Invited Commentary: The Promise and Pitfalls of Causal Inference With Multivariate Environmental Exposures

The accompanying article by Keil et al. (Am J Epidemiol. 2021;190(12):2647-2657) deploys Bayesian g-computation to investigate the causal effect of 6 airborne metal exposures linked to power-plant emissions on birth weight. In so doing, it articulates the potential value of framing the analysis of environmental mixtures as an explicit contrast between exposure distributions that might arise in response to a well-defined intervention-here, the decommissioning of coal plants. Framing the mixture analysis as that of an approximate target trial is an important approach that deserves incorporation into the already rich literature on the analysis of environmental mixtures. However, its deployment in the power plant example highlights challenges that can arise when the target trial is at odds with the exposure distribution observed in the data, a discordance that seems particularly difficult in studies of environmental mixtures. Bayesian methodology such as model averaging and informative priors can help, but they are ultimately limited for overcoming this salient challenge.

Automated summary

The study uses Bayesian g-computation to investigate the causal effect of 6 airborne metal exposures on birth weight linked to power-plant emissions, advocating for framing the analysis of environmental mixtures as an explicit contrast between exposure distributions and focusing on the target trial approach. However, challenges arise in deploying this method in the power plant example when the target trial conflicts with the exposure distribution observed in the data.