“What the Meta Is Going On?”—A Scoping Review of the Different Methods and Methodology of Qualitative Synthesis

Qualitative evidence synthesis has become an essential component of evidence-based practice, particularly for lifespan development and mental-health research, where understanding lived experiences, perceptions, and contextual factors is critical. While searching for qualitative studies is a foundational step in any synthesis, and appraisal of study quality is variably applied depending on philosophical stance and review purpose, the process of synthesis itself remains the most conceptually challenging and least standardized aspect of qualitative reviews. Currently, there is an abundance of terms used to describe the synthesis process; for instance, the term meta is accompanied by a high number of variations illustrating different synthesis approaches. This includes, but is not limited to, Meta-ethnography [1], Meta-study [2], Meta-synthesis [3], Meta-aggregation [4], and Meta-interpretation [5]. The term ‘synthesis’ faces similar problems due to the sheer number of terms used in academic literature. For instance, the following terms have been identified describing approaches to synthesis, including interpretive synthesis [6], aggregative synthesis [7], concept synthesis [8], narrative synthesis [9], textual narrative synthesis [10], thematic synthesis [11], and translation synthesis [12].

Compared to empirical methods for synthesis, review-based approaches are often not linked closely and clearly linked to accompanying methodologies or philosophies. For instance, thematic analysis could be used within a study situated within a subtle realist world view and using interpretive hermeneutic phenomenology; alternatively, social constructivist grounded theory is clearly associated with pragmatism [13] and has very specific, clear steps and considerations for analysis. Further empirical approaches could be named as such. The important point from this is that empirical approaches are easily operationalizable by knowing a methodology and philosophy, whereas the development of review synthesis approaches does not always appear to be. The inability to link methodology with synthesis approaches consistently means there is a greater need to understand the philosophical foundation of the synthesis approach to ensure the product of synthesis is what is intended by the approach. Recently, Soundy and Heneghan [14] identified an approach that combines an empirical method of analysis (social constructivist grounded theory) with a well-established review technique (meta-ethnography), which may be one way of answering this problem and retaining or honouring key analytical strategies [15]. However, the extent to which this problem is considered and identified in past reviews on synthesis approaches e.g., ref. [16] is limited. A review that examines the underlying philosophical considerations to the meta-synthesis approach would be useful to allow scholars to establish these links and ensure a higher standard of practice for reviewing articles related to lifespan development and mental-health research.

Given the above, it is possible that reviewers of qualitative research are confused by an extremely high number of terms that don’t necessarily clearly fit with a methodology or philosophy. In addition to this, there appear to be several articles that talk across synthesis techniques but, in themselves, may not provide enough guidance to enable the different named approaches to be operationalised. For instance, Hannes and Macaitis [17], Nye [18] do an excellent job of introducing approaches, but the ability to operationalise each may be limited by what literature is available. A previous scoping review by Tricco et al. [8] highlighted that many synthesis methods described in the literature were difficult to operationalise. A more recent scoping review by Perlman et al. [19] identifies an excellent consideration of some of the processes; however, the details considered around the synthesis methods are lacking. This lack of clarity around methods of synthesis poses challenges for researchers attempting to apply these methods rigorously and transparently. Given this, it is important that an understanding of the aims of research that details different approaches is better understood.

To illustrate the relevance of this review to lifespan development and mental-health research, it is important to anchor associate it within a specific area of psychology. For example, qualitative studies exploring adolescents’ lived experiences of depression would represent a good health psychology example topic. This topic includes interesting domains of psychology, including identity formation, peer relationships, emotional regulation, and developmental transitions. These domains are contextual and interpretive and provide opportunity for different approaches to be selected and chosen based on the overall outcome of the review which could be to identify and aggregate descriptions from papers, to use an established framework found in an area like emotional regulation to represent results of a review, to develop a new theory by examining experiences identified in papers or to reveal important and contextual difference that may exist across cultures and populations. Further to this, the area has been selected as recent examples of different review methodologies are available e.g., [20,21,22,23,24].

