Joe C. Wen School of Population & Public Health

Air pollution, including particulate matter (PM10, PM2.5) and airborne metals (Cd, Hg, As, Pb) are a systemic health hazard, but osteotoxic effects remain underrecognized. We systematically synthesized human, animal, and in vitro evidence on airborne contaminants and outcomes to skeletal health. Following Office of Health Assessment and Translation (OHAT) guidance, we searched PubMed and Google Scholar (2014-2024; final search November 24, 2024) for English-language studies relating PM or airborne metals to skeletal outcomes, excluding studies limited to adult occupational exposures, age-related bone disease, or non-peer-reviewed reports. Studies were screened and data extracted in Covidence. Risk of bias was assessed using tools from the Quality Assessment and Risk of Bias Tool Repository, and certainty of evidence was graded with GRADE. We used a random-effects meta-analyses model to estimate relative risks (RRs) interpreted as directional consistency hazards for bone loss, impaired osteoblast differentiation, enhanced osteoclast activity, bone marrow adiposity, epigenetic changes, and altered bone microarchitecture. Fifty-three studies met the inclusion criteria. Across outcomes, meta-analyses showed higher hazard risk in exposed versus control groups, with RRs ranging from 7 to 10; for example, bone loss (RR 7.12, 95% CI 4.20-12.06) and altered bone microarchitecture (RR 10.28, 95% CI 4.67-22.61). Epidemiologic and experimental evidence may implicate oxidative stress, inflammation, and epigenetic dysregulation as key mechanisms. Overall certainty was low to moderate due to residual confounding, exposure misclassification, and heterogeneity. Growing evidence indicates that airborne PM and metals adversely affect bone health, particularly during critical developmental windows, underscoring the need to integrate skeletal endpoints into health assessment and policy.

Bicycle infrastructure preferences vary by user experience and comfort, but evidence suggests both cyclists and drivers perceive separated bike lanes as safer than shared lane markings (sharrows). Little is known about sharrows impact in real-world settings, as no observational studies have evaluated their before and after installation impact. Santa Ana, California, incorporated sharrows into its active transportation plan to increase and encourage safe riding in a downtown commercial corridor where separated bike lanes were not feasible. Researchers conducted a natural experiment using a pre-post study design, collecting baseline data in 2015 before installation, and follow-up data in 2016 and 2017. A total of 54 hours of direct bicycle observation was conducted across two target areas over three years (2015 – 2017). A modest increase in bicycling was observed from 2015 to 2016, but was not sustained into 2017. Sidewalk and wrong-way riding increased slightly over time, while participation by female riders and older adults (65+) declined post-intervention. Sharrows offer a low-cost, scalable approach for promoting road sharing and guiding cyclists toward safer positioning, particularly in resource-constrained urban areas. This study’s real-world evaluation underscores the importance of tailoring active transportation infrastructure to meet safety and accessibility needs for all. Findings suggest sharrows are effective when paired with complementary strategies, such as targeted education, driver awareness efforts, or enhanced infrastructure, to better serve diverse populations and promote long-term engagement in bicycling. Further research can build on these insights by integrating user-centered evaluation methods and emerging data technologies.

Over 25% of active NASA astronauts are women who will be exposed to low daily doses and dose rates of galactic cosmic rays (GCR) in space. We hypothesized that exposing mice to a preliminary simulated GCR mixed heavy ion beam composed of iron, silicon, and titanium ions induces follicle depletion and dose-dependent ovarian tumors. Female mice were exposed to 10, or 20 cGy each of Fe, Si, and Ti ions or sham-irradiation in quick succession within 15 min for total doses of 0, 30, or 60 cGy of the three beams. 16 months later, their ovaries were removed. Hyperplasia of the ovarian surface epithelium (OSE) was noted in 13%, 59%, and 22% of the 0, 30, and 60 cGy irradiated mice, respectively. The prevalence of mixed ovarian tumors was 0, 6, and 89%, respectively, in the 0, 30, and 60 cGy groups. Low numbers of Ki67 positive OSE and tumor cells supported a benign tumor phenotype. In a separate study, Si ion irradiation alone at 32 cGy did not induce ovarian tumors in mice; however, the mixed heavy ions at all doses and Si ion irradiation alone reduced the total number of healthy ovarian follicles. Mixed heavy ion exposure reduced lipid peroxidation, fibrosis, inflammation, and lipofuscin accumulation at 60 cGy compared to 0 cGy, but elevated inflammation and lipofuscin accumulation at 30 cGy compared to 60 cGy. Preliminary simulated GCR exposure causes ovarian follicle death and tumorigenesis. This study provides insight into space-radiation induced ovarian damage and cancer risk in females.

