Heterogeneous Distributed Lag Models to Estimate Personalized Effects of Maternal Exposures to Air Pollution.

Children's health studies support an association between maternal environmental exposures and children's birth outcomes. A common goal is to identify critical windows of susceptibility-periods during gestation with increased association between maternal exposures and a future outcome. The timing of the critical windows and magnitude of the associations are likely heterogeneous across different levels of individual, family, and neighborhood characteristics. Using an administrative Colorado birth cohort we estimate the individualized relationship between weekly exposures to fine particulate matter (PM 2.5) during gestation and birth weight. To achieve this goal, we propose a statistical learning method combining distributed lag models and Bayesian additive regression trees to estimate critical windows at the individual level and identify characteristics that induce heterogeneity from a high-dimensional set of potential modifying factors. We find evidence of heterogeneity in the PM 2.5 -birth weight relationship, with some mother-child dyads showing a 3 times larger decrease in birth weight for an IQR increase in exposure (5.9 to 8.5 PM 2.5 µg/m3) compared to the population average. Specifically, we find increased vulnerabilitity for non-Hispanic mothers who are either younger, have higher body mass index or lower educational attainment. Our case study is the first precision health study of critical windows.

Cumulative Exposures to Environmental and Socioeconomic Risk Factors in Milwaukee County, Wisconsin.

The environmental justice literature demonstrates consistently that low-income and minority communities are disproportionately exposed to environmental hazards. In this case study, we examined cumulative multipollutant, multidomain, and multimatrix environmental exposures in Milwaukee County, Wisconsin for the year 2015. We identified spatial hot spots in Milwaukee County both individually (using local Moran's I) and through clusters (using K-means clustering) across a profile of environmental pollutants that span regulatory domains and matrices of exposure, as well as socioeconomic indicators. The cluster with the highest exposures within the urban area was largely characterized by low socioeconomic status and an overrepresentation of the Non-Hispanic Black population relative to the county as a whole. In this cluster, average pollutant concentrations were equivalent to the 78th percentile in county-level blood lead levels, 67th percentile in county-level NO2, 79th percentile in county-level CO, and 78th percentile in county-level air toxics. Simultaneously, this cluster had an average equivalent to the 62nd percentile in county-level unemployment, 70th percentile in county-level population rate lacking a high school diploma, 73rd percentile in county-level poverty rate, and 28th percentile in county-level median household income. The spatial patterns of pollutant exposure and SES indicators suggested that these disparities were not random but were instead structured by socioeconomic and racial factors. Our case study, which combines environmental pollutant exposures, sociodemographic data, and clustering analysis, provides a roadmap to identify and target overburdened communities for interventions that reduce environmental exposures and consequently improve public health.

Relative humidity, temperature, and hypertensive disorders of pregnancy: Findings from the Project Viva cohort.

BACKGROUND: Preeclampsia is a multi-system hypertensive disorder of pregnancy that is a leading cause of maternal and fetal morbidity and mortality. Prior studies disagree on the cause and even the presence of seasonal patterns in its incidence. Using unsuitable time windows for seasonal exposures can bias model results, potentially explaining these inconsistencies. OBJECTIVES: We aimed to investigate humidity and temperature as possible causes for seasonal trends in preeclampsia in Project Viva, a prebirth cohort in Boston, Massachusetts, considering only exposure windows that precede disease onset. METHODS: Using the Parameter-elevation Relationships on Independent Slopes Model (PRISM) Climate Dataset, we estimated daily residential temperature and relative humidity (RH) exposures during pregnancy. Our primary multinomial regression adjusted for person-level covariates and season. Secondary analyses included distributed lag models (DLMs) and adjusted for ambient air pollutants including fine particulates (PM2.5). We used Generalized Additive Mixed Models (GAMMs) for systolic blood pressure (SBP) trajectories across hypertensive disorder statuses to confirm exposure timing. RESULTS: While preeclampsia is typically diagnosed late in pregnancy, GAMM-fitted SBP trajectories for preeclamptic and non-preeclamptic women began to diverge at around 20 weeks' gestation, confirming the need to only consider early exposures. In the primary analysis with 1776 women, RH in the early second trimester, weeks 14-20, was associated with significantly higher odds of preeclampsia (OR per IQR increase: 1.81, 95% CI: 1.10, 2.97). The DLM corroborated this window, finding a positive association from weeks 12-20. There were no other significant associations between RH or temperature and preeclampsia or gestational hypertension in any other time period. DISCUSSION: The association between preeclampsia and RH in the early second trimester was robust to model choice, suggesting that RH may contribute to seasonal trends in preeclampsia incidence. Differences between these results and those of prior studies could be attributable to exposure timing differences.

