Air Pollution Exposure Assessment Research Articles

Calibration of Spatiotemporal Missing Data Imputation Algorithm in Distributed Space-Time Expectation-Maximization with Application in Recovering of Air Pollution Missing Data in Multi-Site Monitoring Network

TPS 691: Methods of measurement, design and data analysis, Exhibition Hall, Ground floor, August 28, 2019, 3:00 PM - 4:30 PM Background: A challenge in epidemiological studies of air pollution data is the issue of missing data. Air quality monitoring stations typically have some amount of missing data. A promising tool for spatiotemporal missing imputation of environmental data is Distributed Space-Time Expectation-Maximization (D-STEM) software. The D-STEM provides a general framework for concurrent modeling and missing imputation of air pollution data, which uses the prediction of a specified hierarchical space-time model in place of missing data in the iterative steps of Expectation-Maximization algorithm. Methods: We used the regression model to calibrate the prediction of the D-STEM software in each monitoring station, which considers the observed data as the response of the calibrating regression and the corresponding prediction of the D-STEM model as the only predictor. A simulation study that induced 10% extra missing data and repeated 50 times was used to compare the performance of missing data imputation algorithms. Moreover, two different underlying hierarchical space-time models were considered, which were (a) the simplest D-STEM model (a space-time model with just an intercept), and (b) an enriched model that benefited from a set of carefully selected predictors. These models were applied to data of fine particle matter, measured in 30 fixed monitoring stations in Tehran, Iran, where 48% of data was missing. To compare the competence of these methods in missing imputation, the Mean Absolute Percentage Error (MAPE) criterion was used. Results: The MAPEs of the D-STEM simplest model and the D-STEM enriched model were 26.1% and 25.3% in the original D-STEM spatiotemporal missing data imputation. The MAPEs of these two underlying models in the proposed calibration for the D-STEM spatiotemporal missing data imputation were 20.6% and 20.4%, respectively. Conclusions: The proposed calibration method substantially improved the spatiotemporal missing data imputation, which could benefit future exposure assessment and epidemiological studies of air pollutants.

Health risk and exposure assessment of indoor air pollutants in Shenzhen, China

Health risk and exposure assessment of indoor air pollutants in Shenzhen, China

Assessment of long-term exposure to airborne pollution in France (1990-2010)

TPS 652: Air pollution exposure modeling 2, Exhibition Hall, Ground floor, August 28, 2019, 3:00 PM - 4:30 PM Background: Long-term exposure to air pollution has been consistently associated with adverse health effects and reduced life expectancy. Air pollution is a mixture of multiple pollutants originating from a large variety of sources and includes various substances classified as carcinogenic or probably carcinogenic to humans. Identifying the long-term effects of these airborne pollutants requires the computation of the spatio-temporal variability of concentration to estimate the exposure of the population. Aim To provide an accurate assessment of long-term air pollution exposure (including airborne PM10, PM2.5, NO2, O3, BaP, PCB, PCDD/F and cadmium (cd) exposure in the whole France and in a case-control study nested in the French national prospective women’s cohort (E3N) Methods: Hourly airborne pollutant concentrations over the study period (1990-2010) were computed using a chemistry- transport model (CHIMERE) with a spatial resolution of 7x7km2. Cohort’s subjects were manually geocoded and annual concentrations were attributed to each subject’s address (n=19669). Results: Simulation scenarios of air pollutant concentrations show the variability of the pollutant sources over time with general decreasing levels of pollutant concentrations in air. At the E3N subject’s address, we observed a decline of concentration for all pollutants (PM10:-41%; PM2.5:-45%; NO2:-34%; PCB:-74%; PCDD/F:-87%, Cd:-68%; BaP:-45%) except for O3 (-0.04%). The percentage of subject exposed to an annual concentration exceeding WHO recommendation (PM10: 20 μg/m3; PM2.5: 10 μg/m3) decrease form 74% to 8% for PM10 and from 99% to 64% for PM2.5. We also observed strong territorial inequalities with differences in atmospheric concentrations that could be greater than a factor of five between regions. Conclusions: Findings from this study will be essential to increase the accuracy of the assessment of long-term airborne pollution exposure and will improve the results of epidemiological studies. The spatial resolution of concentrations will be improved by using complementary models such as land use regression.

