Qualitative synthesis: A guide to conducting a meta-ethnography

Abstract

Introduction

Qualitative meta-synthesis draws together data from multiple studies and can enrich understandings of client experiences and inform health policy. Meta-ethnography is the most commonly utilised qualitative synthesis approach in healthcare; however, there is variation in the conduct and quality of many published meta-ethnographies. This paper aims to guide occupational therapists on the process of completing a meta-ethnography.

Method

A systematic search of articles published from 1986–2016 was conducted. Studies which described an evaluation or critique of meta-ethnography as a research method, and consequently provided reflections or guidance on the method, were included. Two authors independently screened papers and nine relevant papers and one key text were identified. Two authors independently extracted data from identified sources.

Results

The seven stages of meta-ethnography described by Noblit and Hare (1988) are presented and integrated with key advancements from subsequent publications. Stages include starting the meta-synthesis, consideration of relevant studies, reading the studies and deciding on how they should be put together, translating the studies into one another, synthesising translations and expressing the synthesis.

Conclusion

Given the proliferation of qualitative methods in occupational therapy research, rigorous use of meta-ethnography has great potential to contribute to the research evidence base.

Keywords

Meta-ethnography meta-synthesis research methods reporting qualitative synthesis qualitative health research

Introduction

Qualitative research methods have been identified as congruent with the philosophy of occupational therapy (Hammell, 2001) and there has been growth in the use of research situated within the qualitative paradigm in the discipline (Stanley and Nayar, 2015). Drawing together the findings of multiple primary studies is becoming increasingly common and is considered an important source of evidence for healthcare and policy (Classen and Alvarez, 2015). A range of approaches to synthesising qualitative research have been described, including meta-narrative, critical interpretative synthesis, meta-study, meta-ethnography, grounded theory, thematic synthesis, textual narrative synthesis and framework synthesis, and ecological triangulation (Barnett-Page and Thomas, 2009). The existence of various approaches makes it confusing for the novice researcher when deciding how to operationalise different meta-synthesis approaches (Tricco et al., 2016), and few comprehensive descriptions are available to guide researchers in the pragmatics of the synthesis process with the aim of producing a rigorous, clear report (Tong et al., 2012). In this paper we describe the processes involved in conducting a meta-ethnography, the most commonly used qualitative synthesis approach. We systematically searched for recent publications on the development of meta-ethnography, to inform and guide both novice and experienced occupational therapists and researchers in the use of this evolving approach.

Meta-ethnography was one of the first synthesis methods developed, and since its inception in the late 1980s it has become increasingly popular and influential in healthcare research (Tong et al., 2012). Meta-ethnography is now the most frequently used qualitative synthesis approach (Hannes and Macaitis, 2012). George Noblit and Dwight Hare are widely considered to be the founders of meta-ethnography, originally developing their interpretative seven-step qualitative method of synthesis for a small number of ethnographic studies (five) in the area of education. They published a key text on meta-ethnography in 1988, where they described three trajectories of synthesis that a meta-ethnography could follow: reciprocal translation; refutational synthesis and line of argument synthesis (Noblit and Hare, 1988). Reciprocal translation occurs when different accounts are translated into one another; in contrast, refutational synthesis allows interpretation of conflicting accounts, whereas line of argument synthesis is used to build an account of the overarching interpretation of the studies (Noblit and Hare, 1988). While Noblit and Hare’s seven-step process continues to be the main process for conducting meta-ethnography (Lee et al., 2015), it is an evolving research approach, with many researchers using, adapting and developing the method (France et al., 2014; Tong et al., 2012).

Meta-ethnography offers a means of rigorously synthesising qualitative research, with many potential benefits to clients, clinicians and policy makers. Tong et al. (2012: 1) identified that synthesising findings from multiple qualitative studies can provide ‘a range and depth of meanings, experiences, and perspectives of participants across health-care contexts’. Meta-ethnographic studies offer greater description of methods and higher order interpretation when compared with conventional narrative literature reviews and may be particularly useful in synthesising client experiences of their illness and the care received (Atkins et al., 2008). Through the process of translating studies into one another (Noblit and Hare, 1988), the aim of meta-ethnography is to create novel interpretations and conceptual innovation of the phenomenon being studied (Malpass et al., 2009).

