Development of methodology and methods in the field of qualitative synthesis is growing at a significant rate. There is an increasing number of published qualitative reviews as well as books and papers related to methodology. The field of qualitative synthesis is now entering the phase where comparative evaluation research is not only possible but increasingly necessary. A small number of comparative studies has also been published, examining the differences within and between methodologies.1-3

As a field of scholarship, qualitative synthesis is characterized by a rich collection of methods, a strong theoretical basis (albeit not always referred to in published qualitative reviews), and ongoing debates on methods and goodness of fit for purpose. However, very few methodologists have given real consideration to the information needs of end users.4-7

Qualitative systematic reviews provide answers to complex questions related to subjective or humanistic perspectives. Much like primary qualitative research, qualitative synthesis has been framed as identifying the meanings people attribute to their experiences, the beliefs that underpin behavior associated with adherence to a therapy: questions that begin with "what is it like[horizontal ellipsis]", or "what is the experience of[horizontal ellipsis]" and "why do people[horizontal ellipsis]" are often used to demonstrate the areas that qualitative synthesis addresses. Equally important yet rarely addressed are considerations when choosing a method, what the results of the review are to be used for, and who the intended end users and what their information needs are.8

The reason these are relevant is embedded in the categorization of methods of qualitative synthesis. Since Noblitt and Hare's work on meta-ethnography, methods of qualitative synthesis have been categorized as being interpretive or integrative.9 More recently, this same categorization has been described as inductive or deductive, where the inductive approach leads to new theoretical understandings that inform modelling of how mid-level theory fits or extends conceptual understanding of a topic.10 Qualitative methods of synthesis that fit within this broad categorization include meta-ethnography and realist synthesis. The second grouping, described as integrative or deductive, includes meta-aggregation and qualitative meta-synthesis where the evidence is qualitatively compiled, but using methods that do not lose the "parts in the emerging whole". The misconceptualization that may arise from this classification is that while one method is described as inductive and the other as deductive, both are grounded in interpretivism. Whether the methodology is meta-aggregation or ethnographic, the defining characteristic is that it fits within the interpretive paradigm. It is the paradigm, and not just the vernacular associated with the methods, that identifies reviews as interpretive.

If this misconceptualization can be put aside, the focus can increasingly give consideration to prioritization of the choice of methodology based on goodness of fit with the information needs of the end users. The information needs of end users extend beyond the findings or recommendations; they also include transparency and auditability in the review methods to establish the degree of confidence that can be attributed to the findings and recommendations.

A quantitative systematic review is understood to consist of a comprehensive set of methods for the development of an a-priori protocol, leading to a comprehensive search, critical appraisal to establish study credibility and dependability, and data extraction and synthesis leading to conclusions or recommendations. The Joanna Briggs Institute and many groups adhere to these internationally recognized characteristics of a high quality systematic review for the synthesis of evidence arising from all paradigms and any empirical method.1,2,11 The decision-making steps are reported, and an audit trail demonstrates the review author's adherence to the process. The results of a review intended to inform decision-making in policy or practice need to be clearly attributable to the data from which they were derived, and recommendations arising need to be graded by level, with an indication of the degree of confidence for each recommendation that clearly addresses the dependability of included studies, the plausibility of the original study findings, and the strength of the paper.8,12,14

Meta-ethnography has a focus on constructing new interpretations from existing qualitative literature. The reviewer is an interpretive agent who facilitates conceptual development expressed through new models, theories and concepts, often expressed as mid-level theory. The interpretation is always from an incomplete data set (published research), raising questions about whether mid-level theory should be subject to further research and evaluation prior to implementation in policy or practice.8-10

Realist synthesis specifically aims to develop mid-range theory, where theory is understood on the basis of Mertons' (1967) definition of theoretical abstraction (but not to the extent where propositions cannot be constructed and subject to empirical testing).8 This definition demonstrates where realist synthesis arguably has its best fit, that is, in the development of mid-level theory for subsequent empirical validation.8

Meta-aggregation is considered an integrative methodology: it brings together literature without an emphasis on re-interpretation. Meta-aggregation includes a formalized a-priori protocol, a comprehensive search, and critical appraisal; it aligns with internationally recognized standards for systematic review methods. The end product of meta-aggregative reviews is practice level theory that is directly derived from the aggregation of qualitative research findings, and therefore is suited to the information and knowledge needs of policy and practice. Meta-aggregation transparently provides evidence of methodological rigor and auditability of methods: information that is critical to the information needs of policy and practice if there is to be confidence in the evidence and how it was synthesized. The integration of ConQual within meta-aggregation further enhances the transparency of meta-aggregative review reporting, adding a measurable rating for confidence for each synthesized finding.13

Academics continue to debate the issue of how to select which approach for qualitative synthesis, raising issues of question type and the information needs of end users. However, distinction between the outputs of types of qualitative synthesis has not been a significant part of the debate, and end user perspectives in relation to the transparency of reporting as well as the utility of findings is under-represented in the literature. Meta-aggregation provides practice level theory and ConQual adds further methodological transparency. Qualitative systematic review questions are situated within humanistic frameworks, and they ask "what is it like[horizontal ellipsis]" or "why do people[horizontal ellipsis]" or "what is the experience of[horizontal ellipsis]"'. These can be intended to develop mid-level theories for empirical testing, or can be grounded in integrative synthesis methodologies such as meta-aggregation or qualitative meta-synthesis to inform policy or practice through "lines of action".

Director, Implementation Science, The Joanna Briggs Institute

Conflicts of interest:

Associate Professor Craig Lockwood is the program director for Implementation Science in the Joanna Briggs Institute.

References