Sensitive development windows of prenatal air pollution and cognitive functioning in preschool age Mexican children.

Introduction: Neurotoxicity resulting from air pollution is of increasing concern. Considering exposure timing effects on neurodevelopmental impairments may be as important as the exposure dose. We used distributed lag regression to determine the sensitive windows of prenatal exposure to fine particulate matter (PM2.5) on children's cognition in a birth cohort in Mexico. Methods: Analysis included 553 full-term (≥37 weeks gestation) children. Prenatal daily PM2.5 exposure was estimated using a validated satellite-based spatiotemporal model. McCarthy Scales of Children's Abilities (MSCA) were used to assess children's cognitive function at 4-5 years old (lower scores indicate poorer performance). To identify susceptibility windows, we used Bayesian distributed lag interaction models to examine associations between prenatal PM2.5 levels and MSCA. This allowed us to estimate vulnerable windows while testing for effect modification. Results: After adjusting for maternal age, socioeconomic status, child age, and sex, Bayesian distributed lag interaction models showed significant associations between increased PM2.5 levels and decreased general cognitive index scores at 31-35 gestation weeks, decreased quantitative scale scores at 30-36 weeks, decreased motor scale scores at 30-36 weeks, and decreased verbal scale scores at 37-38 weeks. Estimated cumulative effects (CE) of PM2.5 across pregnancy showed significant associations with general cognitive index (CE^ = -0.35, 95% confidence interval [CI] = -0.68, -0.01), quantitative scale (CE^ = -0.27, 95% CI = -0.74, -0.02), motor scale (CE^ = -0.25, 95% CI = -0.44, -0.05), and verbal scale (CE^ = -0.2, 95% CI = -0.43, -0.02). No significant sex interactions were observed. Conclusions: Prenatal exposure to PM2.5, particularly late pregnancy, was inversely associated with subscales of MSCA. Using data-driven methods to identify sensitive window may provide insight into the mechanisms of neurodevelopmental impairment due to pollution.

Estimating cutoff values for diagnostic tests to achieve target specificity using extreme value theory.

BACKGROUND: Rapidly developing tests for emerging diseases is critical for early disease monitoring. In the early stages of an epidemic, when low prevalences are expected, high specificity tests are desired to avoid numerous false positives. Selecting a cutoff to classify positive and negative test results that has the desired operating characteristics, such as specificity, is challenging for new tests because of limited validation data with known disease status. While there is ample statistical literature on estimating quantiles of a distribution, there is limited evidence on estimating extreme quantiles from limited validation data and the resulting test characteristics in the disease testing context. METHODS: We propose using extreme value theory to select a cutoff with predetermined specificity by fitting a Pareto distribution to the upper tail of the negative controls. We compared this method to five previously proposed cutoff selection methods in a data analysis and simulation study. We analyzed COVID-19 enzyme linked immunosorbent assay antibody test results from long-term care facilities and skilled nursing staff in Colorado between May and December of 2020. RESULTS: We found the extreme value approach had minimal bias when targeting a specificity of 0.995. Using the empirical quantile of the negative controls performed well when targeting a specificity of 0.95. The higher target specificity is preferred for overall test accuracy when prevalence is low, whereas the lower target specificity is preferred when prevalence is higher and resulted in less variable prevalence estimation. DISCUSSION: While commonly used, the normal based methods showed considerable bias compared to the empirical and extreme value theory-based methods. CONCLUSIONS: When determining disease testing cutoffs from small training data samples, we recommend using the extreme value based-methods when targeting a high specificity and the empirical quantile when targeting a lower specificity.

