EUVIMEDDécouvrir EUVIMED
Open access

A novel causal mediation analysis approach for zero‐inflated mediators

Meilin Jiang, Seonjoo Lee, A James O'Malley iD, Yaakov Stern, Zhigang Li iD

Statistics in Medicine · 2023

DOI: 10.1002/sim.9689 ↗

At a glance

Journal article in Statistics in Medicine, 2023. Written by Meilin Jiang and 4 co-authors. Institution: Department of Biostatistics, University of Florida, Gainesville, Florida. Major MEDLINE topics: Models, Statistical, Mediation Analysis. Cited 5 times so far (Crossref); 33 references. The full text is freely available in PubMed Central. Funded by: NIGMS NIH HHS, NIA NIH HHS, NIAAA NIH HHS.

RésuméSubjectsRéférences bibliographiquesCiteRelated studies

Abstract (original language)

Mediation analyses play important roles in making causal inference in biomedical research to examine causal pathways that may be mediated by one or more intermediate variables (ie, mediators). Although mediation frameworks have been well established such as counterfactual‐outcomes (ie, potential‐outcomes) models and traditional linear mediation models, little effort has been devoted to dealing with mediators with zero‐inflated structures due to challenges associated with excessive zeros. We develop a novel mediation modeling approach to address zero‐inflated mediators containing true zeros and false zeros. The new approach can decompose the total mediation effect into two components induced by zero‐inflated structures: the first component is attributable to the change in the mediator on its numerical scale which is a sum of two causal pathways and the second component is attributable only to its binary change from zero to a non‐zero status. An extensive simulation study is conducted to assess the performance and it shows that the proposed approach outperforms existing standard causal mediation analysis approaches. We also showcase the application of the proposed approach to a real study in comparison with a standard causal mediation analysis approach.

Subjects

MeSH terms (MEDLINE)

Models, StatisticalMediation AnalysisHumansComputer SimulationLinear ModelsCausality

Author keywords

causal inferencemediationzero-inflated log-normalzero-inflated mediatorzero-inflated negative binomial

Authors and institutions

  • Department of Biostatistics, University of Florida, Gainesville, Florida.
  • Mental Health Data Science, New York State Psychiatric Institute, New York, New York.
  • Departments of Biostatistics and Psychiatry, Columbia University, New York, New York.
  • Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire.
  • The Dartmouth Institute, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire.
  • Department of Neurology, Columbia University, New York, New York.

ORCID profiles: A James O'Malley, Zhigang Li

Funding

  • NIGMS NIH HHS
  • NIA NIH HHS
  • NIAAA NIH HHS
  • National Institutes of Health

Cite

APA

Jiang, M., Lee, S., O'Malley, A. J., Stern, Y., & Li, Z. (2023). A novel causal mediation analysis approach for zero‐inflated mediators. Statistics in Medicine. https://doi.org/10.1002/sim.9689

Vancouver

Jiang M, Lee S, O'Malley AJ, Stern Y, Li Z. A novel causal mediation analysis approach for zero‐inflated mediators. Statistics in Medicine. 2023. doi:10.1002/sim.9689

BibTeX
@article{Jiang2023,
  author = {Jiang, Meilin and Lee, Seonjoo and O'Malley, A James and Stern, Yaakov and Li, Zhigang},
  title = {A novel causal mediation analysis approach for zero‐inflated mediators},
  journal = {Statistics in Medicine},
  year = {2023},
  doi = {10.1002/sim.9689}
}
Start for free →