in this project and on the ease of data capture by clinicians and the existing data systems. The MDS was then piloted in three of the study sites and amended according to the feedback provided. Data were collected on emergency patients for four to six weeks at the start of the project and for the same time before the final ALS. The first data collection and analysis were presented as part of the pilot sites’ organisational diagnostic profiles using frequencies (for example, ‘proportion of patients seen by nurse’) and cross-tabulations (for example, ‘proportion of patients seen by a nurse by triage category’). The second data collection before the third ALS enabled comparative analysis to measure the effect of actions on key indicators, such as the ‘number of presentations managed by nurses without doctors’.
Participants in sample design A Designated nurses > 5,000 presentations
B No of designated nurses with more than 2,500 presentations
C No of designated nurses with fewer than 2,500 presentations
D No of designated nurses with fewer than 1,000 presentations
E Remote area nurse (867 presentations)
Total no of participants
Registered nurse, division one
4
3
4
2
1
14
Visiting medical officer
0
1
1
1
0
3
Pharmacist
1
1
1
0
0
3
CEO
1
1
1
1
0
4
Director of nursing
1
1
1
1
0
4
Paramedic
0
0
0
1
0
1
Total
7
7
8
6
1
29
Pilot site
10 November 2013 | Volume 21 | Number 2
08-13 NR Nov 2013.indd 10
© RCN PUBLISHING / NURSE RESEARCHER
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Service redesign Figure 2 Mar 2007 Engage
Study action cycles, data collection and analysis Apr
May
Jun
Jul
Aug
Sep
Diagnostic phase
Pre-data collection ■■ Emergency presentations profile ■■ Semi-structured interviews Data analysis ■■ Service diagnostic ■■ Profile – service activity, issues and potential solutions Qualitative data collection and analysis Qualitative data were gathered from semi-structured interviews, the planning forum, ALSs and focus groups. Convergent interviewing To simultaneously gather and analyse data, the research facilitator engaged participants in semi-structured interviews of approximately one hour’s length using the convergent interviewing technique described by Dick (2005b). Convergent interviewing is said to suit action research because it allows the interview ‘process to be driven by the (participants) and the data they provide’ (Dick 2005b). Convergent interviewing also enables the researcher to interpret the data while interviewing participants, and ensures that the interpretations and theory that evolve from the interviews reflects the participants’ reality (McDowell et al 1996). The results of the data analysis are continually validated and challenged by subsequent interviews. Convergent interviewing also follows the action research cycles by allowing interviewers to adjust their interviewing style, the questions and even the participants following reflection about previous interviews (Dick 2005b). For example, a paramedic was invited to participate in the project after participants identified her role as important in the emergency service. Each participant was asked about his or her background, role and perspective on how the emergency service operated, identifying areas of
08-13 NR Nov 2013.indd 11
Nov
Dec
Jan 2008
Feb
Implementation phase
Planning forum
© RCN PUBLISHING / NURSE RESEARCHER
Oct
Action learning set
Mar
Apr
Withdraw Action learning set
Data collection ■■ Emergency presentations profile Data analysis ■■ Comparative analysis of process/ output indicators
Action learning set
Data collection Focus groups Data analysis ■■ Nominal group technique
strength and opportunities for improvement. This gave the participants an opportunity to present their views on the subject. The researcher identified issues that emerged following each interview. Dick (2005b) suggested that where there is agreement between interviewees on issues, a probe question for the following interviews should be devised ‘to find the exception (Dick 2005b). Questioning became more specific in subsequent interviews, focusing on the themes that emerged in previous interviews in an attempt to understand the different perspectives (Dick 2005b). Data and analysis from the planning forum and ALSs Participants generated and analysed a significant amount of data in the ALS. ALSs represent ‘communities of inquiry’ (Dewar and Sharp 2006) or ‘self-critical communities’ (McTaggart 1991). The data gathered from these forums were: project outcomes agreed by the participants; action plans developed and refined in these sessions; progress assessments of strategies; factors that enabled or obstructed progress; and participant evaluations of the planning forum and ALS. Focus groups – nominal group technique A focus group was held at each of the four health services and the remote area nurse was re-interviewed before the last ALS. The purpose of these final focus groups and the interview was to identify emerging issues and areas for November 2013 | Volume 21 | Number 2 11
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Nurse Researcher Box 1 Planning forum aims
Action research success factors
■■ To provide a detailed understanding of the pilot, how it fits into the state picture and what commitment is required for it to succeed. ■■ To provide an opportunity to network and form working relationships in and across pilot sites. ■■ To provide opportunities to learn from other participants and challenge assumptions that may get in the way of improvement. ■■ To provide an understanding of the boundaries to improving practice, including the medicolegal boundaries and enablers, as well as the practice boundaries relating to the nurses’ training programme. ■■ To develop achievable plans of action to improve the emergency care service and practice. ■■ To identify the synergies and actions best pursued across the pilot sites or by the government.
