Mitigating unobserved spatial confounding bias with mixed models

by   Patrick Schnell, et al.

Confounding by unmeasured spatial variables has received some attention in the spatial statistics and causal inference literatures, but concepts and approaches have remained largely separated. In this paper, we aim to bridge these distinct strands of statistics by considering unmeasured spatial confounding within a formal causal inference framework, and estimating effects using modifications of outcome regression tools popular within the spatial literature. First, we show that using spatially correlated random effects in the outcome model, an approach common among spatial statisticians, does not mitigate bias due to spatial confounding. Motivated by the bias term of commonly-used estimators, we propose an affine estimator which addresses this deficiency. We discuss how unbiased estimation of causal parameters in the presence of unmeasured spatial confounding can only be achieved under an untestable set of assumptions which will often be application-specific. We provide one set of assumptions that is sufficient for identification of the causal effect based on the observed data. These assumptions describe how the exposure and outcome of interest relate to the unmeasured variables. Estimation of the model components necessary for unbiased estimation of the causal effect proceeds using tools common in the spatial statistics literature, and specifically via a regularized restricted maximum likelihood approach employing weakly informative priors to avoid degenerate estimates. This work is motivated by and used to estimate the causal effect of county-level (a) exposure to emissions from coal-powered electricity generating units, and (b) relative humidity on particulate matter across the New England area in the United States, and to investigate the potential threat from unmeasured spatial confounders in this context.


Spatial causal inference in the presence of unmeasured confounding and interference

Causal inference in spatial settings is met with unique challenges and o...

Causal Simulation Experiments: Lessons from Bias Amplification

Recent theoretical work in causal inference has explored an important cl...

An Improved Doubly Robust Estimator Using Partially Recovered Unmeasured Spatial Confounder

Studies in environmental and epidemiological sciences are often spatiall...

Bayesian data fusion for unmeasured confounding

Bayesian causal inference offers a principled approach to policy evaluat...

On the definition of a confounder

The causal inference literature has provided a clear formal definition o...

Discussion of the manuscript: Spatial+ a novel approach to spatial confounding

I congratulate Dupont, Wood and Augustin (DWA hereon) for providing an e...

Statistics for Spatially Stratified Heterogeneous Data

Spatial statistics is dominated by spatial autocorrelation (SAC) based K...

Please sign up or login with your details

Forgot password? Click here to reset