DNA methylation profiles reveal sex-specific associations between gestational exposure to ambient air pollution and placenta cell-type composition in the PRISM cohort study.

BACKGROUND: Gestational exposure to ambient air pollution has been associated with adverse health outcomes for mothers and newborns. The placenta is a central regulator of the in utero environment that orchestrates development and postnatal life via fetal programming. Ambient air pollution contaminants can reach the placenta and have been shown to alter bulk placental tissue DNA methylation patterns. Yet the effect of air pollution on placental cell-type composition has not been examined. We aimed to investigate whether the exposure to ambient air pollution during gestation is associated with placental cell types inferred from DNA methylation profiles. METHODS: We leveraged data from 226 mother-infant pairs in the Programming of Intergenerational Stress Mechanisms (PRISM) longitudinal cohort in the Northeastern US. Daily concentrations of fine particulate matter (PM2.5) at 1 km spatial resolution were estimated from a spatiotemporal model developed with satellite data and linked to womens' addresses during pregnancy and infants' date of birth. The proportions of six cell types [syncytiotrophoblasts, trophoblasts, stromal, endothelial, Hofbauer and nucleated red blood cells (nRBCs)] were derived from placental tissue 450K DNA methylation array. We applied compositional regression to examine overall changes in placenta cell-type composition related to PM2.5 average by pregnancy trimester. We also investigated the association between PM2.5 and individual cell types using beta regression. All analyses were performed in the overall sample and stratified by infant sex adjusted for covariates. RESULTS: In male infants, first trimester (T1) PM2.5 was associated with changes in placental cell composition (p = 0.03), driven by a decrease [per one PM2.5 interquartile range (IQR)] of 0.037 in the syncytiotrophoblasts proportion (95% confidence interval (CI) [- 0.066, - 0.012]), accompanied by an increase in trophoblasts of 0.033 (95% CI: [0.009, 0.064]). In females, second and third trimester PM2.5 were associated with overall changes in placental cell-type composition (T2: p = 0.040; T3: p = 0.049), with a decrease in the nRBC proportion. Individual cell-type analysis with beta regression showed similar results with an additional association found for third trimester PM2.5 and stromal cells in females (decrease of 0.054, p = 0.024). CONCLUSION: Gestational exposure to air pollution was associated with placenta cell composition. Further research is needed to corroborate these findings and evaluate their role in PM2.5-related impact in the placenta and consequent fetal programming.

Prenatal ambient air pollutant mixture exposure and neurodevelopment in urban children in the Northeastern United States.

BACKGROUND: Studies of prenatal air pollution (AP) exposure on child neurodevelopment have mostly focused on a single pollutant. We leveraged daily exposure data and implemented novel data-driven statistical approaches to assess effects of prenatal exposure to a mixture of seven air pollutants on cognitive functioning in school-age children from an urban pregnancy cohort. METHODS: Analyses included 236 children born at ≥37 weeks gestation. Maternal prenatal daily exposure levels for nitrogen dioxide (NO2), ozone (O3), and constituents of fine particles [elemental carbon (EC), organic carbon (OC), nitrate (NO3-), sulfate (SO42-), ammonium (NH4+)] were estimated based on residential addresses using validated satellite-based hybrid models or global 3-D chemical-transport models. Children completed Wide Range Assessment of Memory and Learning (WRAML-2) and Conners' Continuous Performance Test (CPT-II) at 6.5 ± 0.9 years of age. Time-weighted levels for mixture pollutants were estimated using Bayesian Kernel Machine Regression Distributed Lag Models (BKMR-DLMs), with which we also explored the interactions in the exposure-response functions among pollutants. Resulting time-weighted exposure levels were used in Weighted Quantile Sum (WQS) regressions to examine AP mixture effects on outcomes, adjusted for maternal age, education, child sex, and prenatal temperature. RESULTS: Mothers were primarily ethnic minorities (81% Hispanic and/or black) reporting ≤12 years of education (68%). Prenatal AP mixture (per unit increase in WQS estimated AP index) was associated with decreased WRAML-2 general memory (GM; ß = -0.64, 95%CI = -1.40, 0.00) and memory-related attention/concentration (AC; ß = -1.03, 95%CI = -1.78, -0.27) indices, indicating poorer memory functioning, as well as increased CPT-II omission errors (OE; ß = 1.55, 95%CI = 0.34, 2.77), indicating increased attention problems. When stratified by sex, association with AC index was significant among girls, while association with OE was significant among boys. Traffic-related pollutants (NO2, OC, EC) and SO42- were major contributors to these associations. There was no significant evidence of interactions among mixture components. CONCLUSIONS: Prenatal exposure to an AP mixture was associated with child neurocognitive outcomes in a sex- and domain-specific manner.

