The conduct and reporting of mediation analysis in recently published randomized controlled trials: results from a methodological systematic review

Abstract

Objectives: To describe the methodological characteristics of mediation analyses (MAs) reported in recent randomized controlled trials (RCTs) and to propose recommendations on the planning, conduct, and reporting of MAs in practice. Study Design and Setting: We conducted a systematic review by searching MEDLINE (January 1, 2017, to December 1, 2018) for all reports of RCTs or secondary analyses of previously published RCTs that reported a MA. Two reviewers independently screened the title, abstracts, and full texts of the identified reports and extracted the data from the 98 eligible studies. Results: MAs were nearly always (96%) based on a traditional mediation approach. Most studies did not report a sample size calculation for the MA (96%) or assess potential treatment-by-mediator interactions (96%). In 53% of studies, mediators and outcomes were simultaneously measured. In 57% of studies, mediator-mediator and mediator-outcome confounders were adjusted for in the analysis, although adjustment was often limited to few potential confounders. About 30% of studies discussed the assumptions underlying the MA. Conclusion: The conduct and reporting of MAs remained quite heterogeneous in practice. Future MAs could benefit from a consensusbased planning, conduct, and reporting guideline for MA.

Development of A Guideline for Reporting Mediation Analyses (AGReMA)

“…The number of studies using mediation analysis to understand the mechanisms of health exposures and interventions is increasing [1,2]. However, the reporting of these studies remains heterogenous and incomplete [1,[5][6][7][8][9][10][11]. Methods for synthesising mediation studies are also under development so a reporting guideline is timely to help reduce reporting heterogeneity and facilitate the synthesis and pooling of mediation studies.…”

“…Assessing the quality of reporting will provide important insights into the prevalence of potential sources of bias in studies that use mediation analysis and on reporting items to be considered for the eventual guideline [18]. Vo et al 2019 reviewed the reporting quality of randomised trials that use mediation analysis [11]. We will systematically review the reporting quality of non-randomised observational studies that have used mediation analysis to investigate causal mechanisms.…”

“…The results of this review will provide evidence on the quality of reporting of non-randomised observational studies that use mediation analysis. These findings will be used alongside existing systematic reviews [5,10,11] to inform the Delphi study (stage 2), and the consensus meeting (stage 3) to decide on the core reporting items for AGReMA. The findings of this review will be submitted for publication.…”

“…While the use of mediation analysis has become increasingly common in recent years, there is growing recognition that the reporting of studies that use mediation analysis to investigate causal mechanisms of healthcare interventions is heterogenous and often incomplete [1,[5][6][7][8][9][10][11]. A recent overview of reviews across 11 health care fields and 26 healthcare conditions showed that mediation studies often did not report effect sizes and precision estimates (14/ 54, 26%), the theoretical rationale for the mechanism being tested (7/54, 13%), and essential details of the analytical techniques (4/54, 7%) [1].…”

“…A recent overview of reviews across 11 health care fields and 26 healthcare conditions showed that mediation studies often did not report effect sizes and precision estimates (14/ 54, 26%), the theoretical rationale for the mechanism being tested (7/54, 13%), and essential details of the analytical techniques (4/54, 7%) [1]. Reviews of primary mediation studies also show that the reporting of effect estimates and assumptions are varied across the literature [5] and that most mediation analyses of randomised trials did not report sample size calculations [11]. These limitations stifle clinical application, reproducibility, and evidence synthesis.…”