While action research has been used to help bring about change in education (Wang et al 2010), health care (Elsey and Lathlean 2006), geriatric care (Lindeman et al 2003, Dewar and Sharp 2006), and child and family support services (Crane and Richardson 2000), this study is probably the only published account of action research being used as a collaborative strategy by government to drive service delivery and legislative change. We believe that the success of this study in achieving change in practice, service and legislation can be attributed to the action research approach, which: ■■ Provides a politically safe approach to service, policy and legislative change. ■■ Ensures collaboration permeates the endeavour. ■■ Shifts the focus from a technical to an emancipatory approach to action research. Each of these success factors will now be discussed.
Box 2 Action learning set aims ■■ To provide opportunities for critical appraisal and discussion about issues that get in the way of delivering effective emergency care, tapping into the wide variety of perspectives and experiences present at the ALSs. ■■ To develop strategies, processes and attitudes that enable effective change management. ■■ To create opportunities to network and form working relationships in and across pilot sites. ■■ To provide opportunities to learn from others and challenge assumptions that may get in the way of improvement. ■■ To review achievements so far, and what blocked and enabled progress. ■■ To refine action plans and decide on the next steps for the pilot. improvement, strategies to address these, and processes to sustain this continuous improvement process. Nominal group technique (NGT), developed by Delbecq and Van de Ven (1971), was used with the focus groups. NGT is an approach to solving complex problems that involve a wide range of stakeholders with different perspectives of the solution (Van de Ven and Delbecq 1974). NGT ensures that the greatest number of ideas is generated, by creating a safe environment in which individuals can share their ideas and thoughts. Initially, group members work alongside each other individually, considering and generating their ideas, rather than interacting. These are then ranked and discussed by the group to reach a collective decision (Delbecq and Van de Ven 1971, Van de Ven and Delbecq 1974). 12 November 2013 | Volume 21 | Number 2
08-13 NR Nov 2013.indd 12
Provides a politically safe approach to service system and legislative change Previously, attempts to advance nursing and increase nurses’ ability to practise more autonomously have been met with significant resistance from the medical profession (Willis 1989, Ball and Cox 2004, Bryant-Lukosius et al 2004, Lindeke and Jukkala 2005). Given that this project focused on potentially contested ground between medicine and nursing, action research was considered the most politically safe approach because it actively engaged people from medicine and nursing, tested their tolerance for the new model, and actively involved doctors in implementing the new model. Furthermore, the project brought about change incrementally, which assured participants that changes could be reversed if the change was ineffective or caused stakeholders to have negative reactions. The evidence and support from stakeholders generated by the study contributed to a successful case for changing the Victorian Drugs, Poisons and Controlled Substances Act (1981), enabling nurses to supply medicines without a doctor’s order under certain circumstances. Ensures collaboration permeates the endeavour True to its name, the ‘collaborative practice model pilot’ used the action research principles and processes to establish genuine collaboration between participants at every stage of the project, from design to delivery. This meant that the forums and ALSs encouraged collaboration between members of the health disciplines, clinicians and managers, clinicians and government, and health services. There is an abundance of evidence that stakeholder engagement and collaboration are key to successful change (for example, Kotter and Schlesinger 1979). © RCN PUBLISHING / NURSE RESEARCHER
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
Service redesign Shifts the focus from a technical to emancipatory approach to action research The degree to which participants ‘participate’ is said to depend on the facilitator’s orientation to the research or practice development, and whether it is technical or emancipatory (Grundy 1982, Manley and McCormack 2003). The main difference between technical and emancipatory orientations to facilitation is that the former places the facilitator as the expert authority on the ‘technique’, providing participants with a topic to research, direction and expertise. The facilitator who adopts an emancipatory orientation creates the conditions for open, critical and reflective discussion by all participants, and the development of ideas and motivation for action. The participants retain the power and responsibility for change (Grundy 1982, Manley and McCormack 2003). It was found in this study that a facilitator’s orientation can shift over the course of a project as participants engage and develop their understanding of the project methods and aims. This study started as technical action research and progressed to emancipatory action research. Initially, the research was driven by the researchers’ critical intent to create a ‘negotiated division of labour’ between nurses and doctors, as defined by Freidson (1976) and interpreted
by Sullivan et al (2008). However, to test this theory, the health services and clinicians involved had to be willing and enabled to question their practice and systems. The research facilitator provided the ‘technical expertise’ regarding the public policy process, the rural and remote nursing model that was piloted, and the action research methodology. From the point that the participants were recruited, the project became emancipatory action research, as it was the participants’ critical intent as much as the researchers’ that drove the project.
Conclusion The Victoria Department of Health funded an action research project to engage a selection of rural health services and a bush nursing centre in developing and testing a new, more sustainable approach to the provision of emergency care. Action research was central to the success achieved by the participants in changing clinical practice, service delivery and the Drugs Poisons and Controlled Substances Act (1981). The factors that make action research ideally suited to driving service, policy and legislative reform are its incremental, cyclic nature, the engagement of stakeholders and the empowerment of participants to drive change.
Online archive For related information, visit our online archive and search using the keywords
Conflict of interest None declared Acknowledgements The authors would like to thank the Victoria Department of Health, the former Victorian Ministerial Health Services Management Innovation Council and the Australian Royal College of Nursing, which funded the project, as well as the research participants and the Victoria Department of Health colleagues for their support
References Ball C, Cox CL (2004) Part two: the core components of legitimate influence and the conditions that constrain or facilitate advanced nursing practice in adult critical care. International Journal of Nursing Practice. 10, 1, 10-20. Bryant-Lukosius D, Dicenso A, Browne G et al (2004) Advanced practice nursing roles: development, implementation and evaluation. Journal of Advanced Nursing. 48, 5, 519-529. Crane P, Richardson L (2000) Reconnect Action Research Kit. tinyurl.com/mycghaa (Last accessed: October 10 2013.) Delbecq AL, Van de Ven AH (1971) A group process model for problem identification and program planning. The Journal of Applied Behavioral Science. 7, 4, 466-492. Dewar B, Sharp C (2006) Using evidence: how action learning can support individual and organisational learning through action research. Educational Action Research. 14, 2, 219-237. Dick B (2005a) Areol: Action Research And Evaluation On Line – Paper 38. www.aral.com.au/ DLitt/DLitt_P38areol.pdf (Last accessed: August 19 2013.) Dick B (2005b) Convergent Interviewing. www.aral.com.au/areol/areol-session08.html (Last accessed: October 22 2013.) Elsey H, Lathlean J (2006) Using action research to stimulate organisational change within health services: experiences from two community-based
studies. Educational Action Research. 14, 2, 171-186. Fowles S (2006a) A&E crisis meeting. Portland Observer. May 15. Fowles S (2006b) Mayor takes A&E crisis to Canberra. Portland Observer. July 23. Fowles S (2006c) No A&E doctor tomorrow. Portland Observer. May 31. Freidson E (1976) The division of labour as social interaction. Social Problems. February 23, 304-313. Grundy S (1982) Three modes of action research. Curriculum Perspectives. 2, 3, 23-34. Holter IM, Schwartz-Barcott D (1993) Action research: what is it? How has it been used and how can it be used in nursing? Journal of Advanced Nursing. 18, 2, 298-304. Kenny A, Duckett S (2004) A question of place: medical power in rural Australia. Social Science and Medicine. 58, 6, 1059-1073. Kotter JP, Schlesinger LA (1979) Choosing strategies for change. Harvard Business Review. 57, 2, 106-114. Lewin K (1947) Frontiers in group dynamics. II: Channels of group life; social planning and action research. Human Relations. 1, 2, 143-153. Lindeke LL, Jukkala AJ (2005) Rural NP practice barriers and strategies for success: one state’s story. American Journal for Nurse Practitioners. 9, 3, 11-18.