Incorporating biological knowledge in analyses of environmental mixtures and health.

A key goal of environmental health research is to assess the risk posed by mixtures of pollutants. As epidemiologic studies of mixtures can be expensive to conduct, it behooves researchers to incorporate prior knowledge about mixtures into their analyses. This work extends the Bayesian multiple index model (BMIM), which assumes the exposure-response function is a nonparametric function of a set of linear combinations of pollutants formed with a set of exposure-specific weights. The framework is attractive because it combines the flexibility of response-surface methods with the interpretability of linear index models. We propose three strategies to incorporate prior toxicological knowledge into construction of indices in a BMIM: (a) imposing directional homogeneity constraints on the weights, (b) structuring index weights by exposure transformations, and (c) placing informative priors on the index weights. We propose a novel prior specification that combines spike-and-slab variable selection with an informative Dirichlet distribution based on relative potency factors often derived from previous toxicological studies. In simulations we show that the proposed priors improve inferences when prior information is correct and can protect against misspecification suffered by naïve toxicological models when prior information is incorrect. Moreover, different strategies may be mixed-and-matched for different indices to suit available information (or lack thereof). We demonstrate the proposed methods on an analysis of data from the National Health and Nutrition Examination Survey and incorporate prior information on relative chemical potencies obtained from toxic equivalency factors available in the literature.

Prenatal Ambient Air Pollutant Mixture Exposure and Early School-age Lung Function.

Research linking prenatal ambient air pollution with childhood lung function has largely considered one pollutant at a time. Real-life exposure is to mixtures of pollutants and their chemical components; not considering joint effects/effect modification by co-exposures contributes to misleading results. Methods: Analyses included 198 mother-child dyads recruited from two hospitals and affiliated community health centers in Boston, Massachusetts, USA. Daily prenatal pollutant exposures were estimated using satellite-based hybrid chemical-transport models, including nitrogen dioxide(NO2), ozone(O3), and fine particle constituents (elemental carbon [EC], organic carbon [OC], nitrate [NO3 -], sulfate [SO4 2-], and ammonium [NH4 +]). Spirometry was performed at age 6.99 ± 0.89 years; forced expiratory volume in 1s (FEV1), forced vital capacity (FVC), and forced mid-expiratory flow (FEF25-75) z-scores accounted for age, sex, height, and race/ethnicity. We examined associations between weekly-averaged prenatal pollution mixture levels and outcomes using Bayesian Kernel Machine Regression-Distributed Lag Models (BKMR-DLMs) to identify susceptibility windows for each component and estimate a potentially complex mixture exposure-response relationship including nonlinear effects and interactions among exposures. We also performed linear regression models using time-weighted-mixture component levels derived by BKMR-DLMs adjusting for maternal age, education, perinatal smoking, and temperature. Results: Most mothers were Hispanic (63%) or Black (21%) with ≤12 years of education (67%). BKMR-DLMs identified a significant effect for O3 exposure at 18-22 weeks gestation predicting lower FEV1/FVC. Linear regression identified significant associations for O3, NH4 +, and OC with decreased FEV1/FVC, FEV1, and FEF25-75, respectively. There was no evidence of interactions among pollutants. Conclusions: In this multi-pollutant model, prenatal O3, OC, and NH4 + were most strongly associated with reduced early childhood lung function.

