Disparities in the association of ambient air pollution with childhood asthma incidence in the ECHO consortium: A US-wide multi-cohort study

Abstract

Background: Characterization of US sociodemographic disparities in air pollution respiratory effects has often been limited by lack of participant diversity, geography, exposure characterization, and small sample size.

Methods: We included 34 sites comprising 23,234 children (born 1981-2021) from the Environmental influences on Child Health Outcomes (ECHO) Program with data on asthma diagnosis until age 10 (182,008 person-years). Predicted annual exposure to fine particulate matter (1988-2021), nitrogen dioxide (2000-2016), and ground ozone (2000-2016) were assigned based on residential histories. For each pollutant, we fitted time-varying Cox models adjusted for time trend, site, and several area- and individual-level sociodemographic features that were separately considered as modifiers via an interaction with exposure.

Results: The hazard ratio of incident asthma by age 10 years was 1.19 (95% CI = 1.10, 1.28), 1.19 (95% CI = 1.05, 1.34), and 1.11 (95% CI = 1.01, 1.22) of an interquartile range increase in prior-year exposure to fine particulate matter (6.17 µg/m³), nitrogen dioxide (15.37 ppb), and ozone (6.87 ppb), respectively. For both fine particulate and nitrogen dioxide, children from areas with a higher proportion of Black residents or with a higher population density had greater pollution-associated risks of incident asthma. For ozone, asthma risks were enhanced in less dense areas.

Conclusions: US efforts to mitigate childhood asthma risk by reducing air pollution would benefit from addressing root structural causes of vulnerability and susceptibility, including spatial patterning in air pollution sources and exposures as well as social and economic disadvantage.

Figure 1.

Association between year prior annual air pollution and asthma incidence in the ECHO program. The black-filled circles and error bars denote the hazard ratio and corresponding 95% confidence interval of asthma for an interquartile range (IQR) increase in annual PM_2.5 (6.17 µg/m³), NO₂ (15.37 ppb), and O₃ (6.87 ppb) (left to right).

Figure 2.

Modification of the association between annual air pollution and asthma incidence by area-level characteristics. The black-filled circles and error bars denote the hazard ratio and corresponding 95% confidence interval of asthma for an interquartile range (IQR) increase in annual PM_2.5 (6.17 µg/m³), NO₂ (15.37 ppb), and O₃ (6.87 ppb) (top to bottom). An asterisk with a line spanning the different modifier levels represents the statistical significance of the interaction term between air pollution and modifier at the α = 0.05 level.

Figure 3.

Modification of the association between annual air pollution and asthma incidence by individual-level characteristics. The black-filled circles and error bars denote the hazard ratio and corresponding 95% confidence interval of asthma for an interquartile range (IQR) increase in annual PM_2.5 (6.17 µg/m³), NO₂ (15.37 ppb), and O₃ (6.87 ppb) (top to bottom). An asterisk with a line spanning the different modifier levels represents the statistical significance of the interaction term between air pollution and modifier at the α = 0.05 level.