Low-dose metal exposure contributes to chronic disease, yet differences in metal exposure and metabolism across diverse U.S. populations remain poorly characterized. Geographic location, drinking-water sources, diet, lifestyle, and culturally specific product use may shape multi-metal exposure. To characterize urinary multi-metal biomarkers among Asian Indian adults, compare them with profiles with other western U.S. populations, and evaluate serum–urine relationships for these metals among Asian Indian adults to distinguish metal clearance from physiological regulation. We measured 19 urinary metals by inductively coupled plasma mass spectrometry (ICP-MS) in harmonized subsamples of Asian Indians in the Mediators of Atherosclerosis in South Asians Living in America (MASALA) pilot study (San Francisco Bay Area, 2010–2013; n = 111), White, Black, Hispanic/Latino, and Chinese in the Multi-Ethnic Study of Atherosclerosis, Los Angeles (MESA-LA, 2010–2012; n = 444), and American Indians in the Strong Heart Family Study (SHFS, 2006–2009; n = 111). Metals were creatinine-corrected and reported as µg/g creatinine. In MASALA, the same 19-metal panel was measured in serum. Median age (years)/percent female were 56/50% (MASALA), 62/49% (MESA-LA), and 56/49% (SHFS). Compared with MESA-LA and SHFS, MASALA participants had the lowest urinary Zn (354 µg/g) and Mn (0.11 µg/g), and the highest Mo (66.5 µg/g) and Tl (0.26 µg/g). Median urinary Cs in MASALA (8.55 µg/g) was comparable to MESA-LA Chinese (9.38 µg/g), while Ba was highest in MASALA (2.57 µg/g) relative to both MESA-LA and SHFS. MASALA participants had higher As, Pb, and Cd than SHFS (As 7.60 vs 5.08 µg/g; Pb 0.88 vs 0.59 µg/g; Cd 0.51 vs 0.47 µg/g), but lower levels than MESA-LA Chinese (As 43.9 µg/g; Pb 1.12 µg/g; Cd 0.81 µg/g). In MASALA, serum–urine correlations were strongest for Mo (r = 0.83), As (r = 0.79), Co (r = 0.76), and Cs (r = 0.67); Zn, Cu, and Mn were weakly correlated. Multi-metal biomarker patterns differed across study populations, influencing exposure assessment and biomonitoring interpretation. Serum–urine analyses distinguish elements reflecting recent exposure versus physiological regulation and support population-specific metal mixtures linked to chronic disease risk. Harmonized multi-metal biomarker data across diverse U.S. populations remain limited, which constrains interpretation of exposure patterns in epidemiologic research. This study provides one of the first cross-cohort comparisons of urinary metal profiles across diverse study populations in the western United States and includes paired urine–serum measurements for a broad 19-metal panel in MASALA adults, a population with elevated cardiometabolic disease risk but limited metal exposure data. The results support population-specific biomonitoring approaches and future studies evaluating metal mixtures and chronic disease.
BACKGROUND:Environmental noise is linked to adverse cardiovascular effects. However, few studies have examined people's heart rate variability (HRV) changes related to noise sources, despite the elevated cardiovascular risk associated with reduced HRV. OBJECTIVES:We investigated HRV changes following short- and long-term lagged environmental and headphone sound exposure, assessed using the standard deviation of normal-to-normal intervals (SDNN). We also examined effect modification by individual characteristics. METHODS:Eligible participants were a subset of consented adults from the Apple Hearing Study, a nationwide prospective cohort. Analyses were constructed into four analytical cohorts defined by sound source (environmental vs. headphone) and exposure window (short-term: 20 min vs. long-term: 160 min). In each cohort, twenty participants were randomly sampled from each unique age-sex-ethnicity group. Smartwatches and smartphones measured sound levels and 1-min SDNN estimates. Distributed lag non-linear hierarchical Bayesian models estimated lag-specific and overall changes in first-stage analyses, adjusting for time of day and physical activity, with random effects to capture individual-specific variation. Second-stage linear regression models summarized the average change by age, sex, BMI, stress, self-rated hearing ability, and tinnitus. RESULTS:The stratified sampling ranged from 980 to 1180 participants across analyses. An overall SDNN reduction of 6.8% (short-term) to 16.0% (long-term) was observed following a 10-decibel increase in environmental noise exposure, and 7.1% (short-term) to 7.4% (long-term) for headphone sound. Lag-specific prediction showed brief initial SDNN increases followed by sustained reductions. Greater SDNN reductions were seen in older adults, individuals with tinnitus, or individuals who reported excellent hearing (for environmental noise) or those who reported poor hearing (for headphone sound). SIGNIFICANCE:Findings suggested noise-HRV relationships differed by exposure window, noise source, and individual characteristics, highlighting the importance of stratified risk assessment, especially in the elderly and individuals with hearing loss. However, generalizability may be limited by cohort characteristics and reliance on consumer-grade wearables. IMPACT:The physiological stress associated with noise exposure can persist from minutes to hours, as reflected by decreasing heart rate variability associated with elevated noise exposure levels. Given the association between chronically low heart rate variability and higher risk of cardiovascular diseases, noise source-specific risk assessments and noise control efforts are warranted for future studies on the non-auditory health effects, particularly for vulnerable populations.
