Qualitative evidence synthesis services

Trials can establish that an intervention works on average and still say nothing about why patients abandon it, why clinicians resist it, or why it succeeds in one setting and fails in another. Those are questions about experience, acceptability, and implementation, and the evidence that answers them is qualitative. A qualitative evidence synthesis brings that evidence together with the same discipline a quantitative review applies to trials: a pre-specified question, a comprehensive and documented search, screening in duplicate against explicit inclusion and exclusion criteria, formal appraisal, and a synthesis method named and justified in advance rather than improvised at the write-up stage.

The searching itself differs from a trial-focused review, because qualitative studies are indexed inconsistently and titles rarely declare their methods. We build sensitive strategies with qualitative filters across the right databases, drawing on our literature search expertise, and run dual independent screening so the included set is as defensible as the synthesis built on it.

There is no single correct method, only a correct match between the question, the evidence base, and the intended use. We recommend one of four established approaches and justify the choice in the protocol.

For meta-ethnography in particular, the reach towards a line of argument, a new explanation that no single study contains, is the whole point of the method, and it is where weak syntheses fail. We develop it explicitly: reciprocal and refutational translations are tabulated, the argument is built from them step by step, and the audit trail shows how each interpretation traces back to the primary studies.

1. 1

   Protocol and method selection

   We frame the question, select and justify the synthesis method, and pre-specify the appraisal tool, the sampling approach if the yield is very large, and the confidence-rating plan.

2. 2

   Search, screen, and appraise

   A sensitive multi-database search with qualitative filters, dual independent screening, and CASP appraisal of every included study, recorded per criterion.

3. 3

   Extract and code the findings

   Findings and supporting quotations are extracted verbatim and coded systematically, with the codebook and coding decisions documented for the audit trail.

4. 4

   Synthesise and develop the argument

   Codes are built into themes, translations, framework charts, or aggregated statements according to the chosen method, with interpretation pushed beyond summary where the method allows.

5. 5

   Rate confidence and report

   Each synthesised finding receives a GRADE-CERQual assessment across its four components, and the review is written to ENTREQ, with eMERGe followed for meta-ethnography.

Appraisal in qualitative synthesis is about the trustworthiness of each study’s findings, not a hunt for statistical bias. We appraise every included study with the Critical Appraisal Skills Programme qualitative checklist, recording a judgement and justification per criterion, and we report how appraisal influenced the synthesis, whether through sensitivity analysis, weighting of contribution, or transparent discussion. Those appraisals then feed the methodological limitations component of GRADE-CERQual, alongside coherence, adequacy of data, and relevance, producing a confidence rating for every synthesised finding. The result is a summary of qualitative findings table a guideline panel can act on, the qualitative counterpart of the certainty tables in our GRADE certainty-of-evidence service.

Qualitative syntheses are examined on their transparency, and the reporting standards exist precisely because early meta-syntheses were criticised as unauditable. We write to ENTREQ, the statement for Enhancing Transparency in Reporting the synthesis of Qualitative research, covering the search, screening, appraisal, coding, and synthesis decisions, and to eMERGe for meta-ethnographies, which adds the translation and line-of-argument stages. The coding itself follows the discipline set out in our guide to coding data in qualitative reviews, so every theme in the final synthesis can be traced back through the codebook to the words of the primary studies.

Many commissioners need both strands: a pooled estimate of effect and an explanation of how the intervention behaves in the real world. In a mixed-methods review, we run the qualitative synthesis alongside a quantitative meta-analysis and integrate the two, typically by using the qualitative findings to explain heterogeneity, interpret uptake and adherence, or test whether the factors participants describe match the moderators the numbers suggest. Where a broader programme of work spans several review types, our evidence synthesis services page sets out how the pieces fit together. Integration is planned in the protocol rather than bolted on at the end, so the two strands answer one coherent question instead of sitting side by side as separate reports.

Everything is delivered done-for-you and audit-ready. The synthesis itself arrives as a manuscript-quality document, and behind it sits the full working record: the search strategies, the screening decisions, the appraisals, the codebook, and the development of every theme or translation. Nothing in the final findings is left as an assertion; each synthesised statement can be traced back through the audit trail to the primary studies that support it, which is exactly the traceability that examiners, editors, and guideline panels test.

• A protocol naming and justifying the synthesis method
• A documented, reproducible search with qualitative filters, screened in duplicate
• CASP appraisals for every included study, recorded per criterion
• The full audit trail: codebook, translations or framework charts, and theme development
• GRADE-CERQual confidence ratings and a summary of qualitative findings table
• A manuscript-ready synthesis reported to ENTREQ, and eMERGe where applicable

We quote a fixed fee shaped by the breadth of the question, the expected yield and how many studies survive to synthesis, the method chosen, since a meta-ethnography demands deeper interpretive work than a framework synthesis, and whether a quantitative strand runs alongside. Send us the question, the intended use, and any protocol or search work already done, and we will recommend a method honestly, including telling you if your question is better served by a different review type, and return a fixed quote with no obligation to proceed.