EUVIMEDScopri EUVIMED
Open access

A Calibrated Sensitivity Analysis for Weighted Causal Decompositions

Andy A Shen iD, Elina Visoki, Ran Barzilay, Samuel D Pimentel

Statistics in Medicine · 2025

DOI: 10.1002/sim.70010 ↗

At a glance

Journal article in Statistics in Medicine, 2025. Written by Andy A Shen and 3 co-authors. Institution: Department of Statistics, University of California, Berkeley, California, USA. Major MEDLINE topics: Causality, Health Status Disparities. Cited 5 times so far (Crossref); 64 references. The full text is freely available in PubMed Central. Funded by: NIMH NIH HHS, National Science Foundation, American Foundation for Suicide Prevention.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Disparities in health or well‐being experienced by minority groups can be difficult to study using the traditional exposure‐outcome paradigm in causal inference, since potential outcomes in variables such as race or sexual minority status are challenging to interpret. Causal decomposition analysis addresses this gap by positing causal effects on disparities under interventions to other intervenable exposures that may play a mediating role in the disparity. While invoking weaker assumptions than causal mediation approaches, decomposition analyses are often conducted in observational settings and require uncheckable assumptions that eliminate unmeasured confounders. Leveraging the marginal sensitivity model, we develop a sensitivity analysis for weighted causal decomposition estimators and use the percentile bootstrap to construct valid confidence intervals for causal effects on disparities. We also propose a two‐parameter reformulation that enhances interpretability and facilitates an intuitive understanding of the plausibility of unmeasured confounders and their effects. We illustrate our framework on a study examining the effect of parental support on disparities in suicidal ideation among sexual minority youth. We find that the effect is small and sensitive to unmeasured confounding, suggesting that further screening studies are needed to identify mitigating interventions in this vulnerable population.

Subjects

MeSH terms (MEDLINE)

CausalityHealth Status DisparitiesHumansModels, StatisticalAdolescentSuicidal IdeationSexual and Gender MinoritiesFemaleMaleMediation AnalysisConfounding Factors, EpidemiologicConfidence Intervals

Author keywords

causal decompositionscausal inferencedisparitiessensitivity analysisweighting

Authors and institutions

  • Department of Statistics, University of California, Berkeley, California, USA.
  • Children's Hospital of Philadelphia, Pennsylvania, USA.
  • Perelman School of Medicine, University of Pennsylvania, Pennsylvania, USA.

ORCID profiles: Andy A Shen

Funding

  • NIMH NIH HHS
  • National Science Foundation
  • American Foundation for Suicide Prevention
  • National Institute of Mental Health

Cite

APA

Shen, A. A., Visoki, E., Barzilay, R., & Pimentel, S. D. (2025). A Calibrated Sensitivity Analysis for Weighted Causal Decompositions. Statistics in Medicine. https://doi.org/10.1002/sim.70010

Vancouver

Shen AA, Visoki E, Barzilay R, Pimentel SD. A Calibrated Sensitivity Analysis for Weighted Causal Decompositions. Statistics in Medicine. 2025. doi:10.1002/sim.70010

BibTeX
@article{Shen2025,
  author = {Shen, Andy A and Visoki, Elina and Barzilay, Ran and Pimentel, Samuel D},
  title = {A Calibrated Sensitivity Analysis for Weighted Causal Decompositions},
  journal = {Statistics in Medicine},
  year = {2025},
  doi = {10.1002/sim.70010}
}
Start for free →