The COVID-19-wildfire smoke paradox: Reduced risk of all-cause mortality due to wildfire smoke in Colorado during the first year of the COVID-19 pandemic.

BACKGROUND: In 2020, the American West faced two competing challenges: the COVID-19 pandemic and the worst wildfire season on record. Several studies have investigated the impact of wildfire smoke (WFS) on COVID-19 morbidity and mortality, but little is known about how these two public health challenges impact mortality risk for other causes. OBJECTIVES: Using a time-series design, we evaluated how daily risk of mortality due to WFS exposure differed for periods before and during the COVID-19 pandemic. METHODS: Our study included daily data for 11 counties in the Front Range region of Colorado (2010-2020). We assessed WFS exposure using data from the National Oceanic and Atmospheric Administration and used mortality counts from the Colorado Department of Public Health and Environment. We estimated the interaction between WFS and the pandemic (an indicator variable) on mortality risk using generalized additive models adjusted for year, day of week, fine particulate matter, ozone, temperature, and a smoothed term for day of year. RESULTS: WFS impacted the study area on 10% of county-days. We observed a positive association between the presence of WFS and all-cause mortality risk (incidence rate ratio (IRR) = 1.03, 95%CI: 1.01-1.04 for same-day exposures) during the period before the pandemic; however, WFS exposure during the pandemic resulted in decreased risk of all-cause mortality (IRR = 0.90, 95%CI: 0.87-0.93 for same-day exposures). DISCUSSION: We hypothesize that mitigation efforts during the first year of the pandemic, e.g., mask mandates, along with high ambient WFS levels encouraged health behaviors that reduced exposure to WFS and reduced risk of all-cause mortality. Our results suggest a need to examine how associations between WFS and mortality are impacted by pandemic-related factors and that there may be lessons from the pandemic that could be translated into health-protective policies during future wildfire events.

Infinite hidden Markov models for multiple multivariate time series with missing data.

Exposure to air pollution is associated with increased morbidity and mortality. Recent technological advancements permit the collection of time-resolved personal exposure data. Such data are often incomplete with missing observations and exposures below the limit of detection, which limit their use in health effects studies. In this paper, we develop an infinite hidden Markov model for multiple asynchronous multivariate time series with missing data. Our model is designed to include covariates that can inform transitions among hidden states. We implement beam sampling, a combination of slice sampling and dynamic programming, to sample the hidden states, and a Bayesian multiple imputation algorithm to impute missing data. In simulation studies, our model excels in estimating hidden states and state-specific means and imputing observations that are missing at random or below the limit of detection. We validate our imputation approach on data from the Fort Collins Commuter Study. We show that the estimated hidden states improve imputations for data that are missing at random compared to existing approaches. In a case study of the Fort Collins Commuter Study, we describe the inferential gains obtained from our model including improved imputation of missing data and the ability to identify shared patterns in activity and exposure among repeated sampling days for individuals and among distinct individuals.

Bayesian multiple index models for environmental mixtures.

An important goal of environmental health research is to assess the risk posed by mixtures of environmental exposures. Two popular classes of models for mixtures analyses are response-surface methods and exposure-index methods. Response-surface methods estimate high-dimensional surfaces and are thus highly flexible but difficult to interpret. In contrast, exposure-index methods decompose coefficients from a linear model into an overall mixture effect and individual index weights; these models yield easily interpretable effect estimates and efficient inferences when model assumptions hold, but, like most parsimonious models, incur bias when these assumptions do not hold. In this paper, we propose a Bayesian multiple index model framework that combines the strengths of each, allowing for non-linear and non-additive relationships between exposure indices and a health outcome, while reducing the dimensionality of the exposure vector and estimating index weights with variable selection. This framework contains response-surface and exposure-index models as special cases, thereby unifying the two analysis strategies. This unification increases the range of models possible for analysing environmental mixtures and health, allowing one to select an appropriate analysis from a spectrum of models varying in flexibility and interpretability. In an analysis of the association between telomere length and 18 organic pollutants in the National Health and Nutrition Examination Survey (NHANES), the proposed approach fits the data as well as more complex response-surface methods and yields more interpretable results.

