Guidance for umbrella reviews of observational studies: A scoping review

Abstract

BackgroundUmbrella reviews, or overviews of reviews, synthesize information using systematic reviews (SRs) as their unit of analysis. Although a formal guideline exists for reporting umbrella reviews of healthcare interventions (i.e. Preferred Reporting Items for Overviews of Reviews [PRIOR]), no formal guideline exists for conducting and/or reporting umbrella reviews of observational studies that examine epidemiological associations.ObjectiveTo review the existing guidance on conducting and/or reporting umbrella reviews of observational studies on epidemiological associations, as part of the process of developing a formal reporting guideline.MethodsWe reviewed the scoping review conducted in the context of PRIOR development and identified documents through forward citation search in PubMed, Scopus, and manual search in Google Scholar, Google Search up to December 22, 2024. Documents, regardless of format, were included if they provided guidance for conducting and/or reporting umbrella reviews of observational studies (including meta-research studies of their features). Title/abstract screening and data extraction were performed independently and in duplicate and summarized narratively by stages of the umbrella review process.ResultsThe search retrieved 4491 unique records, with 96 full texts assessed and eight documents included. These documents, published between 2014 and 2023, offered guidance across seven topic areas, but overall guidance on conducting and/or reporting is limited. These areas include the answerable questions, prerequisite considerations, the scope of umbrella reviews, searching for SRs, primary data collection, analysis, presentation, and assessing the certainty/quality of the body of evidence.ConclusionThere is a need for dedicated, practical, and evidence-based formal reporting guidelines for umbrella reviews of observational studies on epidemiological associations. This review lays the groundwork for developing the PRIOR-extension for such studies: the Preferred Reporting Items for Umbrella Reviews of Cross-sectional, Case-control, and Cohort Studies.