Commentary on Booth J, Tolson D, Hotchkiss R and Schofield I (2007) Using action research to construct national evidence-based nursing care guidance for gerontological nursing. Journal of Clinical Nursing 16, 945–953
Peter Lewis, Registered Nurse, Nursing Research and Practice Development Unit, The Children's Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145, Australia. E-mail: firstname.lastname@example.org
The framework of evidence based practice manifests in various ways. Booth et al. (2007) applied the theory of evidence based practice to the development of a Best Practice Statement (BPS) to be used by gerontological nurses in Scotland. While this is undoubtedly a worthwhile undertaking, clarity around certain aspects of the project within the article may enhance overall understanding of the outcomes of the project. Much of the ambiguity in the article arises from the discussion of definitions of evidence. The issue is compounded by the perfunctory treatment of study design and methodology which leads to unsubstantiated claims about the results of the study. This commentary will concentrate on these aspects of the article.
Booth et al. (2007) concur with other authors (Rycroft-Malone et al. 2004) that a myopic view of evidence as exclusively propositional knowledge is outdated and unnecessary. Nor is such a view justified when rigorous qualitative processes for developing evidence are widely reported and accepted. Booth et al. (2007) identify various sources of evidence agreed upon by the Community of Practice (CoP) engaged in the development of the BPS. We would argue that, rather than incorporating a wider definition of evidence into the creation of the BPS, Booth et al. (2007) have relied upon an alternative definition of evidence, one that suits their preconceptions about which evidentiary sources are most highly valued by nurses. A higher value leads to a more likely adoption of the guidelines.
While Booth et al. (2007) give preference to sources of evidence that are non-propositional, the article is unclear about why this preference was agreed. A clear preference for qualitative evidence is stated: ‘… given that SIGN methodology does not include qualitative research it was resolved to avoid a positivistic bias and develop a consensual approach to judging the perceived importance of qualitative evidence for nursing practice.’
A preference for non-propositional knowledge in the development of a BPS ignores the non-generalisability of that source of knowledge by ignoring its context specificity. Titchen and Ersser (2001) make a case for the generalisability of professional or craft knowledge after it has been articulated by individual practitioners, then debated, contested and verified by the wider CoP. The structure of Booth et al's. (2007) article deprives the wider CoP of the opportunity to debate and contest the generalisability of the data used by denying the reader access to it.
The implication of this is that in attempting to limit the influence of propositional evidence on the development of a BPS, a bias towards non-propositional evidence was introduced. This begs the question, is nursing an art or a science? In trying to close down this old debate, the authors inadvertently reignite it by seeking to diminish the importance of ‘proper’‘rational’‘scientific’ data in favour of ‘tacit’‘craft knowledge’ data; thereby acting on an assumption that nursing is an art. Recent work around the art and science debate has been directed at reconciling this perceived exclusivity (Mitchell & Cody 2002, Freshwater 2004, Ramsay & Bunkers 2006).
The judgement Booth et al. (2007) make in attributing greater value to qualitative evidence than to quantitative evidence is based on an ontological perspective not made explicit within the article. In other words, the claim that this type of evidence has more relevance to nursing is unsubstantiated by the article. Furthermore, the constitution of the evidence base considered is not explicit. This appears to be a major flaw in an article that purports to highlight the value of explicating the tacit. The flaw lies not in the intent of the method used but in its disclosure within the article.
It might be surmised from the title of Booth et al's. (2007) article that an action research method made a significant contribution to the development of the BPS. The title of the article leads the reader to believe that the focus will be on methodology (action research) in the context of ‘how to’ or ‘how did’ (utility). With this in mind, the method section of the article is inadequate in making clear the significance of action research. Action research is not defined and, while action research cycles are alluded to, neither are these made explicit. Recent articles provide good examples of how this method might be explained to the reader (Joyce 2005, Walsgrove & Fulbrook 2005). To make the cycles explicit would assist the reader to understand how the BPS was created. Furthermore, it would contribute to an appreciation of the rigour of the process of developing the BPS. For example, the article claims that ‘Members of the CoP then met in real time to identify aspects of the evidence base that they considered to be relevant to nursing care …’ (Booth et al. 2007) But the article does not state on what basis the evidence base was considered relevant. The reader is left, ultimately, to wonder why the process undertaken by the CoP to develop the BPS was any more valid than processes previously undertaken. The absence of an adequately articulated method is detrimental to the value of the article in that it opens to question the rigour of the process.
This article seeks to address some valuable points including which sources of evidence contribute most to the development of BPS as well as which methodology is most suited to testing the applicability of the final product. Both these aims are worthy and although some of the limitations of the article are acknowledged by the authors in their conclusion, more detail in certain sections of the article may have contributed to a deeper understanding by the reader about the value of the BPS described. More significantly, the process might be subject to replication in other contexts and its effectiveness made more rigorous. Ultimately, we do not have access to the level of detail necessary to explain the ‘low-cost, involving methodology’ (Booth et al. 2007), and so can't make a judgement about its effectiveness in producing a BPS which truly contributed to the achievement of best practice within the CoP, never mind the broader practice community.