Estimating perinatal critical windows of susceptibility to environmental mixtures via structured Bayesian regression tree pairs.

Maternal exposure to environmental chemicals during pregnancy can alter birth and children's health outcomes. Research seeks to identify critical windows, time periods when exposures can change future health outcomes, and estimate the exposure-response relationship. Existing statistical approaches focus on estimation of the association between maternal exposure to a single environmental chemical observed at high temporal resolution (e.g., weekly throughout pregnancy) and children's health outcomes. Extending to multiple chemicals observed at high temporal resolution poses a dimensionality problem and statistical methods are lacking. We propose a regression tree-based model for mixtures of exposures observed at high temporal resolution. The proposed approach uses an additive ensemble of tree pairs that defines structured main effects and interactions between time-resolved predictors and performs variable selection to select out of the model predictors not correlated with the outcome. In simulation, we show that the tree-based approach performs better than existing methods for a single exposure and can accurately estimate critical windows in the exposure-response relation for mixtures. We apply our method to estimate the relationship between five exposures measured weekly throughout pregnancy and birth weight in a Denver, Colorado, birth cohort. We identified critical windows during which fine particulate matter, sulfur dioxide, and temperature are negatively associated with birth weight and an interaction between fine particulate matter and temperature. Software is made available in the R package dlmtree.

Bias Amplification and Variance Inflation in Distributed Lag Models Using Low-Spatial-Resolution Data.

Distributed lag models (DLMs) are often used to estimate lagged associations and identify critical exposure windows. In a simulation study of prenatal nitrogen dioxide (NO2) exposure and birth weight, we demonstrate that bias amplification and variance inflation can manifest under certain combinations of DLM estimation approaches and time-trend adjustment methods when using low-spatial-resolution exposures with extended lags. Our simulations showed that when using high-spatial-resolution exposure data, any time-trend adjustment method produced low bias and nominal coverage for the distributed lag estimator. When using either low- or no-spatial-resolution exposures, bias due to time trends was amplified for all adjustment methods. Variance inflation was higher in low- or no-spatial-resolution DLMs when using a long-term spline to adjust for seasonality and long-term trends due to concurvity between a distributed lag function and secular function of time. NO2-birth weight analyses in a Massachusetts-based cohort showed that associations were negative for exposures experienced in gestational weeks 15-30 when using high-spatial-resolution DLMs; however, associations were null and positive for DLMs with low- and no-spatial-resolution exposures, respectively, which is likely due to bias amplification. DLM analyses should jointly consider the spatial resolution of exposure data and the parameterizations of the time trend adjustment and lag constraints.

Using non-parametric Bayes shrinkage to assess relationships between multiple environmental and social stressors and neonatal size and body composition in the Healthy Start cohort.