BACKGROUND:In 2023, Chicago experienced unprecedented levels of fine particulate matter (PM2.5) caused by the long-range transport of wildfire smoke (WFS). OBJECTIVE:Examine how time trends in WFS-related fine PM2.5 varied across regions of the city with varied demographic, socio-economic, and health profiles. METHODS:The City of Chicago includes 792 census tracts grouped into nine regions. Hourly PM2.5 concentrations estimated on 500-m grids across the city (May 1-July 31, 2023) were used to calculate daily 24-h average PM2.5 levels at the tract level. To compare across city regions, interrupted time-series (ITS) models estimated PM2.5 levels for: (1) Pre-peak WFS (May 1-June 26); (2) Peak WFS (June 27-June 28); and (3) Post-peak WFS (June 29-July 31). RESULTS:Across Chicago, 24-h average tract-level PM2.5 concentrations during the pre-peak, peak, and post-peak periods were 14.5 μg/m3, 99.7 μg/m3, and 16.6 μg/m3, respectively. Baseline PM2.5 concentrations (May 1, 2023) in the northern and central regions (higher socioeconomic status) were lower than those in the southern regions (lower socioeconomic status). Differences ranged from -2.23 to -1.09 μg/m3. During the 2-day peak WFS period, northern regions of the city (e.g., Northwest Side, North Side) exhibited higher PM2.5 concentrations (differences of 2.29 and 1.81 μg/m3, respectively) compared to the Southwest Side. After the peak, PM2.5 levels in southern regions decreased more slowly than in northern and central Chicago. SIGNIFICANCE:WFS is degrading air quality even in areas far from wildfires. During WFS peak periods, no region was spared from marked increases in PM2.5 from transported WFS. Lower income and higher minority communities had higher baseline PM2.5 levels and took longer to return to pre-peak citywide average PM2.5 concentrations, indicating that although WFS impacts are widespread, some regions may experience more severe effects than others. IMPACT:This is one of the first studies to examine 2023 wildfire smoke exposure patterns in Chicago, a major urban area that is typically distant from wildfires. Using exposure modeling and region-specific temporal analysis, this study reveals that wildfire smoke (WFS) from distant fires exposed all Chicago communities to substantial increases in exposure to fine particulate matter (PM2.5). However, lower-income neighborhoods with a higher proportion of Black or Hispanic residents, shorter life expectancy, and higher burdens of chronic disease faced both higher baseline PM2.5 levels and longer time until PM2.5 concentrations returned to pre-peak levels. These findings underscore the importance of considering within-city variation in pollution levels and recovery patterns when assessing the public health impacts of wildfire-related air pollution events.
BACKGROUND:Extreme heat is an intensifying environmental stressor, with pregnant women identified as a highly vulnerable population. Although prior researches have examined heat exposure and preterm birth (PTB), inconsistencies in heat definitions and limited evaluation of effect modifiers hinder clear interpretation. OBJECTIVE:To systematically evaluate the association between extreme heat exposure (including temperature percentiles and heatwaves) and PTB risk. METHODS:We systematically searched five major databases (through June 2025) for studies on extreme heat ( ≥ 90th percentile) or heatwaves and PTB. Risk of bias was assessed using the WHO tool for epidemiologic studies. Random-effects meta-analysis estimated pooled relative risks (RRs), and subgroup analyses were conducted to evaluate potential modifiers. Certainty of evidence was evaluated using the WHO-adapted framework. (PROSPERO: CRD420251066785). RESULTS:In total, 49 studies from 13 countries were included. Both short-term exposure ( ≤ 4 weeks before delivery; RR = 1.07, 95% CI: 1.05-1.09) and long-term exposure ( > 4 weeks before delivery; RR = 1.27, 95% CI: 1.18-1.36) significantly increased PTB risk. For short-term exposure, the 90th percentile threshold yielded the largest effect (RR = 1.20, 95% CI: 1.12-1.29; p < 0.001). Long-term associations were stronger in studies spanning multiple climate zones (RR = 1.42; 95%CI: 1.24-1.62; p = 0.024). Key modifiers included exposure window, geographic region, study design, and country income level. SIGNIFICANCE:Extreme heat exposure is significantly elevated PTB risk, particularly during late pregnancy and through cumulative long-term exposure. These findings emphasize the urgent need for targeted public health interventions and climate adaptation strategies to protect maternal and neonatal health in heat-vulnerable and low-resource regions. IMPACT:This study provides the first systematic review and meta-analysis that specifically applies standardized percentile-based definitions of extreme heat to examine its association with preterm birth (PTB). Crucially, it quantifies risk for both short-term and long-term exposure, finding that long-term exposure throughout pregnancy poses a significantly higher risk. By identifying specific vulnerability windows and the heightened risk in multi-climate zones, these results offer a framework for clinicians and policymakers to implement timely heat-health alerts and protective infrastructure for pregnant populations.
BACKGROUND:Indoor air pollution is a major contributor to personal exposure to air pollution, yet accurately modeling indoor concentrations remains a major challenge in epidemiological studies. This issue is especially relevant since we spend a great portion of our time indoors. OBJECTIVE:To develop a two-stage model to estimate indoor nitrogen dioxide (NO2) concentrations. METHODS:The study was conducted in the Barcelona Life Study Cohort (2018-2021), a cohort of pregnant individuals in Barcelona, Spain. Indoor and outdoor NO2 concentrations were measured using passive NO2 samplers placed in participants' homes at weeks 12 and 32 of pregnancy. A total of 1695 indoor measurements and 1577 outdoor measurements were collected. In the first stage, we modeled the indoor-outdoor (I/O) NO2 ratio as a proxy for infiltration using a linear mixed-effects model with repeated measures using 1528 1-week integrated I/O NO2 measurements. In the second stage, we used a random forest algorithm to estimate weekly indoor NO2 concentrations covering the full pregnancy period, incorporating the predicted I/O ratios, a previously developed hybrid-model outdoor NO2 estimate, meteorological data, and home characteristics. RESULTS:The I/O NO2 ratio model showed moderate performance, with a cross-validated R2 of 0.27 under leave-one-subject-out (generalized model) validation and 0.70 under leave-one-observation-out validation (cohort performance). The indoor NO2 model achieved a 10-fold cross-validated R2 of 0.53 and RMSE of 1.28 µg/m3. Finally, we estimated indoor NO2 concentrations for each week of pregnancy using a quantile random forest (QRF) algorithm incorporating uncertainty estimation through prediction intervals. SIGNIFICANCE:The findings demonstrate that accounting for dynamic infiltration processes and individual-level behaviors improves the estimation of indoor NO2 exposure. IMPACT:Integrating indoor and outdoor NO2 measurements in a two-stage modeling framework and combining it with key indoor determinants improves indoor NO2 estimation.
BACKGROUND:Population-specific, age- and sex-stratified reference intervals (RIs) for many serum trace elements remain limited, constraining the clinical interpretation of inductively coupled plasma mass spectrometry (ICP-MS)-based measurements. OBJECTIVE:To derive age- and sex-stratified serum reference intervals-not previously available for a large Japanese screening cohort-for eight trace elements (Co, As, Se, Rb, Sr, Mo, Ag, and Cs), and to assess between-facility and between-year variation. METHODS:A predefined 17-element panel was measured by ICP-MS in adults participating in an optional screening initiative at 12 sites in Japan (2017-2020). After age and eligibility exclusions, deduplication to one sample per participant, and removal of two sites with fewer than 20 participants, 17,523 individuals from 10 sites were retained for analysis. From the 17-element panel, the eight target analytes were pre-specified for RI derivation. RIs were estimated using a CLSI EP28-A3c-oriented statistical framework with Box-Cox transformation and bootstrap 90% confidence intervals. Pooling decisions were guided by Kruskal-Wallis effect sizes and RI width ratios. RESULTS:All eight elements showed age- and/or sex-related variation. As and Mo were generally higher in older age groups in both sexes, and Ag and Cs also tended to be higher in older groups (non-monotonic patterns), whereas Rb was generally lower in older groups. Se was relatively stable in men but increased up to the 60-69-year age group in women. Co was higher in women than in men, with a higher upper reference limit at ages 20-49. Sr upper reference limits generally increased with age in both sexes, especially in women aged ≥70 years. Between-year and between-site effect sizes were small overall, supporting pooled RI construction. SIGNIFICANCE:This study provides large-scale, age-decade- and sex-stratified descriptive serum RIs for eight trace elements that previously lacked established Japanese reference intervals. These intervals support contextual interpretation of ICP-MS biomonitoring data. IMPACT STATEMENT:This study provides large-scale, age-decade- and sex-stratified, descriptive serum concentration intervals for eight trace elements that lack established age- and sex-stratified serum reference intervals for Japanese adults (Co, As, Se, Rb, Sr, Mo, Ag, Cs). Derived from a multicentre screening population rather than a formally defined healthy cohort, these descriptive intervals support the contextual interpretation of ICP-MS-based biomonitoring data and provide a basis for future toxicological and nutritional assessments.
BACKGROUND:Communities near mine tailings face elevated environmental risks, but age-specific factors, behaviours, and household characteristics remain insufficiently characterised. Understanding community-specific patterns of indoor and outdoor time spent, as well as residential factors, is essential for human health risk assessment across different demographic groups. OBJECTIVES:This study examined age-related time-spent patterns among residents living near mine tailings storage facilities in the West Rand district, South Africa. METHODS:A cross-sectional survey was conducted among 330 households in Kagiso, West Rand. Data were collected on demographics, housing characteristics, and indoor and outdoor time using a pre-tested structured questionnaire. Descriptive statistics were computed to summarise participant characteristics and time-spent patterns. Group differences across five age categories were assessed using Kruskal-Wallis tests with Bonferroni-adjusted Dunn's post-hoc comparisons after testing normality with the Shapiro-Wilk test. RESULTS:Indoor time accounted for 66.6-84.8% per day on weekdays and 76.0-82.7% on weekends, averaging 18.28 ± 3.02 h on weekdays and 19.00 ± 1.99 h on weekends. Young children and elderly participants spent the most time indoors, significantly more than other age groups (p < 0.05). Weekend indoor time increased significantly among children, adolescents, and adults (p < 0.05), but not among young children and elderly participants. Outdoor and recreational time varied significantly by age (p = 0.015), with higher participation among children and adolescents than among other age groups. Post-hoc analyses confirmed significant age-group differences in indoor time and recreation between younger and older participants. SIGNIFICANCE:This study revealed that the time West Rand residents spent indoors was nearly 10 percentage points lower than that reported by the USEPA. Significant demographic group differences suggest that time-spent patterns may lead to differential exposure to indoor and outdoor pollutants. The findings highlight the need for age-specific exposure assessments and tailored environmental health interventions in mine-adjacent communities. IMPACT STATEMENT:This study presents the characterisation of age-specific time-spent patterns among residents of a community near mine tailings. In a South African sample, we showed that indoor and outdoor time-spent patterns differ from the default assumptions in the USEPA Exposure Factors Handbook. Variation observed across age groups indicates that reliance on generic exposure parameters may misrepresent population-specific risks. These findings demonstrate that exposure assessments and interventions must be age-specific, reflecting the distinct patterns and profiles of different demographic groups. Furthermore, they emphasize the importance of locally generated demographic, time-spent, and housing data for accurately characterizing environmental risks in mining-affected communities.