Radon is a naturally occurring radioactive gas that poses a serious health risk as the primary cause of lung cancer in non-smokers. Despite the well-known adverse association with health outcomes, current radon exposure assessments are limited to county-level or average-level estimates, which fail to capture regional variability. This study uses Machine Learning models, including Random Forest (RF) and Quantile Regression Forest (QRF), to estimate the indoor radon concentrations at the ZCTA (Zip code tabulation area)-level and characterize uncertainties in model estimates. Incorporating geological, meteorological, and building-specific data, the models aim to improve radon risk assessment by capturing mean exposure, variability, and extreme concentration levels. Processed radon test data (n = 718,111) were analyzed using average, variability, and quantile prediction methods. Models that estimate the average radon exposure at the ZCTA-level can yield promising model-fit results, but they do not capture the underlying variability of indoor radon exposure within a ZCTA. We utilize volatility analyses to identify characteristics indicative of high variability of indoor radon exposure. We also show that a QRF model can be used to estimate upper quantiles of residential radon exposure, thereby uncovering localized areas of elevated exposure that were not apparent in mean estimates. The results highlighted the need for a deep characterization of exposure risk and show that regions with moderate average exposure levels could still harbor extreme outliers with implications for evaluating health risks. Utilizing multiple radon exposure models allows for a deeper characterization of radon risk within a geographic area and can better identify high-risk areas. The results from this study provide a foundation for developing mitigation strategies and examining associations between radon exposure and health outcomes at fine scales. Future research should extend the geographic scope and incorporate additional environmental risk factors to establish a comprehensive framework for risk assessment.

Increasing adoption of zero-emission vehicles (ZEVs) has raised concerns about potential disparities, including increased gentrification in low-income communities. As a first step in unraveling this rarely studied relationship, we examined the association between ZEV adoption and measures of gentrification. We addressed two questions: (1) How do ZEV adoption trends vary by neighborhood-level gentrification and disadvantage in California metropolitan areas? and (2) Among neighborhoods that are not already gentrified, is prior ZEV adoption associated with subsequent changes in housing prices? We obtained census tract-level data on longitudinal counts of battery electric vehicle (BEV) and plug-in hybrid electric vehicle (PHEV) registrations, median home value and monthly rent, disadvantaged communities (DACs) designations, and gentrification status ('none/early,' 'advanced/stable,' or 'exclusive'). Linear mixed models estimated BEV (or PHEV) adoption trajectories (2015-2023) across gentrification status between DAC and non-DAC, allowing nonlinear time trends. Among 'none/early' gentrification tracts, we related prior ZEV adoption (2015-2019) to subsequent (2020-2023) home values and rental prices using linear mixed models. ZEV adoption trends varied by gentrification in non-DAC tracts, while DACs had lower levels of ZEV adoption and less evidence for disparities. Among 'none/early' gentrification tracts, tracts with higher BEV (or PHEV) adoption were associated with larger absolute increases in home value compared to tracts in the lowest adoption quartile. These early findings support community concerns that ZEV adoption may influence housing costs and gentrification. Further research is needed on these complex relations and possible policy interventions.