Due to the proliferation of new synthesis techniques, there is a need to revisit and update the landscape of qualitative synthesis approaches with a special focus on which techniques may be possible to operationalise. This scoping review aims to identify and map the range of qualitative synthesis methods currently described in the literature. Identify those which could be operationalised and explore the approaches for association to methodology as well as philosophical foundations that underpin these approaches.

A scoping review methodology was selected following the framework proposed by PRISMA [25]. This design is appropriate for mapping key concepts, identifying gaps, and clarifying definitions in complex and heterogeneous fields.

The eligibility criteria are based on the acronym PCC.

Qualitative synthesis articles that have been written and published by academics interested in synthesis methodology and produce articles related to synthesis.

Articles that identify a methodological approach to conducting qualitative synthesis and identify some considerations towards the steps or methods involved in that approach. For this review, qualitative synthesis methods are defined as approaches that focus on bringing together research data that represent, or attempt to honour, participants’ voices, behaviours, or actions. Such data must not have been restricted to numerical categories or scales, nor converted for the purposes of statistical or quantitative methods of synthesis. Approaches that do not focus on synthesising qualitative data using qualitative analysis procedures fall outside this definition. Any type of qualitative approach was acceptable, including variations that may be associated with different methodologies or philosophical positions. Approaches that include the analysis of qualitative data using case studies were included. Worked examples of the synthesis procedures were used during the process of analysis. It was identified that the procedure was commonly understood and recognized, and could be operationalised as an approach.

Chapters and books were only included if the methodological named approach could not be accessed via an article. Handbooks or commissioned reports that refer to articles when identifying synthesis and do not produce their own version of synthesis were excluded. Articles that identify an analysis approach designed for empirical studies. For instance, content analysis articles or thematic analysis articles would only be included if specifically designated for the purpose of being included as a step in a qualitative literature review. Quantitative meta-analyses or mixed methods reviews that are not solely focused on qualitative synthesis were excluded. For instance, meta-narrative (a qualitative and mixed methods approach), critical interpretive synthesis (an approach which draws on qualitative principles but is a methodological hybrid used for qualitative and quantitative sources), realist synthesis (a mixed methods approach designed to identify how and why interventions work), grounded meta-analysis (a methodological hybrid approach that can extend beyond qualitative data), qualitative comparative analysis (an approach which combines qualitative case knowledge with formal mathematical, and quasi-quantitative logic) and case survey (an approach which converts case studies to quantitative analysis) were excluded as approaches. Further, for the purpose of this review, umbrella review approaches were excluded. The main reason for this is because this could introduce a different set of literature and some approaches, like mega-ethnography, have just used the phases proposed by meta-ethnography e.g., [26].

Only English language studies were included. Publications from any discipline, provided they focus on synthesizing qualitative data. Where multiple versions of the same article or approach exist by the same authors, only one of those articles was included. This was to avoid duplication and repetition; for instance, multiple references to the emerging guidelines were excluded, and one was selected to represent the approach.

A comprehensive search was conducted across five electronic databases, including MEDLINE, CINAHL, PsycINFO, Scopus, and Web of Science. The first 30 pages of electronic search engines Google Scholar and ScienceDirect were searched. Grey literature and reference lists of included studies were also screened using ProQuest and GreyMatters. Search terms for electronic databases included combinations of: “qualitative synthesis” or “realist synthesis” “meta-ethnography” or “synthesis” or “meta-synthesis” or “framework synthesis” or “descriptive synthesis” or “narrative synthesis” or “critical interpretive synthesis” or “meta interpretation” or “grounded theory synthesis” AND “review” or “overview” or “methods” AND “qualitative” or “framework” or “approach”. A key and critical stage of citation chasing of included articles occurred to elaborate further and consider the approaches identified. The search strategy was intentionally methodological and not topic-specific. To demonstrate practical relevance for lifespan mental-health research, we provide examples drawn from adolescent mental health. These examples were selected post hoc for illustration only and were not used as inclusion criteria above or as keywords within the search.

