Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Correspondence
  • Published:

A reporting guideline for randomized trials and observational studies using mediation analysis: AGReMA

Nature Medicine volume 28, pages 432–434 (2022)Cite this article

Subjects

To the Editor — Over the past decade, the value of quantifying the mechanisms by which health interventions and exposures work has become increasingly apparent. National funding organizations such as the US National Institutes of Health and the UK National Institute for Health Research now advocate for mediation analysis to be routinely conducted in health technology assessments and comparative effectiveness studies. The number of publications that use mediation analysis is increasing, and it is becoming common practice in medicine, epidemiology, psychology and the social sciences1,2.

The findings from mediation analyses are used to advance theory, inform policy, optimize interventions and facilitate the implementation of policies and interventions to clinical and public health practice. For example, in 2019 in this journal, Niewczas et al. embedded a mediation analysis within an observational study to identify circulating inflammatory proteins that increased the risk of developing end-stage renal disease from chronic inflammation associated with diabetes3. The findings point to new therapeutic targets and new prognostic tests to identify individuals at risk of end-stage renal disease, as well as biomarkers to measure responses to treatment of diabetic kidney disease. Schlemm et al., in 2021, also used mediation analysis on data from a randomized trial to investigate the mechanisms through which thrombolysis treatment with recombinant tissue plasminogen activator is associated with improved functional outcome in acute ischemic stroke4. The findings provide evidence on how the treatment works and identify novel neuroimaging markers to assess treatment response.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

References

  1. Vanderweele, T. J. Explanation in Causal Inference: Methods for Mediation and Interaction (Oxford Univ. Press, 2015).

  2. Lee, H., Herbert, R. D. & McAuley, J. H. JAMA 321, 697–698 (2019).

    Article  Google Scholar 

  3. Niewczas, M. A. et al. Nat. Med. 25, 805–813 (2019).

    Article  CAS  Google Scholar 

  4. Schlemm, E. et al. Nat. Commun. 12, 2590 (2021).

    Article  CAS  Google Scholar 

  5. Cashin, A. G. et al. J. Clin. Epidemiol. 111, 60–68.e1 (2019).

    Article  Google Scholar 

  6. Vo, T. T., Superchi, C., Boutron, I. & Vansteelandt, S. J. Clin. Epidemiol. 117, 78–88 (2020).

    Article  Google Scholar 

  7. Lee, H. et al. JAMA 326, 1045–1056 (2021).

    Article  Google Scholar 

  8. Moher, D. et al. PLoS Med. 7, e1000217 (2010).

    Article  Google Scholar 

  9. Cashin, A. G. et al. BMC Med. Res. Methodol. 20, 19 (2020).

    Article  Google Scholar 

  10. Cashin, A. G. et al. BMJ Evidence-Based Med. 26, 106–106 (2021).

    Article  Google Scholar 

Download references

Acknowledgements

We thank the AGReMA group authors and acknowledge the contributions made by the Delphi panelists, the AGReMA international consensus meeting participants and the AGReMA external review experts. AGReMA was supported by project funding from the University of California, Berkeley, Initiative for Transparency in the Social Sciences, a program of the Center for Effective Global Action, with support from the Laura and John Arnold Foundation.

Author information

Authors and Affiliations

  1. Centre for Pain IMPACT, Neuroscience Research Australia, Sydney, NSW, Australia

    Aidan G. Cashin & James H. McAuley

  2. School of Health Sciences, Faculty of Medicine & Health, University of New South Wales, NSW, Sydney, Australia

    Aidan G. Cashin & James H. McAuley

  3. Centre for Statistics in Medicine, Nuffield Department of Orthopaedics Rheumatology and Musculoskeletal Sciences (NDORMS), University of Oxford, Oxford, UK

    Hopin Lee

  4. School of Medicine and Public Health, University of Newcastle, Newcastle, NSW, Australia

    Hopin Lee

Authors
  1. Aidan G. Cashin
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. James H. McAuley
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Hopin Lee
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

All authors conceived the idea for the manuscript. A.G.C. prepared the first draft of the manuscript. Critical edits, further direction and feedback have been attained from all co-authors.

Corresponding author

Correspondence to Hopin Lee.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cashin, A.G., McAuley, J.H. & Lee, H. A reporting guideline for randomized trials and observational studies using mediation analysis: AGReMA. Nat Med 28, 432–434 (2022). https://doi.org/10.1038/s41591-021-01674-2

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41591-021-01674-2

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing