Applying a multiple exposures framework to explore complex spatial inequalities of Covid-19-related excess mortality in Lombardy region of Northern Italy

Abstract

BACKGROUND AND AIM: Spatial inequalities of mortality rates caused by Covid-19 have been observed in many countries, including Italy. While such spatial inequalities may be influenced by a combination of multiple place-based factors, studies thus far have used only conventional multivariable regression approaches to explore spatial determinants of Covid-19-related mortality. METHODS: We use mortality data from all of 2020 and model excess mortality using the previous 5-year mortality average for each municipality in Italy’s Lombardy region. Using Bayesian profile regression (BPR), a non-parametric clustering algorithm, we fit 19 spatial covariates to identify clusters of municipalities with similar ‘spatial exposure profiles’ and explore which exposure profiles are associated with either a higher or lower adjusted risk of excess mortality during the pandemic year (2020). These 19 spatial covariates include six air pollutants and a variety of socio-demographic factors, land use indicators, and health facility contextual information. RESULTS:BPR resulted in 23 clusters of exposure profiles. Controlling for spatial autocorrelation and health protective agency, we find that the highest risk cluster, located in the sparsely populated southern sub-region of Pavia province, has the highest proportion of their population 65 years or older and is characterized with elevated ozone and SO2 air pollutant levels, and has relatively low access to health facilities. Clusters with elevated levels of three or more air pollutants exhibited significantly elevated excess mortality risk only if they were in densely populated urban areas, if the air pollutants appeared to be traffic-related, and if they were located further away from major capital cities of Lombardy provinces. CONCLUSIONS:Our results suggest a complex web of spatial determinants that interact to influence spatial inequalities of Covid-19-related excess mortality in Italy’s Lombardy region. Studies must apply a multiple exposures framework to help unravel the complex and multi-dimensional nature of spatial inequalities of Covid-19 impacts on public health. KEYWORDS: Covid-19, air pollution, inequalities, exposome