Titles and abstracts were screened by the study author. Covidence was used to assist in the storage of identified texts and allow assessment of inclusion criteria to occur with two reviewers. Full texts were retrieved for potentially relevant studies and assessed against inclusion criteria. Discrepancies were resolved through discussion or consultation with a third reviewer.

For any included study, the author developed a standardized data extraction form that captured the following sections of information: name of author, year, aim of paper, and type of synthesis method used.

Descriptive statistics were used to summarize the frequency and distribution of synthesis methods. Graphs were used to document the frequency of terms identified to describe synthesis. The summary aims of the articles were aggregated for the most common aims, with descriptive statistics added. The definition of terms was developed by aggregating study definitions. Further detailed descriptive analysis was undertaken for the most common 10 approaches to synthesis. This included identifying study originators or the earliest reference from results, if frameworks were used, the philosophical foundation (e.g., interpretivist, constructivist) of the approach, identification of synthesis approaches used to create operational guidance, and examples of application.

A total of 806 articles were identified, of which 54 were selected [3,4,5,7,8,9,11,15,16,17,18,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69]. Figure 1 provides the PRISMA flow diagram.

Of the included studies, the UK and the USA had produced the most work centered around qualitative synthesis. This is identified as follows UK: 22, USA: 10, Canada: 5, Australia: 4, Germany: 3, France: 2, Ireland: 2, The Netherlands: 2, Norway: 1, Taiwan: 1, Belgium: 1 and Finland: 1. In addition to this, most of the research (n = 26) was published between 2006 and 2018 when around 3.5 papers/year were produced. Between 1975–1999, there were 2 identified; between 2000–2009, 19; between 2010–2019, 25; and 8 from 2020 onwards.

The main aims of the articles included can be considered in Table 1. The three most common aims mentioned were; (a) to develop an innovative method, (b) to provide an overview of approaches, and (c) guidance for conducting synthesis.

The results are now split into an analysis of qualitative methodologies and methods.

A total of 14 review methodologies were identified, see Figure 2. The three most frequently mentioned types of methodology were: meta-ethnography (n = 22), meta-synthesis (n = 15), framework synthesis (n = 11). Figure 2 identifies the frequency of the different types of review methodologies.

The methodological approaches are presented in Table 2. All approaches named in the table aim to produce higher order insights that transcend individual studies. Many approaches identify the need for iterative analysis and place emphasis on contextual meaning. The most developed approaches which both utilise one name for synthesis were meta-ethnography and meta-synthesis. Meta-ethnography, which has two frameworks developed [15,46] and illustrates a consistent use of steps for undertaking synthesis as well as specific terms used, which have all been derived from the original text [1]. Meta-synthesis had the most articles that identified stages of the approach, although these stages were not consistent, and referenced a similar process.

Readers should be mindful that meta-ethnography, thematic synthesis, and narrative synthesis have clear, agreed stages that help with the process of operationalizing the methodology based solely on the methodological article. In contrast, meta-synthesis and meta-interpretation are more flexible and ideographic. Framework synthesis and meta-study rely on formal frameworks and theories to inform the approach.

Table 3 identifies the philosophical assumptions of each approach, as well as indicating specific findings about the approaches. It is important to locate the ontology and epistemology of approaches, especially when considering the product of the synthesis and what the synthesis is attempting to honor. Inductive theory building appears central to meta-synthesis and meta-ethnography. Whereas pluralism, historical context, and reflexivity are important to other approaches like meta-study. Figure 3 provides a decision aid tool for researchers considering which approach may be most useful.

A total of 8 qualitative synthesis methods were identified, see Figure 3. The three most used qualitative synthesis approaches were meta-grounded theory (n = 10), meta-aggregation (n = 5), and content analysis (n = 5).

The approaches identified represent a broad spectrum of synthesis methods, ranging from five aggregative methods (see Figure 3 and Table 4) which aim to summarise and generalize to inform practice (e.g., meta-aggregation, qualitative meta-summary, content analysis) to ten interpretive methods see (Figure 4 and Table 5) which aim to interpret findings or add new findings as an approach (e.g., interpretive synthesis, meta-grounded theory). Meta-aggregation and content analysis have been designed to be included within and follow systematic review protocols with clearly defined steps. Whereas meta-grounded theory and interpretive synthesis are more iterative and flexible methods.

