Evidence-based practice (EBP) is now a well-established concept in the nursing literature. In this paper, Sacketts et al.'s (2000) definition of EBP is used, defined as an integration of best research evidence with clinical expertise and patient values in clinical decision making. Nevertheless, utilizing evidence-based practice is not straightforward, as witnessed by the citing of a gap between research and practice by various authors (Mullhall 2002; Halfens & Van Linge 2003; Hanberg & Brown 2006; Grol & Wensing 2006; Westert 2006).
The authors of this paper initially consider the factors that help and hinder the implementation of EBP, as well as what is known about the effectiveness of implementation strategies and interventions. Specific attention is then paid to action research as an implementation strategy that may be suitable for EBP. After a theoretical consideration of action research, a review is presented that answers the question: “What is known about the results of implementing EBP in nursing using action research?” The results of the various research projects are dealt with successively, classifying them, paying specific attention to the target group on which the implementation strategy was focused as well as the various strategies used. Finally, the results and limitations of the available research literature are discussed.
Factors that Help and Hinder Implementation
Many authors (Funk et al. 1991a; Kitson et al. 1998; Corrigan et al. 2001; Grol & Wensing 2001, 2006; Fleuren et al. 2002; Rycroft-Malone et al. 2002; Greenhalgh et al. 2004; van Linge 2005; Gerrish et al. 2007; Thompson et al. 2007) have considered the question of why it is so difficult for health care professionals to use or adopt research findings. The Barriers to Research Utilization Questionnaire (BRUQ) developed by Funk et al. (1991b) has often been used to identify the factors that are considered barriers to research implementation in nursing practice (Gerrish et al. 2007). The authors identified 29 barriers and clustered them around four factors, which nurses perceive as obstacles to research utilisation in practice. These four factors are:
- 1Characteristics of the adopter: the nurses’ research values, skills and awareness.
- 2Characteristics of the organisation: setting, barriers and limitations.
- 3Characteristics of the innovation: qualities of the research.
- 4Characteristics of the communication: presentation and accessibility of the research (Funk et al. 1991b).
The importance of the organisational context is also emphasized in the Promoting Action on Research Implementation in Health Services (PARIHS) framework. This framework also describes factors that help the implementation of EBP. Developed in 1998 (Kitson et al. 1998) and later refined (McCormack et al. 2002; Rycroft-Malone et al. 2002; Rycroft-Malone et al. 2004) the framework represents the most successful implementation of evidence occuring when:
- • evidence is scientific and matches professional consensus and patients’ preferences;
- • the context has features of learning organisations, with transformational leaders and appropriate monitoring and feedback mechanisms; and
- • there is an input from skilled facilitators who adapt their facilitation strategies based on the availability of resources, the context's culture and values, and the style of leadership and evaluation activities.
Gerrish et al. (2007) developed and tested a tool, the Developing Evidence-Based Practice Questionnaire, which they used for investigating factors associated with EBP among nurses in England. This tool consists of 10 identifiable factors that help and hinder the implementation of EBP, of which 8 demonstrated high reliability (Cronbach's alpha ≥ 0.7). One of the differences between this tool and the BRUQ (Funk et al. 1991b) is a broader interpretation of the term ‘evidence’ to include documents such as clinical protocols and guidelines, in addition to research evidence. This broader interpretation of evidence can also be found in the PARIHS framework (Rycroft-Malone et al. 2004).
Plas et al. (2006) conclude that the factors that influence the implementation of new knowledge are diverse. This raises the issue of competing terminologies, where authors use their own list of factors (or a different classification of the same factors) and terms, making the sharing and use of this information in day-to-day practice even more difficult.
To prevent possible confusion over terminology, we have chosen to carry out this review using a conceptual framework developed by Plas and colleagues (2006), which will be explained more fully in the methods section.
The Effectiveness of Strategies and Interventions
Bero et al. (1998) conducted an overview of 18 systematic reviews of interventions to promote the implementation of research findings in health care. Most of the included studies focus on physician behaviour, although nurses’ behaviour is also taken into account. Thompson et al. (2007) warn that generalizing findings from existing reviews to the nursing profession is problematic because of the different nature and (social) structure of nursing compared to medicine. Therefore we restricted our search for studies to those focusing on the nursing profession. Halfens and van Linge (2003) explored which strategies are effective for the implementation of guidelines by nurses. They concluded that whilst educational strategies improve nurses’ knowledge, this did not affect their behaviour or patient outcomes. Multiple strategies (a combination of education with one or more other strategies such as participation or aids) could be fairly effective in terms of improving the knowledge and behaviour of carers, but have hardly any effect on patient results.