BACKGROUND: Both environmental and social factors have been linked to birth weight and adiposity at birth, but few studies consider the effects of exposure mixtures. Our objective was to identify which components of a mixture of neighborhood-level environmental and social exposures were driving associations with birth weight and adiposity at birth in the Healthy Start cohort. METHODS: Exposures were assessed at the census tract level and included air pollution, built environment characteristics, and socioeconomic status. Prenatal exposures were assigned based on address at enrollment. Birth weight was measured at delivery and adiposity was measured using air displacement plethysmography within three days. We used non-parametric Bayes shrinkage (NPB) to identify exposures that were associated with our outcomes of interest. NPB models were compared to single-predictor linear regression. We also included generalized additive models (GAM) to assess nonlinear relationships. All regression models were adjusted for individual-level covariates, including maternal age, pre-pregnancy BMI, and smoking. RESULTS: Results from NPB models showed most exposures were negatively associated with birth weight, though credible intervals were wide and generally contained zero. However, the NPB model identified an interaction between ozone and temperature on birth weight, and the GAM suggested potential non-linear relationships. For associations between ozone or temperature with birth weight, we observed effect modification by maternal race/ethnicity, where effects were stronger for mothers who identified as a race or ethnicity other than non-Hispanic White. No associations with adiposity at birth were observed. CONCLUSIONS: NPB identified prenatal exposures to ozone and temperature as predictors of birth weight, and mothers who identify as a race or ethnicity other than non-Hispanic White might be disproportionately impacted. However, NPB models may have limited applicability when non-linear effects are present. Future work should consider a two-stage approach where NPB is used to reduce dimensionality and alternative approaches examine non-linear effects.

KERNEL MACHINE AND DISTRIBUTED LAG MODELS FOR ASSESSING WINDOWS OF SUSCEPTIBILITY TO ENVIRONMENTAL MIXTURES IN CHILDREN'S HEALTH STUDIES.

Exposures to environmental chemicals during gestation can alter health status later in life. Most studies of maternal exposure to chemicals during pregnancy have focused on a single chemical exposure observed at high temporal resolution. Recent research has turned to focus on exposure to mixtures of multiple chemicals, generally observed at a single time point. We consider statistical methods for analyzing data on chemical mixtures that are observed at a high temporal resolution. As motivation, we analyze the association between exposure to four ambient air pollutants observed weekly throughout gestation and birth weight in a Boston-area prospective birth cohort. To explore patterns in the data, we first apply methods for analyzing data on (1) a single chemical observed at high temporal resolution, and (2) a mixture measured at a single point in time. We highlight the shortcomings of these approaches for temporally-resolved data on exposure to chemical mixtures. Second, we propose a novel method, a Bayesian kernel machine regression distributed lag model (BKMR-DLM), that simultaneously accounts for nonlinear associations and interactions among time-varying measures of exposure to mixtures. BKMR-DLM uses a functional weight for each exposure that parameterizes the window of susceptibility corresponding to that exposure within a kernel machine framework that captures non-linear and interaction effects of the multivariate exposure on the outcome. In a simulation study, we show that the proposed method can better estimate the exposure-response function and, in high signal settings, can identify critical windows in time during which exposure has an increased association with the outcome. Applying the proposed method to the Boston birth cohort data, we find evidence of a negative association between organic carbon and birth weight and that nitrate modifies the organic carbon, elemental carbon, and sulfate exposure-response functions.

Multiple exposure distributed lag models with variable selection.

Distributed lag models are useful in environmental epidemiology as they allow the user to investigate critical windows of exposure, defined as the time periods during which exposure to a pollutant adversely affects health outcomes. Recent studies have focused on estimating the health effects of a large number of environmental exposures, or an environmental mixture, on health outcomes. In such settings, it is important to understand which environmental exposures affect a particular outcome, while acknowledging the possibility that different exposures have different critical windows. Further, in studies of environmental mixtures, it is important to identify interactions among exposures and to account for the fact that this interaction may occur between two exposures having different critical windows. Exposure to one exposure early in time could cause an individual to be more or less susceptible to another exposure later in time. We propose a Bayesian model to estimate the temporal effects of a large number of exposures on an outcome. We use spike-and-slab priors and semiparametric distributed lag curves to identify important exposures and exposure interactions and discuss extensions with improved power to detect harmful exposures. We then apply these methods to estimate the effects of exposure to multiple air pollutants during pregnancy on birthweight from vital records in Colorado.

Association between air pollution and COVID-19 disease severity via Bayesian multinomial logistic regression with partially missing outcomes.