This scoping review mapped the landscape of qualitative synthesis methodologies and methods. This revealed a clear selection of methodologies with accompanying guidance documents, as well as a selection of methods, which at times require further consideration if the methods are to be successfully operationalized. The findings underscore several critical issues that merit further discussion, particularly around the operationalisation, philosophical foundations, and terminological clarity of synthesis methods.

Most synthesis development occurred between 2006 and 2018, with the UK and USA leading in publication output. This suggests a period of methodological innovation and consolidation, possibly driven by increased interest in qualitative evidence synthesis within health and social sciences. However, the decline in recent publications may indicate either saturation or a shift toward refining existing methods rather than introducing new ones.

One of the most striking findings is the sheer volume and variation of terms used to describe qualitative synthesis approaches. Across 54 papers, 14 methodologies, 5 methods of aggregative synthesis, and 10 of interpretive synthesis. This terminological diversity had previously added layers of confusion for researchers when decided on a qualitative approach to synthesis. The lack of standardised naming conventions may hinder methodological transparency and reduce the accessibility of synthesis techniques. Research has begun to categorise synthesis into types; for instance, the aggregative and interpretive types provide a useful focus e.g., [92]. However, the current paper identifies methodologies that provide a useful basis for decision making.

While many synthesis approaches are conceptually rich, only a subset appear to be sufficiently developed to allow for practical application. Notably, meta-ethnography, meta-synthesis, framework synthesis, and meta-study were identified as more easily operationalisable due to the presence of frameworks, agreed stages, and detailed guidance. Meta-ethnography, for example, benefits from two established frameworks [15,46] and a consistent set of analytical steps derived from Noblit and Hare [1]. In contrast, other approaches, such as grounded theory synthesis or thematic synthesis, often lack unified guidance, making their application more variable and dependent on reviewer interpretation. This disparity in operational clarity reflects a broader issue: many synthesis methods are introduced or discussed conceptually but lack sufficient detail to be implemented rigorously. This aligns with previous reviews e.g., [8,19], which noted the difficulty of operationalising synthesis techniques due to limited methodological elaboration. It is important to note, however, that the evaluation of operationalisability is based on published procedural guidance rather than empirical evidence of implementation fidelity in applied reviews.

Understanding the philosophical roots of each approach is essential, as these foundations influence not only the synthesis process but also the nature of the findings produced. One useful distinction of methods is by interpretive methods, such as qualitative meta-synthesis and critical interpretivist synthesis, versus aggregative synthesis, grounded in realism and pragmatism, which often seeks to summarize findings for practical application, often avoiding reinterpretation. This is true for both methodologies and methods. One danger of interpretive approaches is that multiple syntheses on the same topic may create nuanced but potentially overlapping insights. This raises concerns about redundancy and the risk of rewording existing knowledge rather than generating genuinely novel understandings.

Given the findings from the current scoping review, several implications emerge:

The methodological distinctions identified in this review can be illustrated by presenting worked examples of reviews that have synthesised qualitative studies on adolescent depression. Based on Figure 5, examples are given by the four purpose statements and outcomes of reviews (generation of theory, conceptual insight, actionable findings for policy or practice, and an analysis of methods, theory, and findings together). Examples are considered against guidelines, and where appropriate, deviations from guidelines are reported to help establish best practice for operationalisability.

Two examples are given around the use of meta-ethnography. Example 1, Lucas et al. [22] developed a meta-ethnography focusing on young women’s perception of mental health during and after pregnancy. A process for identifying participant quotes (first-order translations), author quotes (second-order translations), and their own interpretation across studies (third-order translations) is given. A thematic representation, including perceptions, experiences of support, and social and economic experiences, is identified, and, to aid context, a focus on specific countries is given. Alongside the interpretations and themes, a line of argument analysis is developed (going beyond the data). This suggests that different levels the experience of these women should be considered in terms of the individual personal and physical exercise, the relationship influence, the economic security, the social surveillance, and narrative reparation.