© RCN PUBLISHING / NURSE RESEARCHER
08-13 NR Nov 2013.indd 13
Lindeman MA, Black K, Smith R et al (2003) Changing practice in residential aged care using participatory methods. Education for Health. 16, 1, 22-31. McDowell C, Hine D, Bakker P (1996) Convergent Interviewing: An Application To Small Business Research. Small Business Advancement National Center, Conway AR. McTaggart R (1991) Principles for participatory action research. Adult Education Quarterly. 41, 3, 168-187. Manley K, McCormack B (2003) Practice development: purpose, methodology, facilitation and evaluation. Nursing in Critical Care. 8, 1, 22-29. Popay J, Rogers A, Williams G (1998) Rationale and standards for the systematic review of qualitative literature in health services research. Qualitative Health Research. 8, 3, 341-351. Queensland Government and the Royal Flying Doctor Service (2011) Primary Clinical Care Manual. Seventh edition. Cairns Qld. Roberts K, Taylor B (2002) Nursing Research Processes: An Australian Perspective. Second edition. Thomson Nelson, Southbank Vic. Rural Workforce Agency, Victoria (RWAV) (2006) White Paper – The Viability of Rural And Regional Communities. Resolving Victoria’s Rural Medical Workforce Crisis. RWAV, Melbourne Vic. Schmeiszl R (2006) Phillip Island medical crisis. South Gippsland Sentinel Times. January 10.
Scopelianos S (2006) Hospital screws tighten. Warrnambool Standard. May 13 2006. Stringer ET (1999) Action Research. First edition. Sage Publications, London. Sullivan E (2013) All Hands on Deck: An Autonomous Collaborative Practice Model to Sustain Emergency Care in Rural Victoria. Monash University Faculty of Medicine, Nursing and Health Services, Nursing and Midwifery, Melbourne Vic. Sullivan E, Francis K, Hegney D (2008) Review of small rural health services in Victoria: How does the nursing-medical division of labour affect access to emergency care? Journal of Clinical Nursing. 17, 12, 1543-1552. Susman GI, Evered RD (1978) An assessment of the scientific merits of action research. Administrative Science Quarterly. 23, 4, 582-603. Timmings RW (2006) Rural and isolated practice registered nurse (RIPRN) – Emergency nurses of the Queensland ‘bush’. Australasian Emergency Nursing Journal. 9, 1, 29-34. Van de Ven AH, Delbecq AL (1974) The effectiveness of nominal, Delphi and interacting group decision making processes. Academy of Management Journal. 17, 4, 605-621. Wang Y, Kretschmer RE, Hartman MC (2010) Teacher as researcher: Theory into practice. American Annals of the Deaf. 155, 2, 105-109. Willis E (1989) Medical Dominance. Revised edition. Allen and Unwin Australia, St Leonards.
November 2013 | Volume 21 | Number 2 13
Downloaded from RCNi.com by ${individualUser.displayName} on Nov 29, 2015. For personal use only. No other uses without permission. Copyright © 2015 RCNi Ltd. All rights reserved.
24/10/2013 14:13