There are parallels between meta-synthesis of qualitative studies and systematic review of quantitative research with meta-analysis. Both approaches involve identifying a review question, developing inclusion/exclusion criteria, systematically identifying studies, appraising those studies and then synthesising data (Murray and Stanley, 2015). However, recent methodological reviews have identified that many researchers do not adhere to the conventions of meta-ethnography (France et al., 2014); consequently, meta-ethnographies may not reach the level of analytical depth for which they were intended (Thorne, 2015). This reflects a broader issue in qualitative synthesis, not limited to just the meta-ethnographic approach, where there is a proliferation of qualitative syntheses with meticulous search strategies but little interpretative work (Thorne, 2017).

Given the value of qualitative meta-synthesis in synthesising the findings from multiple studies to findings of a higher conceptual level, it is not surprising that occupational therapists have engaged with the method over the past decade in a variety of research areas, including adults’ experiences of spinal cord injury (Hammell, 2007), challenging experiences in occupational therapy practice (Murray et al., 2015) and engaging in occupation post stroke (Williams and Murray, 2013), amongst others. While these occupational therapy studies used meta-synthesis as the research method, there is frequently a lack of clarity regarding the specific type of synthesis used and at times little description of the data analysis process. For occupational therapy researchers’ contributions in meta-ethnography to be taken seriously, they must be rigorous. Therefore, the aim of this review is to synthesise journal papers that present an evaluation or critique of meta-ethnography as a research method and/or reflections on conducting a meta-ethnography.

Method

The review followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al., 2009).

Study identification

A thorough search of literature was conducted to gather and synthesise relevant papers that presented methodological guidance on conducting a meta-ethnography or critiqued meta-ethnography as a research method. The systematic search was conducted in July 2016. We searched the following databases: Web of Science, CINAHL, Embase, ERIC, Academic Search complete, Social Sciences Index, PsychINFO and Pubmed. A search of abstract or title using the keywords meta ethnograph* OR meta-ethnoghaph* OR metaethnograph* was completed. The entire search string is available on request. Studies were limited to peer-reviewed articles (excluding grey literature) published in English between the years 1986 and 2016, to align with the emergence of meta-ethnography as an approach in the literature.

Study selection

Studies which described an evaluation or critique of meta-ethnography as a research method, and consequently provided reflections or guidance on doing meta-ethnography, were included. Papers which presented the findings from a synthesis of different studies using meta-ethnography as a method along with papers which described the inclusion of quantitative data in a qualitative synthesis were excluded. Two authors independently screened papers by title and abstract with 100% agreement. An assessment of quality was not used as the authors did not wish to exclude any article from the small number of relevant articles which met the inclusion criteria for the review question.

Narrative synthesis

Data were extracted from the included papers and text by two authors independently and discussed in order to present information that advanced the meta-ethnography method from the original text (Noblit and Hare, 1988). The findings are presented using Noblit and Hare’s (1988) original seven-phase process as an overarching framework while acknowledging the seven stage process is iterative and phases may overlap and run in parallel (Noblit and Hare, 1988). These stages include starting the meta-synthesis, consideration of relevant studies, reading the relevant studies and deciding on how they should be put together, translating the studies into one another, synthesising these translations and expressing the synthesis for the target audience.

Results

The initial search produced 1538 results. Duplicates were removed and the search was refined by screening all titles, abstracts and full texts as required. Figure 1 depicts the process of study selection in the review. Nine papers were identified for inclusion (Table 1), along with one key text (Noblit and Hare, 1988). Inclusion of Noblit and Hare’s original text was agreed (by consensus) because of its seminal nature. Despite it being published over 30 years ago, it continues to be highly cited, remaining the methodological cornerstone of contemporary meta-ethnography research. The findings of this review are framed using the seven-phase process described by Noblit and Hare (1988).

Figure 1.

PRISMA 2009 flow diagram inclusion procedure (Moher et al., 2009).

Table 1.

Characteristics of the studies included.