Sub-titles are not listed correctly when compared against the eMERGe guidelines [46]. Detail is missing in both the protocol and the published review. For example, within the synthesising translations sub-title, eMERGe calls for reporting the methods used to develop overarching concepts and potential alternative interpretations—neither of which is provided. Some older references are used to justify stages of extraction and synthesis [30,93], which may have limited or influenced the choice of what to include, despite the authors citing the eMERGe guidelines. A line-of-argument synthesis is presented in the results but not described in the methods, meaning that how it was identified and tested is not clear. No supplementary material or audit trail is provided. No consideration of certainty assessment (e.g., ConQual [94] or CERQual [95]) is given.

Example 2, the basis for a social constructivist meta-ethnographic review could be considered by identifying a specific past area of research such as adolescent coping with mood disorders, include related studies e.g., [96] as well as other searching, that could include past theories and conceptualizations of coping e.g., [37] to allow the iterative process of theory development to take place with the emphasis on developing a substantive theory through identifying a process, model and theory.

Authors must adhere to the social constructivist meta-ethnographic framework [15]. This means appreciating a pragmatist ontology and relativist epistemology, and understanding that the author is an embodied and situated knowledge producer, open to multiple perspectives, including an appreciation of the cultural, historical, social, and environmental factors that shape their inquiry. Attention must be given to aspects such as inequity, injustice, silenced voices, and taken-for-granted meanings, as this ensures a stance suitable for critical inquiry.

Finally, key phases of immersive reading, coding, memoing, focused coding, and development of categories are required and should be clearly presented. Categories, models, or processes identified during the synthesis need to be challenged through a process of iterative idea generation, which involves techniques adopted from social constructivist grounded theory (including, but not limited to, constant comparison, theoretical sampling, and abductive reasoning). Unlike traditional meta-ethnography, multiple searches are possible, and an audit trail is essential (see [74] for a worked example with supplementary material).

Three examples are given that provide conceptual insight. Example 1, a systematic review according to JBI guidance and meta-aggregation study by Twivy et al. [23] was undertaken this which undertook a process the extraction of verbatim text from qualitative studies looking at the lived experience of depression. This review identified several findings (like codes). Focusing on those with the most support (using a ranked system normally). The ranking wasn’t mentioned in the study, but the following steps of grouping findings together to create categories (which need at least 2 findings) with the support of NVivo software [version number not given by paper]. Following these categories are brought together in a group, which was identified as synthesised findings. Three synthesized finding areas were identified, including causes (3 categories), symptoms (8 categories), and coping (5 categories). The resultant synthesis identifies different categories and locates all references that discuss them. This means the work was aggregative and did not seek to interpret the findings.

The use of meta-aggregation within Example 1 references the JBI Manual and related guidance documents, including the JBI checklist [51], the extraction-of-findings tables supported by JBI, and the JBI credibility rating scales (identified as unequivocal, credible and unsupported). This aligns with the guidelines [51] and is further supported by reference to a previous worked example [97]. The authors define these ratings on page 757; however, even with the supplementary material, it is difficult to understand what “findings without question” (unequivocal rating) looked like versus findings that were “partially supported” (credible rating). Greater clarity regarding the steps following the extraction of results would be helpful, ideally in the form of an audit trail. Furthermore, JBI recommends a certainty assessment e.g., ConQual [94], which was not conducted.

Example 2: A thematic synthesis was undertaken by Broad et al. [21] to consider the transition from child mental health services to adult mental health services. The authors identify a thematic synthesis that first independently identified codes, which was followed by meetings to develop a consensus of codes and meaning, and this was completed by the main reviewer, identifying three (complex interplay of multiple concurrent transitions, balancing autonomy and the need for supports and factors impacting your experience of transition) over-arching themes. Each theme has no sub-themes, but example quotes to illustrate points.