NO2 Air Pollution Exposure Assessment in Urban Mysore, India

TPS 651: Air pollution exposure modeling 1, Exhibition Hall, Ground floor, August 27, 2019, 3:00 PM - 4:30 PM Introduction: In developing countries, rapid urbanization has led to decreased air quality. Few studies have applied spatially heterogeneous sampling to assess ambient air pollution levels in cities in India, to understand the important local predictors of air pollution exposure, which may be unique to Indian urban environments. Methods: Seasonal sampling was conducted in four campaigns (2016-2017) at 150 sites throughout Mysore, a small and rapidly growing city in south India. Nitrogen dioxide (NO2) levels were assessed using passive Palmes’ tubes and Ogawa badges. Results: Annual average NO2 ranged from 0.8 to 31.2ppb (1.5 to 58.7 μg / m3) for the study area, with higher concentrations in the center of the city. In the final Land Use Regression (LUR) model, proximity to major roads, point sources of pollution, and religious complexes, as well as high human activity land uses and population density were positively associated with air pollution, while proximity to minor roads and low human activity land uses were inversely associated with air pollution (R2= 0.535). Cross-validation of the results confirmed the reliability of the model. Kriging improved estimates by accounting for spatial correlation. Conclusions: The combination of passive NO2 sampling and LUR / Kriging modeling techniques allowed for characterization of NO2 patterns in Mysore. While traffic pollution has been indicated in the past as a major contributor to ambient air pollution levels in urbanizing centers of Asia, results indicate that other pollution factors (e.g., point sources), as well as highly localized characteristics of the urban environment (such as proximity to religious buildings), influence air pollution exposure patterns in urban India. Despite a reputation as one of the cleanest cities in India, areas of Mysore consistently experience pollution in excess of WHO health-protective guidelines for NO2.

Air pollution and mortality in the Netherlands

PDS 65: Exposure assessment: implications for epidemiology, Exhibition Hall (PDS), Ground floor, August 27, 2019, 1:30 PM - 3:00 PM Background/aim: To assess associations of long-term exposure to ambient air pollution with natural-cause mortality, air pollution exposures are generally estimated using either land-use regression (LUR) or dispersion models (DM). Within the Effects of Low-Level Air Pollution: A Study in Europe (ELAPSE) study, we evaluated associations of annual average air pollution concentrations based both on a Europe-wide hybrid LUR model and on a national DM in a Dutch National cohort. Methods: Concentrations of NO2, PM2.5 and Black Carbon (BC) based on European-wide hybrid LUR model and a national DM were linked to the home address of approximately 10.5 million Dutch adults. We used Cox proportional hazard models to investigate associations with natural-cause mortality. Following a harmonised protocol, models were adjusted for individual covariates and several area-level indicators of socio-economic status. Results: We observed 606,527 natural-cause deaths during the 5 year follow-up (2008-2012). We found strong correlations between LUR NO2 and DM NO2 (Pearson R = 0.79), correlations were slightly lower for PM2.5 (Pearson R = 0.69) and BC (Pearson R = 0.72). In fully adjusted models, we found significant associations with both LUR and DM air pollutants, however, hazard rates (HR) differed. For example, we found a HR of 1.03 per 10 µg/m3 (95% CI: 1.02, 1.04) for LUR NO2 and a HR of 1.01 per 10 µg/m3 (95% CI: 1.01, 1.02) for DM NO2. In two pollutants models of LUR NO2 and PM2.5, associations remained significant for NO2 only, while in two pollutants models of DM NO2 and PM2.5, associations remained significant for PM2.5 only. Conclusion: We found significant associations of annual average air pollution with natural-cause mortality based both on a Europe-wide hybrid LUR model and on a national DM. Different air pollution exposure assessment methods may lead to somewhat different effect estimates.

Opportunities for mobile-phone based air pollution exposure assessment

TPS 651: Air pollution exposure modeling 1, Exhibition Hall, Ground floor, August 27, 2019, 3:00 PM - 4:30 PM Background: Ubiquitous sensing technology provides novel opportunities for integrating activity patterns in exposure assessment. Theoretically promising to generate population-wide activity-based exposures, in reality strong limitations to widespread and large-scale use exist. Methods and results from three contrasting studies are presented. Methods: (1) Mobile phone activity data from 8 million users were obtained from Telecom Italia in 7 cities (aggregate data: presence of users every 15 minutes for 2 months, spatial grid varying from 0.1 to 30km). (2) During one week over a period of 1 year in Barcelona, Spain, 180 commuters were fitted with a smart phone equipped with a geo-tracking and physical activity assessment app. (3) During one week repeated in 3 different seasons, 122 participants from 3 European cities were monitored for BC exposure, geoposition and physical activity. In the first two studies, land use regression air pollution maps were overlaid with the activity data to assess exposures. Activity-weighted exposures were compared to traditional home-based assignments. Results: The scale of spatial aggregation mattered more than accounting for activity patterns in the first study. In the second study, correlations between home-based and activity-based exposures were moderate to high (0.76 Pearson for PM2.5, 0.84 for NO2); travel microenvironments contributed up to 45% of pollution intake, but only 9% of time. Lower contributions from travel (18% intake, 7% time) were found in the 3rd study. Discussion Phone usage data provided by telecom companies has the advantage of enabling representations of movements for broad populations, highlighting exposure hotspots. Aggregated data bypasses issues of privacy which may hinder widespread use of individual phone-based data collection. The level of detail obtained from individual smart phone apps highlights the importance of accounting for activities and intakes, confirmed by much more costly and burdensome air pollution monitoring studies. Methods to bridge approaches need to be investigated.

Incorporation of air pollution dispersion principals into regression models - how far can we take it?

TPS 652: Air pollution exposure modeling 2, Exhibition Hall, Ground floor, August 28, 2019, 3:00 PM - 4:30 PM Background: Regression modelling is the most commonly used method for personalised air pollution exposure assessment. These models often incorporate a large number of land-use variables for which the causal relationship to air pollution is not clear. Moreover, it has been challenging to apply them to exposure assessment at a high temporal resolution. Formulating a regression model which takes into account well-established dispersion principals may improve previous capabilities. Methods: we designed a non-linear regression model which is based on traditional Gaussian dispersion. This model has 9 regression parameters which were optimised to provide the best fit between modelled and observed nitrogen oxides concentrations at the 39 regulatory air quality monitoring locations in our study area. Results: Using a single representative wind direction for the entire study area, the new model performed better than Inverse-Distance-Weighing (IDW) and a simpler regression model (TI-ODM). In particular, the cross-validated Mean Spatial Pearson Correlation (MSPC) was 0.23 over a dataset of 3500 unique half-hourly time-points, versus 0.19 for TI-ODM and 0.14 of IDW. The MSPC is particularly interesting when considering the epidemiological applications of exposure models. It was striking that all models performed much worse in predicting total nitrogen oxides than the nitrogen dioxide predictions shown in previous studies over the same area. A heterogeneous wind field was incorporated through a puff-like regression model. This model was applied both dynamically (i.e. having memory terms that propagate from each half-hourly time-point to the next) and statically (each time-point modelled separately as a “clean slate”). The static version performed much better (Nash-Sutcliffe efficiency=0.3 in the static version vs. -537.9 in the dynamic version). However, the static version could be run only in cross-validation mode due to computational limitations. Conclusions: In future work, this model must be qualitatively validated. If successful, it can be applied to epidemiological studies.

Interactions between environmental pollutants and genetic susceptibility in asthma risk.

Exposure to air pollution is associated with enhanced risk of developing asthma, notably in the presence of genetic risk factors. Interaction analyses have shown that both outdoor and indoor air pollution interact with genetic variability to increase the incidence of asthma. In this review, we summarize recent progress in candidate gene-based studies, as well as genome-wide gene-air pollution interaction studies. Advances in epigenetics have provided evidence for DNA methylation as a mediator in gene-air pollution interactions. Emerging strategies for study design and statistical analyses may improve power in future studies. Improved air pollution exposure assessment methods and asthma endo-typing can also be expected to increase the ability to detect biologically driven gene-air pollution interaction effects.