Author	Year	Journal	Aim
Noblit and Hare	1988	Book	To describe the origins and the theoretical background for meta-ethnography. A seven-step process is described with detailed examples at each stage.
Doyle	2003	Qualitative Research	To discuss weaknesses in the meta-ethnographic process, using three stages of meta-ethnography (case selection, analysis and synthesis). Enhancements are also provided.
Atkins, Lewin, Smith, Egel, Fretheim and Volmink	2008	BMC Medical Research Methodology	To discuss the methodological and practical challenges in conducting a meta-ethnography of qualitative research on adherence to tuberculosis treatment. The methods used are described alongside additional steps found useful in clarifying the process and key methodological challenges encountered in implementing the meta-ethnographic approach.
Campbell, Pound, Morgan, Daker-white, Britten, Pill, Yardley, Pope and Donovan	2011	Health Technology Assessment	To evaluate meta-ethnography as a method for synthesising qualitative research studies in health and healthcare.
Toye, Seers, Allcock, Briggs, Carr, Andrews and Barker	2013	BMC Medical Research Methodology	To explore the processes that determine the ‘quality’ of qualitative research for inclusion in qualitative research syntheses. Qualitative methods were used to explore the process of quality appraisal within a team of qualitative researchers funded to undertake a meta-ethnography of chronic non-malignant musculoskeletal pain.
Lee, Hart, Watson and Rapley	2015	Qualitative Research	To consider the forms of work (the practical action and practical reasoning) comprising the completion of a meta-ethnography and to consider the difference these forms of work make to a meta-ethnography.
France, Ring, Thomas, Noyes, Maxwell and Jepson	2014	BMC Medical Research Methodology	To investigate, through methodological systematic review, the application and reporting of methods in recent meta-ethnography journal papers, focusing on the analysis and synthesis process and output.
Toye, Seers, Allcock, Briggs, Carr and Barker	2014	BMC Medical Research Methodology	To build on the methods of Noblit and Hare (1988) and explore the challenges of including a large number of qualitative studies into a qualitative systematic review.
France, Ring, Noyes, Maxwell, Jepson, Duncan, Turley, Jones and Uny	2015	BMC Medical Research Methodology	To describe the protocol for a study which aims to develop a unique evidence based guideline for reporting qualitative meta-ethnographic studies. (The guideline will be called eMERGe.)
France, Wells, Lang and Williams	2016	Systematic Reviews	To describe why, when and how to update a meta-ethnographic qualitative synthesis and illustrate this process using a worked example.

Phase one: Getting started

Initially, meta-ethnographers identify the area of interest to the author and audience (Noblit and Hare, 1988) and list studies for possible inclusion. Conducting a synthesis must build on existing knowledge (Toye et al., 2014). Therefore, at the outset of the review, the meta-ethnographer needs to ask if a synthesis of the topic is required and if a qualitative synthesis fits with the research question (Toye et al., 2013). Establishing a team of researchers, with differing approaches and opinions and with key skills, to conduct the synthesis adds rigour to the meta-ethnography (Campbell et al., 2011). It has also been recommended that a research advisory group including a clinician, client and public representative, and policy maker is utilised at this stage as they are knowledgeable about the topic and help to ensure that the knowledge translation is relevant (Toye et al., 2014).

Phase two: Deciding what is relevant to the initial interest

Phase two is focused on selecting studies for inclusion in the synthesis and involves defining the focus of the synthesis, locating relevant studies, making decisions on study inclusion or exclusion, and quality assessment of included studies (Atkins et al., 2008). In contrast to systematic reviews where every possible study, on a given intervention, is sought, and in keeping with principles of qualitative research, theoretical sampling is encouraged in meta-ethnography (Atkins et al., 2008). However, a lack of guidance on how to operationalise theoretical sampling at the search stage has been noted (Atkins et al., 2008). Noblit and Hare (1988) advised selecting studies that provide data for the research question and encourage maximum variation sampling, as the focus is not on representativeness and generalisability.

Searching and screening literature can be challenging and time consuming, yet it is a necessary phase and can involve several decision processes. In addition to database searching, citation searching, hand searches and consultation with experts are all recommended (Atkins et al., 2008). Involving a health science librarian in this stage of the meta-ethnography can offer expertise in creating an effective and comprehensive search strategy.

Doyle (2003) suggested enhancements to the process of identifying studies for inclusion by establishing ‘predetermined boundary conditions’ for study selection, that is, narrowing the inclusion criteria for studies. For example, her meta-ethnographic synthesis was to be presented to policy makers, so she narrowed study selection criteria to studies which would withstand scrutiny from this particular audience and only included studies with interpretation and analysis that had evolved from established theories and methods. Noblit and Hare (1988) suggested varying the unit of analysis. Units of analysis are the population or the topic area that is being analysed in the study, and can be individuals, studies, groups or organisations. In Noblit and Hare’s (1988) original meta-ethnography, for example, their unit of analysis was schools in one study and school principals’ behaviour in another, to provide strong contrast for analysis. In contrast, Doyle (2003) recommended considering predetermined boundaries in terms of unit of analysis for practical reasons when designing meta-ethnographic studies. For example, rather than varying the units of analysis, Doyle (2003) focussed on one unit of analysis, a school setting, as she found studies with various units of analysis had strong data in some domains and weak data in other areas.

