On the PM2.5 -- Mortality Association: A Bayesian Model for Spatio-Temporal Confounding

TL;DR

This paper introduces a Bayesian spatio-temporal model to accurately estimate the health impact of PM2.5 on mortality, addressing confounding and non-linear effects in epidemiological studies.

Contribution

It develops a novel Bayesian spatial dynamic generalized linear model that captures non-linear associations and decomposes effects across spatio-temporal scales, reducing confounding bias.

Findings

Identified seasonal peaks in PM2.5 effects on mortality.

Demonstrated model's ability to recover exposure-outcome shape.

Simulations confirmed bias reduction.

Abstract

In epidemiological studies of air pollution and public health, estimating the health impact of exposure to air pollution may be hindered by the unknown functional form of the exposure-outcome association and by unmeasured confounding factors that are linked to both exposure and outcome. These challenges are especially relevant in spatio-temporal analyses, where their joint exploration remains limited. To study the effects of fine particulate matter on mortality among elderly people in Italy, we propose a Bayesian spatial dynamic generalized linear model that captures the non-linear exposure-outcome association and decomposes the exposure effect across fine and coarse spatio-temporal scales of variation. Together, these features allow reducing the spatio-temporal confounding bias and recovering the shape of the association, as demonstrated through simulation studies. The real-data analysis reveals a clear temporal pattern in the exposure effect, with peaks during summer months. We argue that this finding may be due to interactions of particulate matter with air temperature and unmeasured confounders.