A flexible, interpretable framework for assessing sensitivity to unmeasured confounding. (3rd May 2016)

• Record Type: Journal Article
• Authors: Dorie, Vincent; Harada, Masataka; Carnegie, Nicole Bohme; Hill, Jennifer
• Is Part Of: Statistics in medicine. Volume 35:Number 20(2016)
• Page Start: 3453
• Page End: 3470
• Publication Date: 2016-05-03
• Subjects: Bayesian modeling -- causal inference -- nonparametric regression -- sensitivity analysis -- unmeasured confounding; Medical statistics -- Periodicals; Statistique médicale -- Périodiques; Statistiques médicales -- Périodiques; 610.727
• DOI: 10.1002/sim.6973
• Languages: English
• ISSNs: 0277-6715

Abstract

Abstract : When estimating causal effects, unmeasured confounding and model misspecification are both potential sources of bias. We propose a method to simultaneously address both issues in the form of a semi‐parametric sensitivity analysis. In particular, our approach incorporates Bayesian Additive Regression Trees into a two‐parameter sensitivity analysis strategy that assesses sensitivity of posterior distributions of treatment effects to choices of sensitivity parameters. This results in an easily interpretable framework for testing for the impact of an unmeasured confounder that also limits the number of modeling assumptions. We evaluate our approach in a large‐scale simulation setting and with high blood pressure data taken from the Third National Health and Nutrition Examination Survey. The model is implemented as open‐source software, integrated into the treatSens package for the R statistical programming language. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.