The review references a protocol through PROSPERO and the use of the ENTREQ [98] framework and checklist to assist reporting. The authors identify a data analysis section and mention initial ‘content analysis’ followed by thematic synthesis. They identify independent coding to develop a consensus of codes and describe trustworthiness through independent analysis, triangulation of perspectives, and revision of the coding dictionary.

Whilst they acknowledge that their approach adapts methods from meta-ethnography and grounded theory, they do not specifically name these methods or include an audit trail that would allow a reader to follow the analytic processes. This review did not undertake a sensitivity analysis for the critical appraisal, nor did it describe the three stages of analysis using sub-titles or outline the generation of analytical themes as the final stage [60].

The risk with this review is that reporting a protocol and using a checklist such as ENTREQ may create the appearance of quality; however, the description and level of detail provided do not offer sufficient evidence to understand how the synthesis was guided or how it aligns with the stated approach.

Example 3, a meta-synthesis was undertaken by Achterbergh et al. [20] considering the experiences of loneliness among young people with depression. The synthesis is based on the work of Lachal et al. [3] and identifies 6 stages of analysis. The analysis process is identified as involving independent researchers importing all text related to loneliness. This was important into the Invivo software, and the independent researchers then identified what to include from this. A final database of extracted data was then coded independently to develop a coding framework which was then refined through a process of iteration and taxonomy of themes. The thematic synthesis is presented in a similar way to Broad et al. [21] but also included sub-themes. It should be noted that a breakdown of the stages identified by Lachal el al [3] is not provided for instance Lachal et al. [3] identifies active reading in stage 1 (which he identifies as appraising, familiarizing, identifying, extracting, organising, relating, mapping, stimulating comparing and verifying) and then two further stages which are not mentioned by Achterbergh et al. [20].

The authors of Example 3 introduce the term meta-synthesis first as a method based on interpretation, citing four methodological papers. They then identify thematic synthesis, which they state is influenced by meta-ethnography and grounded theory, citing meta-ethnographic texts before returning to the six-step methodological approach to meta-synthesis proposed by Lachal et al. [3]. The reference to different named approaches may confuse readers.

However, inspection of the methodological paper on meta-synthesis by Lachal et al. [3] shows that the steps of synthesis are indeed associated with other named approaches. Four steps of synthesis are identified: reading and re-reading each study (step 1); coding (step 2), linked to Interpretative Phenomenological Analysis [99]; grouping codes into categories and producing a hierarchical tree (step 3), linked to the meta-ethnographic step of translation; and, finally, developing analytical themes (step 4), which corresponds to the meta-ethnographic process of third-order interpretations and a line-of-argument synthesis.

Importantly, within the data analysis section of the article, these four steps are not mentioned. Instead, the authors describe the use of software, the coding of a dataset, and the comparison of codes among four researchers to develop a coding framework, which was then refined through an iterative process to produce a taxonomy of analytical themes. No supplementary file is available, and the protocol demonstrates deviations in analysis and terminology—for example, the protocol specifies percentages for reviewer agreement and names thematic analysis within it.

Quality markers include triangulation from researchers with diverse backgrounds [3], and Example 3 provides evidence of improving external validity through the use of a multidisciplinary team.

Example 1 of Analyzing Methods, Theory, and Findings (Meta-Study). One example of a meta-study is identified. Example 1 was a meta-study by Watkins et al. [80] which examined the meta-theory considering how theories were considered within the topic of interest around black male mental health and well-being (no adolescent example was available). The authors examined the meta-methods and identified the outcome measures and sampling approach procedures used, and finally, the meta-data analysis was able to consider new interpretations from the analysis of qualitative data, although this example identifies using meta-ethnographic principles for this final phase.

