Susceptibility to Air Pollution by Sex in Populations with Different Age Ranges

Abstract

Ambient air pollution is a leading environmental risk factor, contributing to a wide range of adverse health outcomes. Regulatory standards, such as those set by the U.S. Clean Air Act, are designed to protect public health, specifically the most vulnerable populations. However, evidence remains mixed regarding who is most susceptible; and growing research suggests that biological sex may modify the health effects of air pollution in complex, age-dependent ways. While some studies report greater susceptibility in females, others find stronger effects in males, with patterns often shifting across developmental stages. This dissertation evaluated air pollution health effects among different age groups, including infants (Aim 2), older adults (Aim 4), and all-age population (Aim 3), while integrating analyses of sex-specific responses in each age group. This dissertation addresses four aims and contains six Chapters. Chapter 1 is an introduction. Chapters 2 to 5 address Aims 1 to 4, respectively. Chapter 6 presents the overall discussion, conclusions, and implications for future research. Aim 1 was to generate high spatial resolution exposure estimates from sparse regulatory monitoring data. In this aim, we addressed a methodological challenge in exposure assessment: how to improve the accuracy of model-predicted estimates of machine learning models in predicting ambient air pollution in regions with sparse monitoring. We developed and evaluated two approaches, incorporating low-cost sensor data, that enhanced the extrapolating ability of random-forest models to predict daily PM2.5 concentrations in New York and Pennsylvania states. By incorporating satellite, meteorological, and land-use predictors, and data from low-cost sensor monitoring networks, the two-step approach increased the average external validation R² from 0.49 to 0.65 and reduced RMSE from 3.56 to 2.96 µg/m3, compared to basic random forest model. For the regression-enhanced random forest models, the average R2 of the external validation was 0.54, and the RMSE was 3.40 μg/m3. We also observed significant and comparable relationships between urinary 1-hydroxypyrene levels and PM2.5 predictions from both improved models. This PM2.5 model estimation strategy could improve the extrapolating ability of random forest models in areas with sparse monitoring data. Aim 2 was to investigate whether preconception folic acid (FA) supplementation attenuates the adverse effects of maternal air pollution exposure on birthweight outcomes, and to examine potential sex differences in these associations among infants. In this aim, we analyzed data from 10,579 mother–child pairs in the Shanghai Preconception Cohort, assessing modification by preconception FA supplementation and exploring sex-specific differences through stratified analyses by infant sex. Birthweight, birthweight-for-gestational-age z-scores, and risk of small-for-gestational-age (SGA) and low birthweight (LBW) were evaluated in association to PM2.5, NO2, and ozone exposure. Interaction models showed that maternal FA supplementation initiated prior to conception could mitigate the adverse effects of air pollution on birthweight and SGA. Sex-stratified analyses revealed that male infants were more adversely affected than females. We found that a 10 μg/m3 increase in PM2.5 was associated with a 62.8 g (95% CI: -97.0, -28.6) reduction in birthweight and a 1.78 (95% CI: 1.16, 2.73) odds ratio for SGA among women without preconception FA. Larger effects were observed in boys, with an 82.9 g reduction (95% CI: -130.2, -35.8) an odds ratio of 1.95 (95% CI: 1.06, 3.60), compared with a 38.2 g reduction (95% CI: −87.8, 11.3) and an odds ratio of 1.70 (95% CI: 0.94, 3.08) in girls. In contrast, no significant associations were observed among infants born to mothers who took folic-acid supplementation during the preconception period. These findings suggest both a protective role of preconception FA supplementation and an infant-sex-specific susceptibility during fetal development, providing key insights into early-life disparities in air pollution health effects. Aim 3 was to investigate the interaction between air pollution and viral exposure, and to assess potential sex-specific modifications of these effects in an all-age population. In this aim, we conducted a time-series study in residents of Ulaanbaatar, Mongolia, to evaluate the interactive effects of ambient air pollution and viral exposures on hospital visits for influenza over 7 years (n = 16,364). We also explored sex-specific differences in such interactions by stratifying analyses across sex and age groups in the overall population. We found that air pollutants, including PM1, PM2.5, and O3, amplified the effects of viral exposure during both winter and warmer seasons, with a 10-μg/m³ increase in PM1, PM2.5, and O3 during the acute window increasing the effect size by factors of 1.32, 1.16, and 1.68 in the warm season and 1.04, 1.05, and 1.16 in the cold season. Children under five exhibited the strongest synergistic effects. While overall sex differences in the full population were not significant, the lack of sex-specific data within the other age strata limited the statistical power. These results highlight the susceptibility in the interactive effects of air pollution and infection, especially among young children, and underscore the importance of future studies collecting detailed sex- and age-disaggregated data. Aim 4 was to investigate the sex difference in acute cardiorespiratory responses to air pollution exposure among older adults and explored the potential mechanisms of lipid mediators. In this aim, we conducted a secondary analysis of a randomized crossover study involving 80 older adults (42 women, 38 men) in London, UK, comparing cardiorespiratory responses to a 2-hour city traffic exposure. We examined interactions between exposure and sex for respiratory and cardiovascular measurements. We also quantified the eicosanoids and related lipid mediators (e.g., 5-HETE, 15-HDHA, TXB2) in sputum and urine using liquid chromatography–mass spectrometry (LC-MS) and to assess potential sex differences in these pathways. Results showed that compared to males, females showed greater adverse effects of traffic exposure. Only females showed significant increases in pro-inflammatory lipid mediators, particularly those associated with the 15-LOX- and COX- pathways. No sex differences were found in cardiovascular outcomes. The findings provide novel evidence, under real-world exposure conditions (i.e., walking in a city environment), of sex susceptibility in respiratory responses to short-term air pollution exposure, as well as potential biological mechanisms underlying the sex difference. In conclusion (Chapter 6), this dissertation presents an investigation of air pollution health effects, with a specific focus on how biological sex modifies these effects at different age groups. Through the integration of a prospective cohort analysis, a time-series analysis, and a crossover study, this work demonstrates that sex and age are critical modifiers of air pollution health effects. We observed sex-specific effects, which were greater in male infants and older women, and the latter one may be driven by sex-specific activation of lipid mediators. These findings underscore the need for examining sex- and age- differences in environmental health research, which can support targeted interventions and policies to protect susceptible subgroups, particularly during periods of heightened vulnerability such as pregnancy, early childhood, and older age.