Bibliometric analysis of cardiometabolic disorders studies involving NO2, PM2.5 and noise exposure

BackgroundThis study uses bibliometric analysis to describe the state of research about the association of NO2, PM2.5 and noise exposures – three traffic-related pollutants – with cardiometabolic disorders.MethodsWe retrieved references published 1994–2017 from Scopus and classified references with respect to exposure, health outcome and study design using index keywords. Temporal trend, top cited references, used index keywords and the number of hypothesis testing and non-hypothesis testing study design for each group were identified.ResultsResults show PM2.5 is the most frequently studied exposure (47%), followed by both NO2 and PM2.5 exposure (29%). Only 3% of references considered multiple exposures between NO2 and/or PM2.5 and noise, and these were published after 2008. While we observed a growing trend in studies with NO2 and/or PM2.5 and noise and diabetes in the last decade, there is a diminishing trend in studies with noise and diabetes. Different patterns of study designs were found through H/NH ratio, the number of references classified as having a hypothesis (H)-testing design relative to the number of references classified as having a non-hypothesis (NH)-testing design. Studies with NO2 and/or PM2.5 exposure are more likely to have a H-testing design, while those with noise exposure are more likely to have a NH-testing design, such as cross-sectional study design.ConclusionsWe conclude with three themes about research trends. First, the study of simultaneous exposures to multiple pollutants is a current trend, and likely to continue. Second, the association between traffic-related pollutants and diabetes and metabolic symptoms is an area for growth in research. Third, the transition to the use of H-testing study designs to explore associations between noise and cardiometabolic outcomes may be supported by improved understanding of the mechanism of action, and/or improvements to the accuracy and precision of air pollution and noise exposure assessments for environmental health research.

Ambient air pollution exposure and lung function assessment of filling station attendants in Ibadan, Nigeria.

Background There is an increasing range of adverse health effects associated with air pollution at very low concentrations. Few studies have assessed respiratory parameters among filling station attendants. Objectives This study assessed air pollutants; particulate matter (PM10) and total volatile organic compounds (TVOC) concentrations at filling stations as well as determined forced expiratory volume in one second (FEV1) and peak expiratory flow rate (PEFR) levels among filling station attendants. Methods A cross-sectional study was conducted to assess PM10 and TVOC concentrations at 20 systematically selected filling stations in Ibadan North Local Government Area, Ibadan for 2 months using a Thermo Scientific pDR 1500 PM10 monitor and SF2000-TVOC meter. FEV1 and PEFR levels were measured in order to assess the effect of exposure to PM10 and TVOC on lung function of 100 filling station attendants using a PIKO-1 Electronic peakflow/FEV1 meter. Results Total mean PM10 concentrations (μg/m3) in the morning (43.7±16.5) and afternoon (27.8±7.9) were significantly lower (p<0.01) than the World Health Organization (WHO) guideline limit (50 μg/m3). Total mean TVOC concentrations (ppm) in the morning (12.0±3.4) and afternoon (5.6±2.4) were however significantly higher (p<0.01) than the Occupational Safety and Health Administration (OSHA) guideline limit (3 ppm). Mean FEV1 for filling station attendants was 1.63±0.39 and PEFR was 171.7±45.9. Conclusion Filling stations are hotspots for the emission of VOCs and PM10. However, filling station attendants in this study are at risk of exposure to high concentrations of VOCs but not PM10. FEV1 and PEFR values among filling station attendants were very low which could possibly be attributed to extended exposure to air pollutants. Regular medical examinations should also be conducted on filling station attendants in order to aid early detection of deviations in their health status.

