English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In randomized trials, researchers are often interested in mediation analysisto understand how a treatment works, in particular how much of a treatment'seffect is mediated by an intermediated variable and how much the treatmentdirectly affects the outcome not through the mediator. The standard regressionapproach to mediation analysis assumes sequential ignorability of the mediator,that is that the mediator is effectively randomly assigned given baselinecovariates and the randomized treatment. Since the experiment does notrandomize the mediator, sequential ignorability is often not plausible. TenHave et al. (2007, Biometrics), Dunn and Bentall (2007, Statistics in Medicine)and Albert (2008, Statistics in Medicine) presented methods that use baselinecovariates interacted with random assignment as instrumental variables, and donot require sequential ignorability. We make two contributions to thisapproach. First, in previous work on the instrumental variable approach, it hasbeen assumed that the direct effect of treatment and the effect of the mediatorare constant across subjects; we allow for variation in effects across subjectsand show what assumptions are needed to obtain consistent estimates for thissetting. Second, we develop a method of sensitivity analysis for violations ofthe key assumption that the direct effect of the treatment and the effect ofthe mediator do not depend on the baseline covariates.

Mediation Analysis Without Sequential Ignorability: Using Baseline Covariates Interacted with Random Assignment as Instrumental Variables