BACKGROUND:The increasing use of plasticizers as substitutes for heavily regulated phthalates raises concerns about human exposure. However, exposure patterns and suitable biomarkers for exposure assessment remain poorly understood. OBJECTIVE:This study aimed to evaluate temporal variability of plasticizer metabolites in consecutively collected urine samples and to estimate exposure levels using toxicokinetic (TK) modeling. METHODS:Twenty metabolites of six compounds, di-isononyl phthalate (DiNP), di-isodecyl phthalate (DiDP), di-ethylhexyl adipate (DEHA), di-ethylhexyl terephthalate (DEHTP), di-isononyl cyclohexane (DINCH), and tri-octyl trimellitate (TOTM) were quantified in 218 urine samples collected from six adults over one week. Intraclass correlation coefficients (ICC) and bivariate correlations were calculated to evaluate temporal variability across biomarkers. Time-activity diaries were used to identify exposure sources, and TK modeling was applied for biomonitoring-based reverse dosimetry. RESULTS:Metabolites of DiNP, DEHA, and DEHTP were frequently detected (>70%). First morning voids correlated moderately with daily average concentrations for frequently detected biomarkers (Spearman's ρ = 0.34-0.62, p ≤ 0.05). ICCs ranged from 0.16 to 0.73, with carboxylated metabolites demonstrating fair-to-good temporal stability (ICC = 0.57-0.68) for DiNP, DEHA and DINCH, while the hydroxylated metabolite showed the highest reproducibility for DEHTP (ICC = 0.62). Dietary intake, particularly processed foods, appeared to be the major exposure route, as concentration peaks of DiNP and DEHTP metabolites coincided with recorded dietary events. Daily intake estimates reconstructed from TK models under intermittent ingestion scenarios were generally lower than under continuous exposure assumptions, reflecting systematic differences. For most compounds, mass-balance approaches were comparable to TK model estimates, though DEHA showed a discrepancy attributable to the metabolite-specific model structure. IMPACT:Phthalate and non-phthalate substitutes increasingly replace heavily regulated phthalates, yet methodological frameworks for biomonitoring-based exposure assessment remain limited. We combined serial urine sampling, time-activity diaries, and toxicokinetic modeling as a case study for comprehensive exposure assessment. Key findings include: (1) secondary metabolites demonstrated fair-to-good temporal stability, while first morning void samples showed moderate correlations with daily exposure; (2) time-activity diaries identified potential dietary sources; and (3) toxicokinetic modeling accounting for realistic exposure scenarios demonstrated advantages over mass-balance approaches. These findings provide practical guidelines for individual-level assessments when environmental monitoring is limited.
BACKGROUND:Indoor contamination by agricultural pesticides and its determinants remain poorly characterized. OBJECTIVE:This study investigates the influence of outdoor environmental factors, residential characteristics, and occupant behaviors on pesticide contamination inside homes. METHODS:In 2020 and 2021, we collected surface wipe samples from 31 homes near vineyards during the peak pesticide spraying season, including dusty high surfaces (tops of doors, windows, shelves, furniture; n = 116), frequently touched surfaces (n = 272), and floors (n = 128). Eight fungicides were analyzed. Data on household characteristics and behaviors, home features and local environmental factors were gathered through questionnaires and databases. To capture some key processes of pesticide intrusion and removal, we developed specific scores from factors related to cleaning practices, air exchange, and track-in via residents and pets. We applied structural equation modeling to estimate both direct and indirect effects of potential determinants on indoor pesticide contamination for the three surface types. Three latent variables were included in the models: outdoor and indoor contamination, estimated through measured pesticide surface loadings, and household structure, estimated by socio-demographic data on occupants. Results were reported as standardized β coefficients. RESULTS:Only two models achieved satisfactory fit levels (high surfaces and frequently touched surfaces). For high surfaces, pesticide contamination was significantly associated with local annual pesticide purchases (β = 0.48, p = 0.037), occupant track-in (β = 0.22, p = 0.033), and households with active adults and children (β = -0.48, p = 0.031), while air exchange in the room had no significant effect (β = -0.03, p = 0.74). For frequently touched surfaces, contamination was positively associated with local annual pesticide purchases (β = 0.79, p < 0.001), air exchange (β = 0.16, p = 0.02), and occupant track-in (β = 0.12, p = 0.06). In contrast, cleaning was negatively associated with contamination (β = -0.36, p < 0.001) while the presence of active adults and children was not (β = -0.11, p = 0.53). SIGNIFICANCE:These findings refine our understanding of indoor pesticide contamination sources and can inform targeted interventions to reduce exposure among rural populations. IMPACT:Our research provides new insights into residential pesticide contamination in agricultural areas through an extensive collection of surface samples from homes near vineyards. Our results suggest a spatial and temporal gradient of indoor contamination. High surfaces may accumulate long-term pesticide residues, while floors could reflect more recent contamination. Frequently touched surfaces showed intermediate contamination. Outdoor-indoor transfers, driven by air exchange and track-in, appear to affect these surfaces differently. Frequent cleaning seems to help reduce contamination, and household characteristics may influence pesticide distribution. These findings may guide further research efforts to better understand contamination pathways and help refine future mitigation strategies.
BACKGROUND:Ambient volatile organic compounds (VOCs) exist as complex mixtures. Their short-term health associations in children remain poorly characterized. OBJECTIVE:To evaluate short-term associations between ambient VOC mixtures and pediatric atopic dermatitis (AD) and asthma in Seoul, Korea, and to estimate the ecological attributable burden. METHODS:Daily counts of outpatient visits and hospital admissions among individuals aged <19 years in Seoul (2015-2019) were linked with citywide VOC concentrations grouped as n-alkanes (G1), branched/cyclo-alkanes (G2), alkenes/alkynes (G3), and aromatics (G4). Mixture associations were estimated using Weighted Quantile Sum regression (WQSR) and Bayesian Kernel Machine Regression. Population Attributable Fractions (PAFs) were calculated from WQSR-based relative risks under a hypothetical 25th-percentile lower-exposure scenario. RESULTS:Ambient VOC mixtures were associated with higher daily counts of outpatient visits and hospital admissions for pediatric AD and asthma. Toluene (G4) showed the highest WQSR weight (0.88 for AD; 0.75 for asthma), while hexane (G1) and cyclohexane (G2) also showed relatively high weights (AD: 0.67 and 0.38; asthma: 0.55 and 0.25, respectively). Generalized additive models showed broadly consistent single-compound associations. Estimated ecological PAFs under the 25th-percentile scenario were 7.0% for AD (G4) and 4.7% (G1), 4.0% (G2), and 6.5% (G4) for asthma. SIGNIFICANCE:Short-term exposure to ambient VOC mixtures, particularly aromatics, is associated with an increased pediatric burden of AD and asthma in Seoul. Toluene was identified as a key contributor, whereas the roles of other VOCs require further investigation because of potential shared source contributions with other pollutants. IMPACT:This study quantifies the short-term effects of ambient VOC mixtures on pediatric AD and asthma using advanced mixture models. The findings suggest that aromatic VOCs, particularly toluene, were consistently associated with the observed health outcomes. VOC mixture exposure may be associated with a measurable ecological burden of pediatric AD and asthma at the population level.
BACKGROUND:Limited data exist on environmental exposures experienced by preterm infants in the neonatal intensive care unit (NICU). OBJECTIVE:To determine the timing and magnitude of a broad array of metals during gestation and NICU hospitalization using shed teeth as a biomatrix. METHODS:Teeth from participants in the Developmental Impact of NICU Exposures (DINE) cohort were analyzed for Al, As, Ba, Bi, Ca, Cd, Co, Cu, Li, Mg, Mn, Mo, Ni, Pb, Rb, Sn, Sr, and Zn using laser ablation techniques. The neonatal line was used to assign prenatal and postnatal age estimates to metal concentrations relative to Ca in discrete regions of teeth. We evaluated changes in metal ratios across the prenatal/postnatal periods and measurement week using linear mixed effects models and calculated estimated marginal means for each week in the prenatal and postnatal periods. RESULTS:Metals were measured in 138 participants (median gestational age: 27, IQR 25-28 weeks) and linked to developmental weeks in utero or in the NICU. Averaged differences across the prenatal and postnatal periods decreased for Pb, Ba, Sn, Zn, Mn, Al, Mg, and Cu, and increased for Sr, Rb, and Li. Prenatal ratios were highest and larger than postnatal ratios in weeks 28-30 for Al and Ba. Prenatal Zn was highest in weeks 24 and 30 and Mn was highest during weeks 24 and 25. SIGNIFICANCE:We provide the first comprehensive evaluation of metals during gestation and NICU stay in preterm infants, with postnatal levels generally lower than prenatal levels for most metals. This approach has relevance to ascertaining exposure sources, influences on clinical outcomes, and provides insights not previously possible in cohorts where all third-trimester exposures were incurred in utero. IMPACT:This study represents the first analysis of metals spanning gestation and the NICU stay in a preterm infant population using shed teeth as a biomatrix. Metal levels differentiated across the third trimester prenatally or postnatally in the NICU provide unique insights for understanding exposure sources and health outcomes.
BACKGROUND:We are developing a cloud-based neurohealth data platform to aggregate and store data from large, complementary historical and ongoing studies focused on traumatic brain injury (TBI), a condition shaped by complex interactions among biological, environmental and contextual factors, and to enable secure, flexible querying across studies. These multifaceted datasets span clinical, physiological, behavioral, and environmental domains, mirroring the complexity of comparable multisite efforts such as the exposome moonshot. OBJECTIVE:Define a scalable and interoperable framework for organizing and harmonizing multifaceted data derived from different sources, purposes, and time scales, supporting precision neurohealth research and translational applications. METHODS:We designed and have begun implementing the Precision Health for Enhanced Neuro-Operational Modeling (PHENOM) platform, using a hybrid lakehouse architecture deployed on the public cloud. PHENOM employs automated extract-transform-load pipelines, metadata templates, and standardized biomedical terminologies to support FAIR-compliant data integration and cross-domain interoperability. We have mapped our findings to our datasets of interest in traumatic brain injury (TBI) and to the knowledge being generated for moonshot exposome efforts. RESULTS:The PHENOM framework organizes data into four general "bins": (1) One-time capture, (2) regular or continual lab or site visits, (3) personalized electronic health records, and (4) collective measures. The architecture enables ingesting multimodal data and supports AI-driven analytic tools and digital twin modeling through public cloud services. Initial deployment has demonstrated automated harmonization across multiple TBI studies and scalable query performance across datasets. SIGNIFICANCE:Scientists supported through large research programs, such as those funded by government initiatives or consortia, are increasingly encouraged (or required) to share their diverse data in public repositories in FAIR format. The proposed framework provides a flexible, cloud-based platform that meets these requirements by enabling the harmonization and secure sharing of multifaceted data across studies. Designed for scalability and interoperability with established data models, the platform supports both current research needs and a flexible structure responsive to those needs for use across future large-scale precision health initiatives. IMPACT STATEMENT:This Perspective proposes a cloud framework for structuring multifaceted data derived from complementary or consortia-related studies using four general "bins": (1) One-time capture, (2) regular or continual lab or site visits, (3) personalized health records, and (4) collective measures. The proposed framework offers a flexible structure for use across many existing research programs that are increasingly encouraged (or required) to share their diverse data in public repositories in FAIR format.
BACKGROUND:Personal care products (PCPs) are a ubiquitous, modifiable source of chemical exposures. Although numerous studies have characterized behaviors and perceptions around adult PCP use, little is known about PCP use in adolescence, a critical developmental period. OBJECTIVE:Characterize adolescent PCP use in New Jersey, USA in relation to sociodemographic characteristics, precautionary behaviors, and social media consumption. METHODS:With a local high school, we designed and administered a pilot survey to students (n = 143). Students reported on PCP use (within the past 24 h as well as general habits), precautionary behaviors they use when choosing products (e.g., using healthy product apps), and social media habits. For each participant, we tallied the number of products used overall and by product type (e.g., skin care, cosmetics). We fitted multivariable negative binomial models examining the number of products used in relation to precautionary behaviors and social media consumption. Secondarily, we fitted models restricted to females. RESULTS:Females used twice as many products as males in the past 24 h (median (inter-quartile range), females: 14(8), males: 7(3)). Although 41% reported avoiding products containing certain chemicals, few students engaged in precautionary behaviors when selecting products, such as reading ingredient lists (19%) or using healthy product apps or websites (5%). In multivariable models, few associations were observed between precautionary behaviors and the number of PCPs used. However, there were consistent positive associations between social media consumption and the number of products used. For example, above median daily time spent on social media (Incidence Rate Ratio [IRR]: 1.05, 95% CI: 1.00, 1.09), following influencers with product lines (IRR:1.30, 95% CI: 1.11, 1.52) and watching hair/beauty tutorials (IRR:1.40, 95% CI: 1.21, 1.63) were associated with greater PCP use in the last 24 h. Results were robust to analyses of occasional PCP use (at least once in the past 6 months) as well as in analyses restricted to females. IMPACT:In adolescents, PCP use was comparable to prior studies among adults, while engagement in precautionary behaviors that could reduce chemical exposures was infrequent. Higher social media engagement was associated with greater PCP use. Interventions leveraging social media and influencers may be an effective strategy to improve environmental health behaviors related to PCP use.
BACKGROUND:Short-term air pollution exposure may exacerbate respiratory health outcomes in vulnerable populations, particularly those with pre-existing cardiovascular conditions. Given the persistent burden of COVID-19 pneumonia, there is a need to better understand how environmental exposures influence risk of COVID-19 pneumonia. OBJECTIVE:We aimed to explore whether associations between short-term particulate matter (PM2.5) exposure and COVID-19 pneumonia risk are stronger among those with pre-existing cardiovascular conditions. METHODS:We used NYC COVID-19 hospitalization records (March-December 2020; N = 15,361) and defined short-term PM2.5 as a 7-day residential ZIP Code average prior to and including hospitalization, using Environmental Protection Agency (EPA) Community Multiscale Air Quality (CMAQ) modeled data. Pre-existing cardiovascular conditions included 15 ICD-coded circulatory conditions (e.g. myocardial infarction, atrial fibrillation, ischemic heart disease). Modified Poisson regression estimated risks of acute respiratory distress syndrome, pneumonia, ventilation use, and dialysis use, and Cox hazards regression estimated COVID-19 fatality risk and length of stay, adjusting for age, sex, smoking status, and environmental vulnerability. Relative excess risks of interaction (RERI) estimated additive interaction between short-term PM2.5 exposure and cardiovascular conditions on risk of COVID-19 pneumonia during hospitalization. Targeted maximum likelihood estimation estimated average treatment effects (ATE) of short-term PM2.5 exposure on COVID-19 pneumonia, among those with and without cardiovascular conditions. RESULTS:Short-term PM2.5 and pre-existing cardiovascular conditions together increased pneumonia risk from July to December 2020. For example, joint exposure to an interquartile range increase in short-term PM2.5 and pre-existing myocardial infarction amplified pneumonia risk (RR:1.67; 95%CI:1.41,1.99), with smaller increases observed for PM2.5 alone (RR:1.03; 95%CI:0.95,1.12) or myocardial infarction alone (RR:1.34; 95%CI:1.14,1.58). Upper-quartile short-term PM2.5 increased pneumonia risk only among individuals with cardiovascular conditions (ATE:0.05, 95%CI:0.02,0.08). IMPACT:COVID-19 pneumonia drives substantial hospitalizations, but little is known about the link between environmental exposures and pneumonia. We aimed to examine whether short-term air pollution influences pneumonia risk, and if pre-existing cardiovascular conditions modify this link. We found that pre-existing cardiovascular conditions increase vulnerability to short-term air pollution, amplifying COVID-19 pneumonia risk during hospitalization. These study findings may help identify populations that may be more vulnerable to the effects of short-term air pollution exposure during future respiratory infectious disease outbreaks.