Determining the inclusion of studies with consensus reached by two reviewers enhances rigour in this phase (Atkins et al., 2008). One particular issue when determining inclusion can be whether to include mixed methods studies with a qualitative component or not. Atkins et al. (2008) described using two reviewers to reach consensus on decisions of this nature. Conflicts can be resolved through discussion or by a third reviewer.

Opinion differs as to the number of studies that are needed for a viable meta-ethnography. France et al. (2014) found an average of 21 papers (range 3–77) were included in meta-ethnographic reviews. Recently, the possibility of including over 70 papers in a meta-ethnography has been suggested, using Nvivo 9 software to manage the synthesis and to aid collaborative team interpretations (Toye et al., 2014). Including a large number of studies in a meta-ethnography risks reducing the rigour of interpretive work and increases the risk of becoming reliant on procedural approaches to synthesis (Campbell et al., 2011).

Quality appraisal is another contentious topic in meta-ethnography, partly due to a lack of agreement about what quality is in qualitative research (Toye et al., 2013). Indeed, debate continues regarding whether or not to appraise the quality of studies to be included in a meta-ethnography. Toye et al. (2013) suggest that some quality assurance of studies needs to be considered in order for the meta-ethnography to be more than an anecdotal account. Suggested tools for quality assessment that have been used across a number of meta-ethnographies include the Critical Appraisal Skills Programme (CASP) and the Qualitative Assessment and Review Instrument designed by Joanna Briggs (JBI-QARI) (Toye et al., 2013). Issues identified with currently available appraisal checklists include the inconsistent judgements produced by current checklists and the inherently subjective process of qualitative research appraisal (Toye et al., 2013).

A number of authors have offered guidance on quality assessment of included studies in a meta-ethnography. Reflecting on their experience of quality assessment of included studies in a meta-ethnography, Atkins et al. (2008) argued that quality assessment affected the contribution of an individual paper to the synthesis as papers that provided mainly descriptive data offered fewer insights and contributed less substantively to themes than papers that included thick description and rigorous analysis. Toye et al. (2013) suggested two core facets of quality for inclusion in meta-ethnography: (a) conceptual clarity – how clearly has an author articulated a concept that facilitates theoretical insight (can you decipher a translatable concept?) and (b) interpretative rigour – what is the context of interpretation, how inductive are the findings and has the researcher challenged their interpretation (as interpretative rigour encapsulates a reflexive approach to the research).

Phase three: Reading the studies

Noblit and Hare (1988) advised repeated reading of the studies and noting or extracting the key concepts or metaphors that existed in each study. They described metaphors of a study as ‘what others may call themes, perspectives and/or concepts revealed by qualitative studies’ (14). These key concepts or metaphors are the ‘data’ for the synthesis. When reading studies, documenting information in a table captures each study’s context, data collection method, description of participants, organises studies in groups (Pound et al., 2005) and assists with capturing all the primary study author’s concepts and conclusions. Reading the studies in an intensive, repetitive and highly active way is central to the meta-ethnographic process (Lee et al., 2015). Reading can occur with different intentions across the process, for example to familiarise, to identify, to extract, to record, to organise, to map and to verify data (Lee et al., 2015).

Maintaining reflective journals documenting the researchers’ thoughts and insights ensures that richness and depth are captured (Doyle, 2003). Using analysis software, rather than a hard copy data extraction form, can allow the coding of conceptual findings as they appear in the original text and allows all members of the team to view one another’s interpretations. This software also allows ideas to be tracked as they develop (Toye et al., 2013) and is valuable where there are a large number of studies to synthesise. The use of annotations has also been proposed to assist the active reading process (Lee et al., 2015).

Meta-ethnographies often describe data in terms of first order, second order and third order constructs (Toye et al., 2014). First order constructs are the research participants’ own everyday words; second order constructs are the researchers’ interpretations of the first order constructs (key concepts); and third order constructs are the synthesis team’s interpretation of the second order constructs. The raw data of meta-ethnographic studies are the second order constructs or key concepts (Toye et al., 2014).

Atkins et al. (2008) queried the usefulness of the notion of first, second and third order constructs based on their experience of the following issues: first order constructs are already selective and do not represent the totality of participant accounts and it is difficult to decipher first from second order constructs at times. Based on their experience of conducting three meta-ethnographies and the use of methodological memos and field notes, Lee et al. (2015) attempted to articulate and make sense of the hidden labour of meta-ethnography. Two key activities across the stages of meta-ethnography, reading and conceptual innovation, are described in detail by Lee et al. (2015). At different stages, reading had a different intention, including appraisal, extraction, mapping, comparing and verification, and an array of techniques to achieve these purposes were identified, including annotating, coding, listing and tabulating. A challenging dimension of reading described by Lee et al. (2015) was the shift required to move from working within accounts to working across accounts.