Methyl bromide (MeBr) has been widely used as a fumigant to control for pests, fungi and weeds as well as for disinfection of warehouses, shipping containers, and other commodities. MeBr is a known developmental, neurologic and respiratory toxin. Due to its ozone-depleting properties, MeBr was listed under the Montreal Protocol in 1992. While MeBr use was set to phase out by 2005, the Montreal Protocol and the US Clean Air Act allows critical use exemptions, such as fumigation of freight containers for quarantine and preshipment purposes. To evaluate state-wide spatial and temporal patterns, we examine publicly available pesticide data on the use of MeBr in California from 2016–2022. We found that MeBr applications continue in 36 out of 58 CA counties. For non-agricultural fumigation applications (e.g., commodity fumigation) of MeBr from 2016–2022, a total of 582,050 kilograms (1,283,201 pounds) were applied across 25 counties in CA, home to 24 million people. Los Angeles ranks as the highest use county, with a total of 269,571 or 46% of the total kilograms (594,302 pounds) applied from 2016–2022 for non-agricultural fumigation applications. Additionally, we characterized ambient MeBr concentrations in the West Long Beach community of LA County, based on a state monitor active since 2023, observing concentrations exceeding CA standards. This study underscores the importance of evaluating chemical phaseouts and improving enforcement and monitoring to ensure public health protections.

Bicycle infrastructure preferences vary by rider experience and comfort, yet both bicyclists and motorists perceive separated bicycle lanes as safer than shared lane markings (sharrows). Despite widespread adoption, evidence on the real-world behavioral impacts of sharrows remains limited. This study evaluates a natural experiment in Santa Ana, California, where sharrows were installed in a downtown commercial corridor when separated bicycle lanes were not feasible. Researchers collected baseline data in September 2015 and follow-up data in February 2016 and early 2017, totaling 54 hours of direct bicycle observation across two target areas. Results revealed a mixed and cautionary pattern of change. West-to-East mean bicycle counts increased significantly from 0.986 (2015) to 1.734 (2016; p = 0.045) but declined to 1.500 by 2017. East-to-West counts showed modest, non-significant year-over-year growth. Critically, sidewalk riding increased significantly from 2015 to 2016 (p = 0.002) and remained elevated through 2017 (p = 0.009), and wrong-way riding increased significantly from 2015 to 2016 (p = 0.003). Female ridership declined from 2015 to 2016 (p = 0.036) but recovered by 2017. Ridership among adults over 65 declined post-installation without meaningful recovery. Overall, sharrow installation was not associated with lasting ridership gains or improved safe bicycling behavior. Declines among women and older adults suggest sharrows may not adequately meet the needs of all roadway users. Drawing on the present data and prior literature, policymakers should consider pairing sharrows with education programs, driver awareness campaigns, and enhanced infrastructure to support safety, equity, and sustained bicycling engagement.

Developmental exposure to environmental toxicants is a cause of skeletal abnormalities. Yet the molecular mechanisms linking early exposure to impaired bone formation remain undefined. Skeletal tissues arise from both neural crest- and mesoderm-derived lineages that rely on shared osteogenic differentiation programs, suggesting that disruption of common regulatory processes may contribute to diverse skeletal outcomes. MicroRNAs (miRNAs) are key post-transcriptional regulators of gene networks and have appeared as indicators of toxicant-induced perturbation. In this study, we examined whether developmentally relevant toxicants are associated with miRNA regulatory networks during osteogenic differentiation. Using a human embryonic stem cell (hESC)-based osteogenic differentiation model, we assessed the effects of nine toxicants spanning distinct primary mechanisms. Toxicant exposure impaired osteogenic differentiation at their IC50, as reflected by altered expression of osteogenic markers and transcriptional remodeling. Global miRNA profiling revealed dysregulation of miRNAs enriched for bone-related biological processes, including regulators of osteogenic commitment and differentiation timing. Integrated miRNA-mRNA network analysis identified a subset of miRNAs linked to core osteogenic and lineage-associated pathways, including RUNX2-dependent transcription and BMP and Wnt signaling. Modulation of representative miRNAs produced osteogenic outcomes consistent with those observed following toxicant exposure and, in some cases, was associated with partial restoration of differentiation in exposed cultures. Collectively, these findings indicate that chemically diverse developmental toxicants are associated with miRNA-mediated regulatory patterns during osteogenic differentiation. Identification of shared miRNA regulatory features provides mechanistic insight into developmental bone toxicity and supports the use of miRNA network analysis as a human-relevant endpoint for skeletal hazard identification.