BACKGROUND:Mercury, a toxic metal, is added to some cosmetic skin care products as a lightening agent, due to its ability to inhibit melanin production, and it has also been found in products intended for blemish removal, acne, and anti-aging. Skin lightening products containing mercury were first found in New York City (NYC) in 2005 as a result of NYC Health Department biomonitoring. OBJECTIVE:We analyzed the mercury concentrations and product descriptions of skin lightening products collected from NYC stores between 2009 and 2022 as part of public health investigations. We described the results of store surveillance and product testing and characterized mercury-containing products. METHODS:We conducted surveys of shops in NYC, focusing on those located in communities of color. We purchased and tested 197 products using the Manual Cold Vapor Technique for total mercury. Products were categorized by country of manufacture, type, intended use, and other label information. Descriptive statistical analyses were conducted. RESULTS:Mercury was detected in 29% (n = 56) of products, and 18% (n = 33) of products contained mercury above the U.S. Food and Drug Administration permissible limit of 1 part per million (ppm) for mercury in cosmetics. All products containing mercury were manufactured abroad. Eleven products had mercury concentrations above 10,000 ppm, with a maximum concentration of 27,000 ppm. Only 4% (n = 7) of product labels listed mercury as an ingredient. SIGNIFICANCE:Skin lightening products containing mercury were found for sale at NYC stores, reaffirming pre-existing concerns about mercury exposure for product users and their families. This exposure risk is a health equity and environmental justice issue rooted in colorism and racist beauty standards. Increased awareness of this issue among communities of color and health care providers; collaboration between governmental and nongovernmental partners; and stricter regulation, enforcement, and compliance structures are potential strategies to address this complex issue. IMPACT STATEMENT:Mercury was found in approximately one third of skin lightening products purchased and tested by the New York City Health Department between 2009 and 2022. Products containing mercury were all manufactured abroad and were marketed to lighten the skin, treat under eye circles, clear acne or blackheads, reduce aging and wrinkles, and remove spots, marks, blemishes or freckles. With mercury rarely listed as an ingredient, mercury concentrations thousands of times higher than regulatory limits in some products, and marketing towards women of color, these products pose an environmental health and health equity issue rooted in colorism and racist beauty standards.
BACKGROUND:Metals, including lead, are frequently detected in tap water. Biomarker-based studies link elevated levels of lead with poorer child cognitive outcomes, including executive function (EF) skills. Nevertheless, few studies have characterized metal exposures in child care center and home water samples or examined whether tap water concentrations of lead alone are associated with child EFs. OBJECTIVE:We aimed to (1) characterize metal concentrations in tap water from child care kitchens, classrooms, and homes, (2) identify sociodemographic and environmental predictors of waterborne lead, and (3) test associations between lead and child EF skills. METHODS:Tap water from 51 North Carolina (NC) child care kitchens, 120 classrooms, and 138 homes of 297 children (2-6 years old) were collected and analyzed for 15 metals via EPA Method 200.8. Child EF skills were assessed through a performance-based assessment and a parent questionnaire. Regression models related waterborne lead to sociodemographic characteristics and water source. Path models evaluated associations between lead from samples in each setting and EF skills. RESULTS:Lead was detected in 67% of child care kitchen, 34% of classroom, and 57% of home kitchen drinking water samples. While 8% of child care kitchen, 3% of classroom, and 7% of home kitchen water samples exceeded the American Academy of Pediatrics reference level of 1 part per billion (ppb) for lead, none exceeded the NC state action level and U.S. Environmental Protection Agency standard of 10 ppb. Lead concentration in child care kitchen water was negatively associated with inhibitory control. SIGNIFICANCE:Metals were frequently detected in child care center and home water samples. Although waterborne lead concentrations were below the state action level, child care kitchen lead was negatively associated with child EF skills. IMPACT STATEMENT:Key metals, including lead, were frequently detected in water samples collected from child care kitchens (N = 51), classrooms (N = 120), and homes (N = 138) of children ages 2-6 years old. Though lead levels were below state and federal guidelines, the waterborne lead in the child care center was negatively associated with children's executive function skills (N = 297). The absence of a link between home water lead and children's executive function skills may stem from selection bias, as families at higher risk for lead were underrepresented in home sample collection. The results underscore the frequency of exposure to, and potential impact of, neurotoxic contaminants in children's daily environments.