Air pollution modeling and exposure assessment during pregnancy in the French Longitudinal Study of Children (ELFE)

We developed a nation-wide exposure model to NO2, PM10 and PM2.5 at a fine spatial and temporal resolution for France in order to study air pollutants exposure during pregnancy for the French Longitudinal Study of Children (ELFE).The exposure to air pollutants was estimated daily for years 2010 and 2011 by combining three simulation models at the national and regional scale (CHIMERE) and at the local urban scale (ADMS-Urban or SIRANE). The spatial resolution was 4 km for the national scale model, 3–4 km for regional models and from 10 to 200 m for urban-scale models. We developed a confidence index (from 0 to 10) based on the target plot to identify the best model to estimate exposure for a given address, year and pollutant. Air pollution exposure during pregnancy was then estimated using each modeling scale for the 17,427 women participating in the ELFE cohort. We described the exposure of the women during different time windows of pregnancy using each of the three models and using the most suitable model as estimated by the confidence index.The exposure estimates obtained from the three models were quite similar and highly correlated (spearman correlation between 0.64 and 0.96), especially for the national and regional models. For NO2 and PM10 predicted by the urban models, the minimum values were lower and the maximum values and the variability were higher, compared to the regional and national models. The averaged confidence indexes were comprised between 5.6 and 8 depending on the pollutant, year and exposure model considered. The best confidence index was observed for urban modeling (10) and the lowest for the regional modeling (0). In average during pregnancy, using the most suitable model, women were exposed to 21 μg/m3 for NO2, 16 μg/m3 for PM2.5 and 24 μg/m3 for PM10.To our knowledge, this is the first study combining three modeling tools available at different scales to estimate NO2, PM10 and PM2.5 concentrations at a fine spatial and temporal resolution over a large geographical area. The confidence index provides guidance in the choice of the exposure model. These exposure estimates will be used to investigate potential effects of air pollutants on the pregnant woman health and on health of the fetus and development of the child.

Unraveling environmental justice in ambient PM2.5 exposure in Beijing: A big data approach

Air pollution imposes significant environmental and health risks worldwide and is expected to deteriorate in the coming decade as cities expand. Measuring population exposure to air pollution is crucial to quantifying risks to public health. In this work, we introduce a big data analytics framework to model residents' stay and commuters' travel exposure to outdoor PM2.5 and evaluate their environmental justice, with Beijing as an example. Using mobile phone and census data, we first infer travel demand of the population to derive residents' stay activities in each analysis zone, and then focus on commuters and estimate their travel routes with a traffic assignment model. Based on air quality observations from monitoring stations and a spatial interpolation model, we estimate the outdoor PM2.5 concentrations at a 500-m grid level and map them to road networks. We then estimate the travel exposure for each road segment by multiplying the PM2.5 concentration and travel time spent on the road. By combining the estimated PM2.5 exposure and housing price harnessed from online housing transaction platforms, we discover that in the winter, Beijing commuters with low wealth level are exposed to 13% more PM2.5 per hour than those with high wealth level when staying at home, but exposed to less PM2.5 by 5% when commuting the same distance (due to lighter traffic congestion in suburban areas). We also find that the residents from the southern suburbs of Beijing have both lower level of wealth and higher stay- and travel- exposure to PM2.5, especially in the winter. These findings inform more equitable environmental mitigation policies for future sustainable development in Beijing. Finally, or the first time in the literature, we compare the results of exposure estimated from passive data with subjective measures of perceived air quality (PAQ) from a survey. The PAQ data was collected via a mobile-app. The comparison confirms consistencies in results and the advantages of the big data for air pollution exposure assessments.

Optimal Deployment of a Heterogeneous Air Quality Sensor Network

Accurate assessment of air pollution exposure is crucial to better public health. Routine monitoring is done by standardized Air Quality Monitoring (AQM) stations, which are spread thinly due to size and cost. Recent technological developments have made Wireless Distributed Environmental Sensor Networks (WDESNs) that consist of low-cost Micro Sensing Units (MSUs) feasible. These MSUs can be spread more densely and provide higher spatial resolution data. The availability of MSUs, however, poses the challenge of selecting optimal sensors’ locations. Previous attempts assumed prior knowledge on pollution levels in the region of interest, and considered MSUs which measured only one pollutant. This paper presents a scheme for finding an optimal deployment of heterogeneous WDESN, which is based only on MSUs characteristics and land use analysis. To this end, a set of optional deployment locations (OLs) is defined. Each OL is characterized by a set of utilities of placing the various MSUs in that location. The optimization process seeks for the set of locations, under budget and resources constraints, that maximizes the overall utility. Using the suggested method leads to an intelligent deployment under a set of given premises. This is demonstrated vs. a real-world deployment scenario, with multiple types of MSUs.