Conceptual innovation during the synthesis process is the employment of concepts not found in the included studies (Strike and Posner, 1983). Lee et al. (2015) proposed that conceptual innovation is a key meta-ethnography activity which, depending on the interpretative quality of the included studies, may not always be possible. As individual accounts/studies are analysed and a concept is generated, that concept is examined in terms of whether it makes sense when compared to the original accounts. Meta-ethnography is not guaranteed to result in conceptual innovation – the raw data or the included studies may not be of sufficient quality to enable conceptual innovation, in which case a meta-ethnography may not be possible despite an already significant investment of time in searching for, appraising and reading studies (Lee et al., 2015).

Phase four: Determining how the studies are put together

Noblit and Hare (1988) suggested creating a list of metaphors/ideas (phrases or themes) from each paper and juxtaposing them at this stage, to examine the relationships between the metaphors in each study. Juxtaposing the metaphors, ideas, phrases and/or themes enables the research team to look across studies for common and recurring concepts that were encompassed in each study, thereby illustrating the relationships across studies and assisting with the translation process in step five (Noblit and Hare, 1988). ‘Concepts’ are described by France et al. (2014: 9) as ‘a meaningful idea that develops by comparing particular instances. Fundamentally, concepts must explain not just describe the data’. Key concepts are then extracted which encompass the themes, as defined by the authors of the original studies (Toye et al., 2013). The terminology used in stages four, five and six of the meta-ethnography process is often unclear (Toye et al., 2013). Phase four is primarily involved in identifying the key concepts or second order interpretations. At this stage a grid or table can be used to display concepts and themes across all studies. From this grid, themes can be reduced into relevant categories and these categories can be revised and merged through discussion of how they were related and by reference to the original text (Atkins et al., 2008).

Phase five: Translating the studies into one another

Noblit and Hare (1988) described the translation process as treating the studies as analogies. For example, ‘one study is like another except….’ (29). The translation process entails examining the key concepts within and across the studies and is similar to the method of constant comparison. The translations can be conducted in chronological order (Campbell et al., 2011). Each concept from each paper is compared with all the other papers in turn to check for the absence or presence of concepts. This highlights similarities and differences between concepts and metaphors and enables the researcher to organise them into further conceptual categories (Toye et al., 2013). Arranging the research studies chronologically and translating the themes, using the key concepts/themes from paper one to paper two and then translating the synthesis of these two papers to paper three and so on, has been suggested (Atkins et al., 2008), particularly when synthesising a large number of papers over a large timespan. Team discussions regarding the key concepts and their meanings can result in collaborative interpretations. Maintaining a personal journal during this phase of the process ensures that the researcher is aware of their position from a theoretical point of view when analysing the studies (Doyle, 2003).

Phase six: Synthesising the translations

The researcher now views the studies as a whole rather than as parts, with the aim of forming a conceptual framework (Toye et al., 2014) making sense of the concepts and categories across the studies. Noblit and Hare (1988: 28) described this phase as ‘making the whole into something more that the parts alone imply’; however, they did not describe how to operationalise this phase in detail. By referring to a completed table of data, as recommended by Atkins et al. (2008), researchers can establish the relationship between studies and decide if the synthesis is a refutational synthesis (differing accounts) or a reciprocal synthesis (accounts being similar), from which a line of argument can be created. An overarching model or line of argument can then be created, using a matrix to consider the key concepts and categories in chronological order. Each researcher can do this independently, with the themes subsequently merged and discussed by the team. Face-to-face research team meetings and the use of analysis software (when there are a large number of studies to synthesise) can facilitate this translation process and assist with refining the interpretations in the translation process (Toye et al., 2014). A diagram can be used to develop a line of argument collaboratively over time, with team discussions and interpretations being pivotal to this stage (Toye et al., 2013).

Atkins et al. (2008) described at this stage a process where translated themes and subthemes were listed in a table and juxtaposed with secondary themes derived from author interpretations. Each member of the research team independently developed an overarching model that linked together the translations and authors’ interpretations. These models were merged and discussed and used to generate hypotheses in order to produce a ‘line of argument’ synthesis. Reflecting on the process, Atkins acknowledged that the inherent subjective nature of interpretation and the widely varied interpretations of the research team meant compromises had to be made. Providing original authors of included studies with a copy of the draft translation and asking how the interpretation related to their original study was proposed as a useful method of member checking (Doyle, 2003).

