TY - JOUR T1 - Items for consideration in a reporting guideline for mediation analyses: a Delphi study JF - BMJ Evidence-Based Medicine JO - BMJ EBM SP - 106 LP - 106 DO - 10.1136/bmjebm-2020-111406 VL - 26 IS - 3 AU - Aidan G Cashin AU - James H McAuley AU - Sallie Lamb AU - Sally Hopewell AU - Steven J Kamper AU - Christopher M Williams AU - Nicholas Henschke AU - Hopin Lee Y1 - 2021/06/01 UR - http://ebm.bmj.com/content/26/3/106.abstract N2 - Objectives Mediation analysis is a widely used quantitative method for investigating how interventions and exposures in randomised controlled trials and observational studies have an effect on healthcare outcomes. This study aimed to assess the importance of items that should be considered in a consensus meeting aimed at developing a guideline for reporting mediation analyses.Design International online Delphi study.Participants International experts in the development and application of mediation analysis.Main outcome measures The Delphi panel were asked to rate the importance of a list of items for inclusion in a guideline for reporting mediation analyses. Thresholds for disagreement and consensus on importance for inclusion were specified a priori. We used the Research ANd Development/University of California Los Angeles appropriateness method to quantitatively assess the importance for inclusion and panel agreement.Results Nineteen expert panellists (10 female) from seven countries agreed to participate. All panellists contributed to all three rounds conducted between 10 June 2019 and 6 November 2019. The panel reached consensus on 34 unique reporting items for study design, analytic procedures and effect estimates, with three items rated ‘optional’. Panellists added one extra item and provided 60 qualitative comments for item refinement and prioritisation.Conclusion This Delphi study used a rigorous consensus process to reach consensus on 34 reporting items for studies that use mediation analysis. These results will inform a consensus meeting that will consolidate a core set of recommended items for reporting mediation analyses.The data set used and analysed during this study is available from the corresponding author on reasonable request. ER -