Assessment of indoor air pollution exposure in urban hospital microenvironments

Hospitals are microenvironments containing populations with potentially enhanced sensitivity to air pollution. The objectives of this study were to characterize the concentration of indoor and outdoor size-fractionated particulate matter (PM) at two urban hospital sites in Kashan, Iran, and to evaluate the relationship between indoor and outdoor PM levels. PM1.0, PM2.5, and PM10 concentrations were measured over a 3-month period outside each hospital with parallel sampling at four indoor locations in patient wards. The results indicated that mean indoor concentrations at the sampling sites (PM1.0 = 17.8 μg/m3, PM2.5 = 45.5 μg/m3, and PM10 = 162.7 μg/m3) were found to be lower than outdoors levels (PM1.0 = 20.6 μg/m3, PM2.5 = 62.1 μg/m3, and PM10 = 300.6 μg/m3). Outdoor and indoor PM mass concentrations were associated with PM1.0, PM2.5, and PM10.0. Ambient wind speed also influenced the indoor/outdoor relationship for PM1.0 and PM2.5 but not for PM10. The average I/O ratios for PM2.5 in the intensive care unit (ICU) and children’s ward at Shahid Beheshti Hospital were close to or above 1.00. Indoor PM1.0 and PM2.5 concentrations were found to be positively associated with outdoor PM1.0 and PM2.5 concentrations, but no relationship was observed with PM10. The present findings may inform policymakers in implementing evidence-based efforts for the aim of improving the indoor air quality in closed and confined spaces.

Development and performance evaluation of new AirGIS – A GIS based air pollution and human exposure modelling system

AirGIS, a Geographic Information Systems (GIS) based air pollution and human exposure modelling system, is routinely used in conjunction with the Operational Street Pollution Model (OSPM®), across the globe, to assess local- or street-scale air pollution. We developed a substantially revised version of AirGIS (hereafter, new AirGIS) as a new modelling system in open-source GIS i.e. PostgreSQL software with its spatial extension PostGIS to (1) optimize the model performance enabling model calculations for a large number of sites over a large geographical area, with limited computing resources (2) replace the outdated programming language Avenue (3) become independent of commercial GIS software. This paper, therefore, aims to describe the overall structure of new AirGIS modelling system together with its strengths and limitations. Furthermore, the new AirGIS has been evaluated against various measured datasets of ambient air pollution (NOx, NO2, PM10 and PM2.5). In terms of reproducing temporal variation (single location, time series of concentrations e.g. annual, daily etc.) of air pollution, the new model achieved correlations (R) in the range 0.45–0.96. While, in terms of reproducing the spatial variation (several locations, single time interval), the new AirGIS achieved correlations in the range 0.32–0.92. The new model, therefore, can be used for both short- and long-term air pollution exposure assessments to facilitate health related studies. However, the present evaluation of the new modelling system also revealed that the new AirGIS significantly overestimated the observed concentrations for two out of four datasets. The possible reasons for these errors and future directions to reducing the bias in the new model output have been discussed.

Assessment of Household Air Pollution Exposure and Associated Health Effects from Burning Firewood and Yak Dung on the Eastern Tibetan Plateau

