Health professionals are highly discerning consumers of 'knowledge' to inform the decisions they make at the point of care to ensure the best outcomes for their patients. It has long now been recognized that high-quality research evidence is the gold standard 'knowledge' base for this purpose. Historically evidence of effectiveness has been a primary focus but in more recent years other forms of evidence have been recognized as equally important to respond to the diversity of information required in health settings (feasibility, appropriateness and meaningfulness).1 However, in the absence of research evidence what is the best available evidence to inform decisions? Of course the Joanna Briggs Institute (JBI) has long held the view that an inclusive approach to the conceptualization of what counts as evidence is important to the evidence-based movement.2 The Institute moves this definition beyond research to include, where appropriate, textual, nonresearch data.

In 2015, the JBI methodological group devoted to the synthesis of textual, nonresearch data reviewed and updated several steps in the process as they related to inclusion criteria, search strategy, critical appraisal and data extraction.3 The methodology is once again under review by the group which now seeks to reconceptualize the existing approach, drawing upon diverse theoretical perspectives to inform and understand the value of this type of data to the evidence-based healthcare movement. It utilizes both discourse analysis and argumentation theory to reconstruct an understanding of the theory and practice of textual synthesis to both accentuate the critical and socially transacted properties of knowledge production and its use in context.

Although yet to be published, textual, nonresearch data are considered, in this latest version, to be inclusive of three distinctive data types, namely, opinion, narrative and policy (previous iterations of this methodology have articulated the data set to be inclusive of narrative, opinion and text). The explicit inclusion of policy, as opposed to the more general 'text', is an interesting one that I think opens up the utility and scope of this type of review. The historical view has been that reviews of text and nonresearch evidence should only be conducted where there is no research evidence to draw upon. However, there is the possibility that this evidence may offer an interesting adjunct to traditional reviews that provides new or different information to contextualize traditional syntheses. For example, there is the potential that discursive anomalies may occur between published research and unpublished literature in relation to politically sensitive topics where a review that includes both may be of great interest and use and may indeed be beneficial with respect to implementation initiatives. Some of these explanatory data may be identified through qualitative inquiry and synthesis, but in some instances an assessment of nonresearch evidence, such as that found in policy, may also serve us well. In that sense, the inclusion of text and nonresearch evidence could be viewed in a similar manner to realist synthesis in the sense that it has the potential to contextualize complex interventions that are often situated in dynamic, complex systems.4,5

Of course with this view in mind, to reasonably incorporate textual, nonresearch data in syntheses designed to inform clinical decision-making at the point of care, it is critical that reviewers have (in so far as is possible) a transparent framework within which to make assessments regarding the legitimacy, authority, interests, needs, values, ideology, preferences, motivations, world views and obligations that sit behind and inform the text they seek to include. The language of narrative, opinion and policy is therefore the primary 'data' for analysis and the discursive struggle therein that seeks to shape our collective conceptual framing of problems and guide us toward (potential) solutions.

There have long been arguments surrounding the appropriateness of utilizing nonresearch evidence to inform decision-making. However, where research data do not exist, patients still do; and they still require health professionals to make decisions regarding their treatment and care. Equally, the measure of the success of an intervention is not always simply in the identification of its effectiveness in isolation, and thus the continued inclusion of other forms of evidence remains critical to the knowledge translation agenda.

Acknowledgements

Conflicts of interest

The author reports no conflicts of interest.

References