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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Causal mediation analysis can improve understanding of the mechanismsunderlying epidemiologic associations. However, the utility of natural directand indirect effect estimation has been limited by the assumption of noconfounder of the mediator-outcome relationship that is affected by priorexposure---an assumption frequently violated in practice. We build on recentwork that identified alternative estimands that do not require this assumptionand propose a flexible and double robust semiparametric targeted minimumloss-based estimator for data-dependent stochastic direct and indirect effects.The proposed method treats the intermediate confounder affected by priorexposure as a time-varying confounder and intervenes stochastically on themediator using a distribution which conditions on baseline covariates andmarginalizes over the intermediate confounder. In addition, we assume thestochastic intervention is given, conditional on observed data, which resultsin a simpler estimator and weaker identification assumptions. We demonstratethe estimator's finite sample and robustness properties in a simple simulationstudy. We apply the method to an example from the Moving to Opportunityexperiment. In this application, randomization to receive a housing voucher isthe treatment/instrument that influenced moving to a low-poverty neighborhood,which is the intermediate confounder. We estimate the data-dependent stochasticdirect effect of randomization to the voucher group on adolescent marijuana usenot mediated by change in school district and the stochastic indirect effectmediated by change in school district. We find no evidence of mediation. Ourestimator is easy to implement in standard statistical software, and we provideannotated R code to further lower implementation barriers.

Robust and Flexible Estimation of Stochastic Mediation Effects: A Proposed Method and Example in a Randomized Trial Setting