Phase seven: Expressing the synthesis

This involves writing the synthesis, while considering the audience for which it is being created (Noblit and Hare, 1988). Syntheses can usually be presented in textual format, while the use of diagrams and other visual representations along with text have been suggested (Toye et al., 2013). At present no accepted standard of reporting meta-ethnographic studies exists (France et al., 2014). In order to contribute to improving client outcomes and improved healthcare, it is essential that meta-ethnography studies are reported clearly. The development of a specific meta-ethnography reporting tool, eMERGe, is currently underway (France et al., 2014). A website has been created to document the process and provide information on the tool as it develops (www.stir.ac.uk/health-sciences-sport/research/groups/emerge/).

Discussion

Meta-ethnography is evolving and growing in popularity as an effective and useful method of meta-synthesis. Meta-ethnographies have drawn together findings of studies in systematic ways to produce new insights and theoretical advancements in contrast to traditional narrative-type literature reviews (Campbell et al., 2011) and are becoming more prominent in providing evidence to underpin policy and practice (France et al., 2016).

Murray and Stanley (2015) recommended that a meta-synthesis is more suited to experienced qualitative researchers. Within recent critiques and descriptions of meta-ethnography studies, a collaborative team approach is evident, especially when there are a large number of studies for synthesis (Atkins et al., 2008; Campbell et al., 2011; Lee et al., 2015). Forming a team of researchers from different backgrounds to collaborate on the meta-synthesis can facilitate the development and articulation of the different perspectives, interpretations and findings. This has been suggested as a means to enhance the rigour and transparency of the process as well as ensuring that the results are more relevant to the audience for which the meta-ethnography is conducted (Atkins et al., 2008). Furthermore, a team approach offers the opportunity for novice researchers to learn and develop synthesis skills from more experienced researchers.

Within the discipline of occupational therapy, qualitative meta-synthesis and particularly meta-ethnography is growing in popularity (for example, see Classen et al., 2009; Fossey and Harvey, 2010; Frostad et al., 2016; Gibbs et al., 2015; Shamrock et al., 2016; York and Wiseman, 2012). Recent reviews using meta-ethnography illustrate the potential of the method to contribute to occupational therapy practice and research. Gibbs et al. (2015) included 35 papers in their meta-ethnography and the findings provided clinicians with a novel and comprehensive understanding of the process of adapting to the role of parenting a preterm infant. York and Wiseman (2012) included four studies in their meta-ethnography of gardening as an occupation, and developed a line of argument synthesis showing that the natural environment acted as stage on which people could develop their emotional, spiritual, cognitive and physical abilities. Frostad et al. (2016) used meta-ethnography to synthesise findings from 16 studies, and identified four key concepts as important for return to work after an acquired brain injury: empowerment, self-awareness, motivation and facilitation.

These studies demonstrated that the process of meta-ethnography generated novel insights, for example on individual experiences of an event (becoming a parent of a preterm infant) or occupation (experiences of gardening), without a large team.

Clear reporting guidelines such as CONSORT (for randomised control trials) and PRISMA (for systematic reviews and meta-analysis of randomised controlled trials) (Moher et al., 2009) ensure consistency of approach and reporting. In contrast, one common critique of meta-ethnography is that there are few published reporting guidelines for qualitative synthesis (France et al., 2015; Tong et al., 2012). There have been calls for the development of bespoke guidelines to improve reporting quality (France et al., 2015). The ENTREQ (enhancing transparency in reporting the synthesis of qualitative research) statement (Tong et al., 2012) is a recently developed reporting guideline that aims to encourage transparency in the reporting of syntheses. The ENTREQ statement contains 21 items grouped into five domains: introduction; methods and methodology; literature search and selection; appraisal and synthesis of findings. However, since it was published, the ENTREQ statement has only been used infrequently in reporting (France et al., 2015), possibly because it was developed for meta-synthesis more broadly rather than specifically to address reporting a meta-ethnography. A protocol has been published detailing plans to develop specific meta-ethnography reporting guidelines (France et al., 2015). These are based on a methodological systematic review to identify recommendations/guidance in the conduct and reporting of meta-ethnography, a review and audit of published meta-ethnography, and consensus methods with international experts and stakeholders (France et al., 2015). The eMERGe guidelines are expected to be available in 2017 and it is anticipated that the use of these guidelines will be expected by publishers of different healthcare journals when reporting meta-ethnography syntheses (France et al., 2015). Adoption of a standard reporting mechanism for meta-ethnographic studies will increase the clarity of the methodology, analysis and findings to be clearly conveyed, thereby enabling users to assess the quality and trustworthiness of a meta-ethnography (France et al., 2015).