BACKGROUND:Following Superstorm Sandy (SS), domestic fungal growth and the associated respiratory health impacts concerned New York City (NYC) residents, especially those with asthma. OBJECTIVES:Test hypotheses that compared to non-damaged homes, SS-damaged homes would have different fungal profiles and children living in SS-damaged homes would have higher asthma morbidity. METHODS:We recruited 59 families with asthmatic children from SS-damaged NYC homes 6-18 months following the storm to compare with 68 asthmatic children from non-damaged homes in a concurrent study. Home visits included questionnaires, physical health assessments, and settled floor dust collection. We measured fungi in dust using quantitative polymerase chain reaction (qPCR) and next-generation DNA sequencing (NGS). RESULT:Compared with non-damaged homes, SS-damaged homes had lower fungal diversity (Shannon index) and higher total and allergenic fungal concentrations. NGS and qPCR revealed different genera compositions (e.g., elevated Epicoccum and Cladosporium in damaged homes). Repeat measures in damaged homes revealed that the persistence of these changes was variable (e.g. fungal diversity increased; sustained increase in Epicoccum over the following year). There were no statistically significant differences in asthma symptom frequency, atopy to fungi, lung function, or exhaled nitric oxide among children living in SS-damaged vs. non-damaged homes. Findings were consistent when among SS-damaged homes either with mold odors or visible mold. SIGNIFICANCE:qPCR and NGS revealed that SS-damaged homes experienced persistently higher total and allergenic fungal concentrations and lower diversity, compared to non-damaged homes. Such fungal profiles have previously been associated with asthma development and exacerbation. While we identified health-relevant differences in fungal profiles, we did not observe differences in health outcomes between SS-damaged and non-damages homes. We elaborate on potential reasons for the null health findings including: the study population's relatively mild asthma, recruitment timing, statistical power, and ceiling effects. IMPACT:Despite the wide population impacted by Superstorm Sandy, and the popularization of terms like "Sandy cough," there have been no studies, to our knowledge, characterizing changes in the fungal profile in Sandy-damaged homes and their asthma-related health impacts. Comparing Sandy-damaged homes to non-damaged homes, quantitative polymerase chain reaction and next-generation DNA sequencing methods revealed higher total and allergenic fungal concentrations and lower diversity in damaged homes- profiles that have previously been associated with asthma development and morbidity. However, these fungal differences were not consistently associated with asthma outcomes.
BACKGROUND:Although suspected to alter child development, exposure to glycol ethers (GEs) is widespread in the general population and remains insufficiently characterised, especially given their short half-lives. OBJECTIVE:We investigated the variability of urinary concentrations of GE metabolites in pregnant women using different sampling methods. METHODS:Thirty pregnant women were enroled by medical professional around Rennes, France. They completed questionnaires and underwent a repeated urine sampling protocol to collect all urine voids for one day and then two consecutive days, several weeks apart. We then measured the concentrations of nine GE metabolites in three pooled 24-h samples and three first morning void samples per woman, and all individual samples collected over two days (N = 530). We assessed the temporal variability of urinary GE metabolite concentrations for six sampling methods, calculating intraclass correlation coefficients (ICCs), using mixed-effects models with random effects for each day and participant. RESULTS:The highest ICC ( > 0.75) was observed for sampling methods involving the collection of one or three random urine voids within a day for 2-methoxyproprionic acid (2-MPA), and for the first morning void for both 2-MPA and phenoxyacetic acid (PhAA). For other GE metabolites, none of the sampling methods seemed to offer good enough reproducibility, with ICC < 0.63. SIGNIFICANCE:Among the methods assessed, FMV seemed to provide the best balance between reproducibility for PhAA and 2-MPA. For other metabolites, no sampling method seemed to provide sufficient reproducibility. IMPACT:Perinatal health concerns have been reported in association with exposure to glycol ethers (GEs). Reproducible methods for measuring GE exposure in pregnant women are therefore required. Metabolite-specific sampling strategies may be appropriate to investigate exposure-response associations.
BACKGROUND:Lead exposure in U.S. communities is typically attributed to indoor sources, particularly lead-based paint in pre-1978 homes. However, exterior sources may contribute substantially to lead contamination, especially in heavily impacted communities. OBJECTIVES:To determine whether soil lead is tracked into homes in East Trenton, NJ, a community with extensive historical environmental lead contamination, using a community-based participatory approach. METHODS:Through a community-academic partnership, community scientists sampled soil from 122 homes in East Trenton. In a subset of 42 homes, lead-based paint was measured using X-ray fluorescence, and dust lead loadings were measured from interior floors, windowsills, and window wells using a modified HUD method. RESULTS:Of 242 soil samples, 86.4% exceeded EPA's 200 ppm residential lead hazard level. In homes built before 1978 with lead-based paint present (n = 33), mean (SD) dust lead loadings were: floors (n = 72) 135.4 (311.9) μg/ft², windowsills (n = 39) 3,372.2 (19,688.4) μg/ft², and window wells (N = 36) 7,675.1 (34,041.1) μg/ft². In homes without lead-based paint (n = 11), loadings were: floors (n = 20) 263 (1006) μg/ft², windowsills (n = 11) 23.6 (26.2) μg/ft², and window wells (n = 10) 312.5 (307.1) μg/ft². Wilcoxon ranked sum test showed no observed difference in loadings on floors between homes with and without lead paint identified (p = 0.28); however, there were significant differences in lead loadings for both windowsills (p < 0.001) and window wells (p = 0.009), with homes with lead paint having higher windowsill and window well lead loadings. Notably, in homes without lead-based paint, 80% of floor samples exceeded 10 µg/ft², indicating substantial lead hazard without the presence of lead-based paint. SIGNIFICANCE:Elevated dust lead in homes built or renovated after 1978 without interior paint hazards suggests that lead is tracked indoors from exterior sources. Regulatory policies relying solely on housing age to assess lead exposure risk may overlook critical community-level contamination pathways, inadequately protecting residents in heavily impacted areas. IMPACT STATEMENT:Homes with no interior lead-based paint hazards are susceptible to interior dust lead loadings greater than the U.S. Environmental Protection Agency (EPA) hazard level for floor dust if located in a community with elevated lead soil levels.
