EUVIMEDEUVIMED entdecken
Crossref

Bayesian Structured Mediation analysis with Unobserved confounders

Yuliang Xu iD, Shu Yang iD, Jian Kang iD

Biometrics · 09.04.2026 · 82 (2)

DOI: 10.1093/biomtc/ujag110 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, online erschienen am 17.06.2026. Verfasst von Yuliang Xu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics, University of Chicago, Chicago, IL 60637, United States. Hauptthemen laut MEDLINE: Mediation Analysis. Die Arbeit nennt 25 Literaturangaben. Gefördert von: National Science Foundation, National Institute of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We explore methods to reduce the impact of unobserved confounders on the causal mediation analysis of high-dimensional mediators with spatially smooth structures, such as brain imaging data. The key approach is to incorporate the spatial latent subject-specific spatial confounding effects, which influence the structured mediators, as unobserved confounders in the outcome model, thereby potentially debiasing the mediation effects. We develop BAyesian Structured Mediation analysis with Unobserved confounders (BASMU) framework, and establish its model identifiability conditions. Theoretical analysis is conducted on the asymptotic bias of the Natural Indirect Effect (NIE) and the Natural Direct Effect (NDE) when the unobserved confounders are omitted in mediation analysis. For BASMU, we propose a two-stage estimation algorithm to mitigate the impact of these unobserved confounders on estimating the mediation effect. Extensive simulations demonstrate that BASMU substantially reduces the bias in various scenarios. We apply BASMU to the analysis of fMRI data in the Adolescent Brain Cognitive Development study, focusing on four brain regions previously reported to exhibit meaningful mediation effects. Compared with the existing image mediation analysis method, BASMU identifies two to four times more voxels that have significant mediation effects, with the NIE increased by 41%, and the NDE decreased by 26%.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Mediation AnalysisBayes TheoremComputer SimulationMagnetic Resonance ImagingHumansAlgorithmsBrainModels, StatisticalConfounding Factors, EpidemiologicBias

Stichwörter der Autor:innen

Bayesian non-parametricbrain imagemediation analysisspatial structureunobserved confounders

Autor:innen und Institutionen

  • Department of Statistics, University of Chicago, Chicago, IL 60637, United States.
  • Department of Statistics, North Carolina State University, Raleigh, NC 27695, United States.
  • Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109, United States.

ORCID-Profile: Yuliang Xu, Shu Yang, Jian Kang

Förderung

  • National Science Foundation
  • National Institute of Health

Zitieren

APA

Xu, Y., Yang, S., & Kang, J. (2026). Bayesian Structured Mediation analysis with Unobserved confounders. Biometrics, 82(2). https://doi.org/10.1093/biomtc/ujag110

Vancouver

Xu Y, Yang S, Kang J. Bayesian Structured Mediation analysis with Unobserved confounders. Biometrics. 2026;82(2). doi:10.1093/biomtc/ujag110

BibTeX
@article{Xu2026,
  author = {Xu, Yuliang and Yang, Shu and Kang, Jian},
  title = {Bayesian Structured Mediation analysis with Unobserved confounders},
  journal = {Biometrics},
  year = {2026},
  volume = {82},
  number = {2},
  doi = {10.1093/biomtc/ujag110}
}
Kostenlos starten →