Conclusion

Meta-ethnography is an evolving approach to the synthesis of qualitative research that is increasing in popularity in occupational therapy and healthcare research. Recent examples of meta-ethnographies in occupational therapy demonstrate the usefulness of the approach in creating new insights through drawing studies together in a novel and effective way. The possibilities of meta-ethnographic research to positively impact on healthcare and health policy is significant; however, the rigour and transparency of published meta-ethnographies varies and is often poor. A clear understanding of the steps of the meta-ethnography process is pivotal to the successful production of a rigorous synthesis which will be of appropriate depth to inform occupational therapy practice and policy. Most meta-ethnographies follow the seven steps initially proposed by Noblit and Hare in their seminal text (1988). Since this text was published, multiple researchers have proposed methodological advancements. In this review of meta-ethnography methodology journal papers, the seven stages of meta-ethnography are presented alongside advancements from subsequently published journal papers. Two key recommendations from the reviewed papers are the need for researchers undertaking a meta-ethnography to be experienced in qualitative research and the benefits of a collaborative research team with differing skill sets. Ongoing work to develop meta-ethnography reporting guidelines is likely to increase the rigour of future published meta-ethnographies.

Key findings

Meta-ethnography of qualitative research offers the potential to expand insights and knowledge about phenomena. To achieve this aim, meta-ethnography must involve significant interpretative effort.

A clear understanding of the meta-ethnography process is essential for researchers undertaking or interpreting qualitative syntheses.

Expertise in qualitative research methods and team collaboration are recommended for successful completion of a meta-ethnography.

What the study has added

The review presents practical guidance to occupational therapist readers and researchers on the process of completing a rigorous meta-ethnography utilising the seven-phase process described by Noblit and Hare (1988) alongside subsequently published methodological advancements.

Footnotes

Research ethics

Ethical approval was not required for this review. Consent was not relevant to this review.

Declaration of conflicting interests

The authors confirm that there is no conflict of interest.

Funding

This research received no specific grant support from any funding agency in the public, commercial or not for profit sectors.

References

Atkins

Lewin

Smith

et al. (2008) Conducting a meta-ethnography of qualitative literature: Lessons learnt. BMC Medical Research Methodology 8: 21. doi: 10.1186/1471-2288-8-21.

Barnett-Page

Thomas

(2009) Methods for the synthesis of qualitative research: A critical review. BMC Medical Research Methodology 9: 59. doi: 10.1186/1471-2288-9-59.

Campbell

Pound

Morgan

et al. (2011) Evaluating meta-ethnography: Systematic analysis and synthesis of qualitative research. Health Technology Assessment 15(43): 1–164. doi: 10.3310/hta15430.

Classen

Alvarez

(2015) Evidence-Informed Reviews—Moving Occupational Therapy Practice and Science Forward, Los Angeles, CA: SAGE Publications.

Classen

Winter

Lopez

(2009) Meta-synthesis of qualitative studies on older driver safety and mobility. OTJR: Occupation, Participation and Health 29(1): 24–31.

Doyle

(2003) Synthesis through meta-ethnography: Paradoxes, enhancements, and possibilities. Qualitative Research 3(3): 321–344.

Fossey

Harvey

(2010) Finding and sustaining employment: A qualitative meta-synthesis of mental health consumer views. Canadian Journal of Occupational Therapy 77(5): 303–314.

France

Ring

Noyes

et al. (2015) Protocol-developing meta-ethnography reporting guidelines (eMERGe). BMC Medical Research Methodology 15: 103. doi: 10.1186/s12874-015-0068-0.

France

Ring

Thomas

et al. (2014) A methodological systematic review of what’s wrong with meta-ethnography reporting. BMC Medical Research Methodology 14: 119. doi: 10.1186/1471-2288-14-119.

10.

France

Wells

Lang

et al. (2016) Why, when and how to update a meta-ethnography qualitative synthesis. Systematic Reviews 5(1): 44. doi: 10.1186/s13643-016-0218-4.

11.