Thompson et al. (2007) concluded from a recent systematic review that there are very few methodologically strong studies on the implementation of research findings in nursing practice. In the four studies they included, education was the most frequently used form of intervention used for promoting the use of research findings. However, education on its own did not prove to be effective. When education was combined with the training of a local opinion leader, increased research utilisation was observed. The same positive results were also found in the only study not using education as the primary component, but rather researchers and nurses participating in multi-disciplinary committees formed to optimise pain management (Dufault et al. 1995).
Several authors (Denis et al. 2002; Dopson et al. 2002; Halfens & van Linge 2003) advise that characteristics of the context, the new knowledge, actors involved and their possible interactions should be taken into account when implementing change. Action research methodology is a form of implementation that satisfies these points as it directly addresses the problem of the division between research and practice (Noffke & Somekh 2005). Instead of being research “on” a social setting and the people within it, it is research (in collaboration) “with” stakeholders within their natural context. Participation and knowledge of those involved in the context is essential, making a consideration of its potential contribution to the implementation of EBP worthwhile.
What is Action Research?
The origins of action research lie in the first half of the 20th century, and Lewin is often cited as the person who first used the term (Waterman et al. 2001). He was interested in a social science that could help resolve social conflicts.
This aim immediately identifies the differences between action research and other research methodologies: change (action) and research are combined. An action researcher not only wishes to gather knowledge about a particular situation, but also wishes to (help) improve the situation while investigating it.
Through a systematic review of the role of action research in UK health care settings, Waterman et al. (2001 p. 11) arrived at the following core definition:
Action research (AR) is a period of inquiry that describes, interprets, and explains social situations while executing a change intervention aimed at improvement and involvement. It is problem focused, context-specific and future-oriented. AR is a group activity with an explicit critical value basis and is founded on a partnership between action researchers and participants, all of whom are involved in the change process. The participatory process is educative and empowering, involving a dynamic approach in which problem identification, planning, action and evaluation are interlinked.
Waterman et al. (2001) conclude that two criteria are fundamental to action research. First, an intervention must be carried out as part of a cyclic process. Starting with problem identification or diagnosis (including reflection), the cycle moves on to planning, action (implementation of change and monitoring), and rounds off with evaluation/reflection before starting a new situation analysis.
The second fundamental criterion of action research concerns the partnership between the researcher and those being investigated in the research process.
In action research, those being investigated are often referred to as co-researchers. Partnership is seen as essential for developing practical knowledge and for implementing change in practice. This partnership enhances the accessibility of the knowledge created to a wider public than researchers, and so helps achieve the emancipatory intent of action research.
The partnership and the level of participation of those being investigated (the co-researchers) can vary. The minimum level of participation needed to guarantee success is not yet known (Waterman et al. 2001).
Action research is an umbrella term covering various types and, depending on the author, different classifications. In this article, the typology of Hart and Bond (1995) is used. They sketch a continuum in which four types of action research are distinguished on the basis of seven criteria (educative base, individuals in groups, problem focus, change intervention, improvement and involvement, cyclic processes and research relationship). The four types of action research Hart and Bond (1995) distinguish are: experimental, organisational, professionalising, and empowering. The “experimental” type is focused primarily on discovering general patterns that serve as the basis for choices, while the “organisational” type is more concerned with overcoming resistance and creating more-productive working relationships (Hart & Bond 1995). Moving along the continuum is the “professionalising” type, which is focused on practice, and reflects the aspirations of developing professions (such as nursing) to raise their status and develop practice based on research. Finally at the other end is the “empowering” type, which is characterised by the adoption of an anti-oppressive position in which there is collaboration with vulnerable groups in society.
The literature contains some indications that action research is a suitable methodology for bridging the gap between practice and research, and for implementing new knowledge. In their review (70% of the study participants were nurses, the other 30% were medical staff, educators, students and management), Waterman et al. (2001) established the following short-term outcomes:
- • learning results (67% of the studies) split between personal development (29%) and professional development (38%); and
- • changes in working practice, services, provision of training, and the attitude and perceptions of the staff (60%).
Long-term effects (impacts) were found in 54% of the studies. The two most important areas where these effects were achieved were in changes in the provision of training (28%) and in clinical care (22%).
To bring Waterman and colleagues’ review up-to-date, and to place more emphasis on the implementation of EBP within the nursing profession, a new review was carried out with the question: “What is known about the results of implementing EBP in nursing through action research?”