Recent ecological analyses suggest air pollution exposure may increase susceptibility to and severity of coronavirus disease 2019 (COVID-19). Individual-level studies are needed to clarify the relationship between air pollution exposure and COVID-19 outcomes. We conduct an individual-level analysis of long-term exposure to air pollution and weather on peak COVID-19 severity. We develop a Bayesian multinomial logistic regression model with a multiple imputation approach to impute partially missing health outcomes. Our approach is based on the stick-breaking representation of the multinomial distribution, which offers computational advantages, but presents challenges in interpreting regression coefficients. We propose a novel inferential approach to address these challenges. In a simulation study, we demonstrate our method's ability to impute missing outcome data and improve estimation of regression coefficients compared to a complete case analysis. In our analysis of 55,273 COVID-19 cases in Denver, Colorado, increased annual exposure to fine particulate matter in the year prior to the pandemic was associated with increased risk of severe COVID-19 outcomes. We also found COVID-19 disease severity to be associated with interactions between exposures. Our individual-level analysis fills a gap in the literature and helps to elucidate the association between long-term exposure to air pollution and COVID-19 outcomes.

Household air pollution from wood-burning cookstoves and C-reactive protein among women in rural Honduras.

Household air pollution from solid fuel combustion was estimated to cause 2.31 million deaths worldwide in 2019; cardiovascular disease is a substantial contributor to the global burden. We evaluated the cross-sectional association between household air pollution (24-h gravimetric kitchen and personal particulate matter (PM2.5) and black carbon (BC)) and C-reactive protein (CRP) measured in dried blood spots among 107 women in rural Honduras using wood-burning traditional or Justa (an engineered combustion chamber) stoves. A suite of 6 additional markers of systemic injury and inflammation were considered in secondary analyses. We adjusted for potential confounders and assessed effect modification of several cardiovascular-disease risk factors. The median (25th, 75th percentiles) 24-h-average personal PM2.5 concentration was 115 µg/m3 (65,154 µg/m3) for traditional stove users and 52 µg/m3 (39, 81 µg/m3) for Justa stove users; kitchen PM2.5 and BC had similar patterns. Higher concentrations of PM2.5 and BC were associated with higher levels of CRP (e.g., a 25% increase in personal PM2.5 was associated with a 10.5% increase in CRP [95% CI: 1.2-20.6]). In secondary analyses, results were generally consistent with a null association. Evidence for effect modification between pollutant measures and four different cardiovascular risk factors (e.g., high blood pressure) was inconsistent. These results support the growing evidence linking household air pollution and cardiovascular disease.

Prenatal Fine Particulate Matter, Maternal Micronutrient Antioxidant Intake, and Early Childhood Repeated Wheeze: Effect Modification by Race/Ethnicity and Sex.

Fine particulate matter (PM2.5) potentiates in utero oxidative stress influencing fetal development while antioxidants have potential protective effects. We examined associations among prenatal PM2.5, maternal antioxidant intake, and childhood wheeze in an urban pregnancy cohort (n = 530). Daily PM2.5 exposure over gestation was estimated using a satellite-based spatiotemporally resolved model. Mothers completed the modified Block98 food frequency questionnaire. Average energy-adjusted percentile intake of ß-carotene, vitamins (A, C, E), and trace minerals (zinc, magnesium, selenium) constituted an antioxidant index (AI). Maternal-reported child wheeze was ascertained up to 4.1 ± 2.8 years. Bayesian distributed lag interaction models (BDLIMs) were used to examine time-varying associations between prenatal PM2.5 and repeated wheeze (≥2 episodes) and effect modification by AI, race/ethnicity, and child sex. Covariates included maternal age, education, asthma, and temperature. Women were 39% Black and 33% Hispanic, 36% with ≤high school education; 21% of children had repeated wheeze. Higher AI was associated with decreased wheeze in Blacks (OR = 0.37 (0.19-0.73), per IQR increase). BDLIMs identified a sensitive window for PM2.5 effects on wheeze among boys born to Black mothers with low AI (at 33-40 weeks gestation; OR = 1.74 (1.19-2.54), per µg/m3 increase in PM2.5). Relationships among prenatal PM2.5, antioxidant intake, and child wheeze were modified by race/ethnicity and sex.