Background: Cooking and heating with solid fuels produces high levels of household air pollution (HAP) and is a substantial contributor to the disease burden. This study assesses the sources, magnitudes, and chemical compositions of HAP in traditional Eastern Tibetan Plateau households; examines the perturbation of urine metabolome in response to the HAP exposure; and explores how these correlate with cookstove use, fuel type, cooking behaviors, and residents&amp;#8217; overall health.Methods: We measured 24-hour personal exposure to PM2.5 (n &amp;#61; 46) and kitchen area black carbon (BC) concentrations (n &amp;#61; 8), using MicroPEM and microAeth, respectively. We also measured particle-bound polycyclic aromatic hydrocarbon (PAH) (n &amp;#61; 12) and metal concentrations (n &amp;#61; 11) during post-analysis of the MicroPEM filter. We conducted household surveys (n &amp;#61; 24) regarding demographics, cookstove usage, fuel type, cooking behavior, and lifestyles. In addition, we collected fasting urine (n&amp;#61; 46) and applied untargeted metabolite profiling to monitor metabolic features and assess associations with measured HAP exposures. These data will be analyzed with bioinformatics approaches-Manhattan plot and Mummichog pathway analysis.Results: Our results reaffirm that burning firewood and yak dung, result in high PM2.5 and BC exposures. The geometric mean (95% CI) concentration was 92.2 (12.5, 682.4) &amp;#956;g/m3 for PM2.5 and the mean &amp;#177; standard deviation (SD) concentration was 5.8 &amp;#177; 5.5 &amp;#956;g/m3 for BC, 474.3 &amp;#177; 584.3 ng/m3 for EPA standard suite of 16 PAHs, respectively. The mean &amp;#177; SD of concentrations of twenty-five detected metal elements ranged from 0.9 &amp;#177; 1.1 (Co) to 1585.1 &amp;#177; 1621.5 ng/m3 (Si). Our statistical analyses illustrate high concentrations of PM2.5, BC, most PAHs and metals are significantly associated with village characteristics and household fuel types.Conclusion: Results from this study provide useful information for understanding the HAP exposure and related health risks in Tibetan communities.

Assessment of Particle Oxidative Potential as an Air Pollution Exposure Metric: A Systematic Review

Ambient particulate air pollution is linked to numerous acute and chronic health outcomes. The standard approach to measuring particulate exposure has been to estimate particle mass concentration. New measures account for oxidative potential, the ability of particles to cause oxidative stress in the body. There has been no systematic comparison of particulate air pollution exposure assessment methods between those that incorporate oxidative potential and those that do not. The aim of this review is to determine if associations between particulate air pollution exposure and health outcomes are stronger when oxidative potential is used as the exposure metric, in comparison to particle mass concentration. The databases Medline and Embase were searched from inception to 28 March 2018 for studies that reported an association between particulate air pollution exposure and a health outcome in a human population, and in which exposure was measured by both particle mass concentration and oxidative potential. Study quality was assessed using a US National Toxicology Program instrument. We identified 18 publications meeting the selection criteria. The most common oxidative potential assay was antioxidant depletion in synthetic respiratory tract lining fluid, used in 12 articles. The rate of consumption of dithiothreitol was also frequently used as an oxidative potential assay and appeared in 5 studies. The dithiothreitol assay most consistently produced stronger effect estimates, as 10/14 endpoints showed higher point estimates of associations compared to particle mass concentration estimates. Other assays showed mixed results. To date there is no consistent evidence that oxidative potential is more strongly associated with health outcomes than particle mass concentrations; however, these assays do have the advantage of not treating all particles as equally harmful and further work should explore which specific assays are most relevant to evaluating air pollution health risks.

Indirect Personal Household Air Pollution Exposure Assessment for Children under 1 Year of Age Using Bluetooth Low Energy (BLE) Beacon System

Background:Household air pollution (HAP) generated from solid fuel combustion is associated with a large health burden in children under 1 year. However, assessing personal exposure to HAP for chil...

Benefits of Personal Modeling for Air Pollution Exposure Assessment

Benefits of Personal Modeling for Air Pollution Exposure Assessment

Population dynamics based on mobile phone data to improve air pollution exposure assessments.

Air pollution is one of the greatest challenges cities are facing today and improving air quality is a pressing need to reduce negative health impacts. In order to efficiently evaluate which are the most appropriate policies to reduce the impact of urban pollution sources (such as road traffic), it is essential to conduct rigorous population exposure assessments. One of the main limitations associated with those studies is the lack of information about population distribution in the city along the day (population dynamics). The pervasive use of mobile devices in our daily lives opens new opportunities to gather large amounts of anonymized and passively collected geolocation data allowing the analysis of population activity and mobility patterns. This study presents a novel methodology to estimate population dynamics from mobile phone data based on a user-centric mobility model approach. The methodology was tested in the city of Madrid (Spain) to evaluate population exposure to NO2. A comparison with traditional census-based methods shows relevant discrepancies at disaggregated levels and highlights the need to incorporate mobility patterns into population exposure assessments.