In the process of tackling incomplete and inadequate reporting of mediation studies, Aidan Cashin & Hopin Lee share the findings of their recent publication identifying items for consideration in A Guideline for Reporting Mediation Analysis (AGReMA).
Blog entry written on: ‘Items for consideration in a reporting guideline for mediation analyses: a Delphi study’ (bmjebm-2020-111406)
Authors: Aidan Cashin & Hopin Lee.
Why mediation analysis
Mediation analysis is a widely used method for investigating how interventions and exposures in randomised controlled trials and observational studies have an effect on health outcomes. Although the publication of mediation studies is increasing exponentially (Figure 1), recent systematic review evidence has identified problems with the completeness and transparency of reporting. Inadequate and incomplete reporting impedes appraisal, application and synthesis of research findings, and ultimately contributes to avoidable research waste.
We formed an international working group to develop A Guideline for Reporting Mediation Analysis (AGReMA) to tackle the problematic reporting of mediation studies. In our recent publication in BMJ Evidence-Based Medicine, we used a Delphi procedure to seek expert opinions and ratings on items that should be considered in a consensus meeting to consolidate a core set of recommended reporting items for mediation analyses.
Figure 1. An increasing trend in the publication of mediation analysis research. The figure displays publication counts from a search in Google Scholar for articles (excluding patents and citations) with “mediation analysis” anywhere in the text. The search was conducted on 21/01/2020.
What did we do?
We recruited 19 experts in the methodology and application of mediation analysis to take part in a Delphi panel. The panel were asked to rate the importance of a list of items for inclusion in a guideline for reporting mediation analyses, and to contribute additional items where possible and provide suggestions for item refinement.
What did we find?
All panellists contributed to all three Delphi rounds. The panel reached consensus on 34 unique reporting items on the study design, analytic procedures and effect estimates relevant to mediation analyses, with three items rated ‘optional’. The panel also added one additional item and provided 60 qualitative comments for item refinement and prioritisation.
What does this mean?
The Delphi procedure is an important step that engages key stakeholders in the process of developing a reporting guideline. The list of 34 items that reached expert consensus from this Delphi were recently discussed at a consensus and guideline development meeting as part of the wider AGReMA initiative. The development and implementation of AGReMA will improve the standardisation, transparency, and completeness in the reporting of studies that use mediation analysis to understand the mechanisms of health interventions and exposures.
Aidan G. Cashin, University of New South Wales, Australia.
Short bio: Aidan is a Doctor of Philosophy (PhD) candidate in the Centre for Pain IMPACT at Neuroscience Research Australia, The University of New South Wales.
COI: Aidan is funded by the University of New South Wales Prince of Wales Clinical School Postgraduate Research Scholarship and a NeuRA PhD Candidature Supplementary Scholarship. Aidan is a Catalyst for the Berkeley Initiative for Transparency in the Social Sciences.
Hopin Lee, University of Oxford, United Kingdom
Short bio: Hopin is a postdoctoral research fellow at the Centre for Statistics in Medicine at the University of Oxford and the School of Medicine and Public Health at the University of Newcastle.
COI: Hopin has received funding from the Australian National Health and Medical Research Council and the Berkeley Initiative for Transparency in the Social Sciences.
The views and opinions expressed on this site are solely those of the original authors. They do not necessarily represent the views of the BMJ and should not be used to replace medical advice. All information on this blog is for general information, is not peer-reviewed, requires checking with original sources and should not be used to make any decisions about healthcare. No responsibility for its accuracy and correctness is assumed by us, and we disclaim all liability and responsibility arising from any reliance placed on such commentary or content by any user or visitor to the Website, or by anyone who may be informed of any of its content. Any reliance you place on the material posted on this site is therefore strictly at your own risk.