EUVIMEDDescoperă EUVIMED
Open access

A Bayesian Approach to the G‐Formula via Iterative Conditional Regression

Ruyi Liu iD, Liangyuan Hu, Francis Perry Wilson, Joshua L Warren iD, Fan Li iD

Statistics in Medicine · 2025

DOI: 10.1002/sim.70123 ↗

At a glance

Journal article in Statistics in Medicine, 2025. Written by Ruyi Liu and 4 co-authors. Institution: Department of Biostatistics, Yale School of Public Health, New Haven, CT, USA. Major MEDLINE topics: Algorithms, Observational Studies as Topic. 71 references. The full text is freely available in PubMed Central. Funded by: NHLBI NIH HHS, Patient-Centered Outcomes Research Institute, National Heart, Lung, and Blood Institute.

RezumatSubjectsDate bibliograficeCiteRelated studies

Abstract (original language)

In longitudinal observational studies with time‐varying confounders, the generalized computation algorithm formula (g‐formula) is a principled tool to estimate the average causal effect of a treatment regimen. However, the standard non‐iterative g‐formula implementation requires specifying both the conditional distribution of the outcomes and the joint distribution of all time‐varying covariates. This process can be cumbersome to implement and is prone to model misspecification bias. As an alternative, the iterative conditional expectation (ICE) g‐formula estimator solely depends on a series of nested outcome regressions and avoids the need for specifying the full distribution of all time‐varying covariates. This simplicity lends itself to the natural integration of flexible machine learning techniques to develop more robust average causal effect estimators with time‐varying treatments. In this work, we introduce a Bayesian approach that includes parametric regressions and Bayesian Additive Regression Trees to flexibly model a series of outcome surfaces. We fit the ICE g‐formula and develop a sampling algorithm to obtain samples from the posterior distribution of the final causal effect estimator. We illustrate the performance characteristics of the Bayesian ICE estimator and the associated variations via simulation studies and applications to two real world data examples.

Subjects

MeSH terms (MEDLINE)

AlgorithmsObservational Studies as TopicBayes TheoremHumansComputer SimulationLongitudinal StudiesRegression AnalysisModels, StatisticalMachine LearningCausality

Author keywords

Bayesian additive regression treescausal inferenceg‐formulalongitudinal propensity scoreobservational studiestime‐varying confounding

Authors and institutions

  • Department of Biostatistics, Yale School of Public Health, New Haven, CT, USA.
  • Center for Methods in Implementation and Prevention Science, Yale School of Public Health, New Haven, CT, USA.
  • Department of Biostatistics and Epidemiology, Rutgers University, New Brunswick, NJ, USA.
  • Clinical and Translational Research Accelerator, Yale School of Medicine, New Haven, CT, USA.

ORCID profiles: Ruyi Liu, Joshua L Warren, Fan Li

Funding

  • NHLBI NIH HHS
  • Patient-Centered Outcomes Research Institute
  • National Heart, Lung, and Blood Institute

Cite

APA

Liu, R., Hu, L., Perry Wilson, F., Warren, J. L., & Li, F. (2025). A Bayesian Approach to the G‐Formula via Iterative Conditional Regression. Statistics in Medicine. https://doi.org/10.1002/sim.70123

Vancouver

Liu R, Hu L, Perry Wilson F, Warren JL, Li F. A Bayesian Approach to the G‐Formula via Iterative Conditional Regression. Statistics in Medicine. 2025. doi:10.1002/sim.70123

BibTeX
@article{Liu2025,
  author = {Liu, Ruyi and Hu, Liangyuan and Perry Wilson, Francis and Warren, Joshua L and Li, Fan},
  title = {A Bayesian Approach to the G‐Formula via Iterative Conditional Regression},
  journal = {Statistics in Medicine},
  year = {2025},
  doi = {10.1002/sim.70123}
}
Start for free →