Frostad Liaset

Lorås

(2016) Perceived factors in return to work after acquired brain injury: A qualitative meta-synthesis. Scandinavian Journal of Occupational Therapy 23(6): 446–457.

12.

Gibbs

Boshoff

Stanley

(2015) Becoming the parent of a preterm infant: A meta-ethnographic synthesis. British Journal of Occupational Therapy 78(8): 475–487.

13.

Hammell

(2001) Using qualitative research to inform the client-centred evidence-based practice of occupational therapy. British Journal of Occupational Therapy 64(5): 228–234.

14.

Hammell

(2007) Quality of life after spinal cord injury: A meta-synthesis of qualitative findings. Spinal cord 45(2): 124–139.

15.

Hannes

Macaitis

(2012) A move to more systematic and transparent approaches in qualitative evidence synthesis: Update on a review of published papers. Qualitative Research 12(4): 402–442.

16.

Lee

Hart

Watson

et al. (2015) Qualitative synthesis in practice: Some pragmatics of meta-ethnography. Qualitative Research 15(3): 334–350.

17.

Malpass

Shaw

Sharp

et al. (2009) “Medication career” or “Moral career”? The two sides of managing antidepressants: A meta-ethnography of patients’ experience of antidepressants. Social Science & Medicine 68(1): 154–168.

18.

Moher

Liberati

Tetzlaff

et al. (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine 6: e1000097.

19.

Murray C and Stanley M (2015) Meta-synthesis demystified: Connecting islands of knowledge. In: Nayar S and Stanley M (eds) (2015) Qualitative Research Methodologies for Occupational Science and Therapy. New York: Routledge, 174–189.

20.

Murray C, Turpin M, Edwards I, et al. (2015) A qualitative meta-synthesis about challenges experienced in occupational therapy practice. British Journal of Occupational Therapy 78(9): 534–546.

21.

Nayar S and Stanley M (2015) Beginning conversations. In: Nayar S and Stanley M (eds) Qualitative Research Methodologies for Occupational Science and Therapy. New York: Routledge, 1-7.

22.

Noblit

Hare

(1988) Meta-ethnography: Synthesizing Qualitative Studies, Thousand Oaks, CA: SAGE.

23.

Pound

Britten

Morgan

et al. (2005) Resisting medicines: A synthesis of qualitative studies of medicine taking. Social Science & Medicine 61(1): 133–155.

24.

Shamrock

Gray

Cameron

et al. (2016) Physical disabilities in Asia: A meta-synthesis of qualitative literature written in English. British Journal of Occupational Therapy 79(9): 565–573.

25.

Stanley M and Nayar S (2015) Deepening understandings. In: Nayar S and Stanley M (eds) Qualitative Research Methodologies for Occupational Science and Therapy. Oxford: Routledge, 8–20.

26.

Strike

Posner

(1983) ‘Types of Synthesis and their Criteria’. In: Ward

Reed

(eds) Knowledge Structure and Use: Implications for Synthesis and Interpretation, Philadelphia: Temple University Press, pp. 343–362.

27.

Thorne

(2017) Metasynthetic madness: What kind of monster have we created? Qualitative Health Research 27(1): 3–12.

28.

Thorne

(2015) Qualitative metasynthesis: A technical exercise or a source of new knowledge? Psycho-Oncology 24(11): 1347–1348.

29.

Tong

Flemming

McInnes

et al. (2012) Enhancing transparency in reporting the synthesis of qualitative research: ENTREQ. BMC Medical Research Methodology 12: 181. doi: 10.1186/1471-2288-12-181.

30.

Toye

Seers

Allcock

et al. (2013) ‘Trying to pin down jelly’ – Exploring intuitive processes in quality assessment for meta-ethnography. BMC Medical Research Methodology 13: 46. doi: 10.1186/1471-2288-14-80.

31.

Toye

Seers

Allcock

et al. (2014) Meta-ethnography 25 years on: Challenges and insights for synthesising a large number of qualitative studies. BMC Medical Research Methodology 14: 80. doi: 10.1186/1471-2288-14-80.

32.

Tricco

Soobiah

Antony

et al. (2016) A scoping review identifies multiple emerging knowledge synthesis methods, but few studies operationalize the method. Journal of Clinical Epidemiology 73: 19–28.

33.

Williams S and Murray C (2013) The experience of engaging in occupation following stroke: A qualitative meta-synthesis. British Journal of Occupational Therapy 76(8): 370–378.

34.

York

Wiseman

(2012) Gardening as an occupation: A critical review. British Journal of Occupational Therapy 75(2): 76–84.