Bayesian propensity scores for high-dimensional causal inference: a comparison of drug-eluting to bare-metal coronary stents

DOI10.1002/BIMJ.201700305MaRDI QIDQ4584965zbMATH OpenOpenAlexWikidataFDO

Authors Jacob V. Spertus, Sharon-Lise Normand

Publication date 5 September 2018

Published in Biometrical Journal (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1710.03138, http://europepmc.org/articles/pmc6397046

zbMATH Keywords

regularizationBayesian methodscoronary stentBayesian additive regression treespropensity score weighting

Mathematics Subject Classification ID

Bayesian inference (62F15) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Abstract: High-dimensional data can be useful for causal inference by providing many confounders that may bolster the plausibility of the ignorability assumption. Propensity score methods are powerful tools for causal inference, are popular in health care research, and are particularly useful for high-dimensional data. Recent interest has surrounded a Bayesian formulation of these methods in order to flexibly estimate propensity scores and summarize posterior quantities while incorporating variance from the (potentially high-dimensional) treatment model. We discuss methods for Bayesian propensity score analysis of binary treatments, focusing on modern methods for high-dimensional Bayesian regression and the propagation of uncertainty from the treatment regression. We introduce a novel and simple estimator for the average treatment effect that capitalizes on conjugancy of the beta and binomial distributions. Through simulations, we show the utility of horseshoe priors and Bayesian additive regression trees paired with our new estimator, while demonstrating the importance of including variance from the treatment and outcome models. Cardiac stent data with almost 500 confounders and 9000 patients illustrate approaches and compare among existing frequentist alternatives.

Recommendations

• Model feedback in Bayesian propensity score estimation
• Causal inference in high dimensions: A marriage between Bayesian modeling and good frequentist properties
• A two-step Bayesian approach for propensity score analysis: simulations and case study
• Causal Inference With General Treatment Regimes
• A Bayesian view of doubly robust causal inference: Table 1.

Cited in

(4)

• Uncertainty in the design stage of two-stage Bayesian propensity score analysis
• A two-step Bayesian approach for propensity score analysis: simulations and case study
• Causal inference in high dimensions: A marriage between Bayesian modeling and good frequentist properties
• Model feedback in Bayesian propensity score estimation

This page was built for publication: Bayesian propensity scores for high-dimensional causal inference: a comparison of drug-eluting to bare-metal coronary stents

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4584965)