Intersectional Inequalities in Neighbourhood Air Pollution Concentration in England: A Quantitative Analysis of Ecological Data Using Eco-Intersectional Multilevel (EIM) Modelling

TL;DR

This study finds that air pollution levels in England vary greatly based on neighborhood characteristics like ethnicity and deprivation, with some areas having five times more pollution than others.

Contribution

The novel Eco-Intersectional Multilevel (EIM) modeling approach enables simultaneous analysis of multiple intersecting neighborhood characteristics.

Findings

Neighborhoods with high deprivation, high minority ethnic populations, and urban settings had five times higher NOx concentrations than others.

Ethnic inequalities in NOx concentration remained significant even after accounting for area deprivation.

Area deprivation alone did not show independent inequalities once ethnicity was considered.

Abstract

Air pollution is detrimentally associated with many health outcomes, yet its impacts are not equally distributed. Research consistently finds inequalities by ethnicity, area deprivation and age. However, such inequalities are typically investigated separately, potentially underestimating the extent of differential exposures. We aim to investigate inequalities in NOx concentrations across multiple intersecting neighbourhood characteristics in England simultaneously. We do this using the novel Eco-Intersectional Multilevel (EIM) modelling approach, we define analytic “strata” of neighbourhoods based on sociodemographic characteristics. This enables us to quantify NOx concentration inequalities across community types, simultaneously considering area deprivation, ethnicity, education, rurality and age of residents. We find that neighbourhoods belonging to the “most deprived, high proportion minority ethnic, high education, urban and not ageing” stratum had the highest average NOx concentration. This concentration was five times higher than places with the lowest concentration in the mid deprivation, low proportion minority ethnic, high education, rural and ageing stratum. We find clear and striking inequalities by ethnicity. However, we do not find evidence of inequalities by area deprivation that operate independently of community ethnicity, likely due to the strong relationship between ethnicity and deprivation distributions. This study demonstrates the value of taking an intersectional approach to geographical inequalities.

The online version contains supplementary material available at 10.1007/s12061-025-09787-8.

- NOx concentration was analysed across several intersecting place characteristics.

- Eco-Intersectional Multilevel modelling was used to study these inequalities.

- NOx concentration was five times greater in the stratum with the highest versus lowest NOx.

- Large ethnic inequalities were identified after accounting for area deprivation.

The online version contains supplementary material available at 10.1007/s12061-025-09787-8.

Figures