The example of meta-study does not present the six steps (1. formation of the team, objectives and theoretical lens; 2. identification and assessment of existing studies; 3. categorising data based on methodology, sample, publication year or other criteria; 4. analysis of methodologies and how they impact methods and presentation of findings; 5. examination of theory in primary studies and the implications for subsequent theory development; 6. critical synthesis of methodological directions, ideas, and implications) as identified by Paterson et al. [2]

The authors do, however, identify general stages including meta-theory (using Lenski’s [61]) standards for theory construction), meta-method (appraisal of individual studies in “search of proper ‘fit’ with stated research methods and their influence on research findings” (p. 208), as well as an appraisal of themes), meta-data analysis (referred to as the analysis of “processed data”, involving comparative analysis of individual studies and, in this instance, drawing on the meta-ethnographic approach to analysing text [1]), and a synthesis approach of meta-synthesis (described as the final step and involving a new interpretation of the phenomenon that accounts for the data, method, and theory). This was identified in the article as follows Paterson et al.’s [2] three-stage approach (recoding, organising lists of codes, and re-analysing original findings focusing on central study metaphors).

This synthesis process, without an audit trail, is extremely difficult to follow, and the mention of other approaches such as meta-ethnography, or the later reference in the results to “content and constant comparative analyses” (p. 316)—may confuse readers.

Two examples of synthesis that provide actionable findings are given. Example 1, a narrative synthesis is proposed by Ballesteros-Urpi et al. [84], which identifies a specific application using predefined stages. They identify a need for a preliminary synthesis, which will first produce a summary of mental health and adolescent recovery, including definitions, theories, conceptual and theoretical frameworks, domains, and dimensions. Studies will be grouped by setting and type, and work will be tabulated to identify where concepts are similar, different, and where overlap occurs, with reasons for these identified. This preliminary synthesis will be discussed and focus on common themes and sub-themes used to identify an initial conceptual framework. A secondary synthesis will follow this to develop the framework.

This example does not mention the expected four stages of narrative synthesis (developing a theory, preliminary synthesis, exploring relationships, and assessing the robustness of the synthesis) as identified by Popay et al. [9]. The worked example demonstrates some alignment with Popay et al. [9], including intentions to produce a preliminary synthesis and plans to develop a conceptual framework. However, Stage 1 (developing a theory) is not mentioned, and Stage 4 (assessing the robustness of the synthesis) is also omitted. Furthermore, no reference is made to tools or techniques such as grouping and clusters, idea webs, or textual descriptions, which are recommended by Popay et al. [9]. These omissions may make the synthesis process harder to follow.

Example 2, a framework analysis is undertaken by Viduani et al. [24] to understand the experience of depression in adolescents. The framework used in the analysis is based on two theories: illness representation theory [100] and ecological theory [101]. This enabled the researcher the ability to index the interactions identified in studies between individuals and their environment.

Example 2 references Brunton et al. [31]; however, it does not mention all the stages of framework synthesis (familiarisation, framework selection, indexing, charting, mapping, and interpretation) as sub-titles. The article describes indexing studies according to past theories and coding across broad areas (personal meaning of depression, contextual factors influencing depression, and the ways adolescents coped). Line-by-line coding is identified, followed by the development and description of themes.

Further description of familiarisation, charting, mapping, and interpretation, as well as examples of framework matrices, could be provided, as could clarification of the relationship between theoretical coding and emergent codes. As with other articles, a protocol was published, and supplementary materials are available; however, understanding the framework approach remains limited even with these additional resources.

Future research should aim to:

This review did not include books or handbooks that could capture relevant material. It is possible that this review has not captured all types of synthesis approaches. The review is limited by its focus on synthesis approaches and keywords used, and not the fuller or broader steps of review. Only the most elaborate papers from a researcher or research group on a synthesis approach were included. Because not every synthesis paper from each author was included, there may be some limitations in findings. Finally, the boundary for deciding which qualitative synthesis approaches to include meant that the current review shouldn’t be considered as an exhaustive list of synthesis and it may be possible to reconsider limits of eligibility to include other qualitative synthesis techniques that were currently excluded (see eligibility criteria).

Microsoft co-pilot was used to check for typographical errors.

A.S. designed and undertook all processes for this review.

No ethical review was undertaken.

Not applicable.

All data is avaliable via the supplementary material in references.

No funding is declared.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.