Over 1000 terms have been used to describe evidence synthesis: a scoping review

Objective To inform the development of an evidence synthesis taxonomy, we aimed to identify and examine all classification systems, typologies or taxonomies that have been proposed for evidence synthesis methods.

Design Scoping review.

Methods This review followed JBI (previously Joanna Briggs Institute) scoping review methodology and was reported according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). Resources that investigated typologies, taxonomies, classification systems and compendia for evidence synthesis within any field were eligible for inclusion. A comprehensive search across MEDLINE (Ovid), Embase (OVID), CINAHL with Full-Text (EBSCO), ERIC (EBSCO), Scopus, Compendex (Elsevier) and JSTOR was performed on 28 April 2022. This was supplemented by citation searching of key articles, contact with experts, targeted searching of organisational websites and additional grey literature searching. Documents were extracted by one reviewer and extractions verified by another reviewer. Data were analysed using frequency counts and a basic qualitative content analysis approach. Results are presented using bar charts, word clouds and narrative summary.

Results There were 15 634 titles and abstracts screened, and 703 full texts assessed for eligibility. Ultimately, 446 documents were included, and 49 formal classification systems identified, with the remaining documents presenting structured lists, simple listings or general discussions. Included documents were mostly not field-specific (n=242) or aligned to clinical sciences (n=83); however, public health, education, information technology, law and engineering were also represented. Documents (n=148) mostly included two to three evidence synthesis types, while 22 documents mentioned over 20 types of evidence synthesis. We identified 1010 unique terms to describe a type of evidence synthesis; of these, 742 terms were only mentioned once. Facets that could usefully distinguish (ie, similarities and differences or characteristics) between evidence synthesis approaches were categorised based on similarity into 15 overarching dimensions. These dimensions include review question and foci of interest, discipline/field, perspective, coverage, eligibility criteria, review purpose, methodological principles, theoretical underpinnings/philosophical perspective, resource considerations, compatibility with heterogeneity, sequence planning, analytical synthesis techniques, intended product/output, intended audience and intended impact or influence.

Conclusion This scoping review identified numerous unique terms to describe evidence synthesis approaches and many diverse ways to distinguish or categorise review types. These results suggest a need for the evidence synthesis community to organise, categorise and harmonise evidence synthesis approaches and terminology.