An ensemble-based model of PM2.5 concentration across the contiguous United States with high spatiotemporal resolution

Di, Qian; Amini, Heresh; Shi, Liuhua; Kloog, Itai; Silvern, Rachel F.; Kelly, James T.; Sabath, Matthew Benjamin; Choirat, Christine; Koutrakis, Petros; Lyapustin, Alexei; Wang, Yujie; Mickley, Loretta J.; Schwartz, Joel

doi:10.1016/j.envint.2019.104909

Cited by 469 publications

(427 citation statements)

References 69 publications

Supporting

Mentioning

372

Contrasting

Unclassified

Order By: Relevance

“…It is very difficult to compare the performance of so many different methods used in previous studies with the one proposed here. In summary, machine-learning methods seemed to outperform regression-based approaches, and ensemble designs only marginally improved model fit compared with individual base learners [48,49]. In this regard, we expect that the random forest methodology proposed here is the preferable option and a strength point of our study.…”

Section: Comparison With Previous Studiesmentioning

confidence: 75%

“…This is a highly valued characteristic in situations where the joint relationship between daily particulate matter and multiple spatial and spatiotemporal predictors is only marginally understood. In the last years, outputs from dispersion models have been added to the list of potential predictors, and "ensemble" approaches have been proposed, under the assumption that the average of multiple base learners would benefit from the relative advantages of each one of them [48,49].…”

Section: Comparison With Previous Studiesmentioning

confidence: 99%

See 1 more Smart Citation

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

et al. 2020

Self Cite

View full text Add to dashboard Cite

Air pollution is one of the leading causes of mortality worldwide. An accurate assessment of its spatial and temporal distribution is mandatory to conduct epidemiological studies able to estimate long-term (e.g., annual) and short-term (e.g., daily) health effects. While spatiotemporal models for particulate matter (PM) have been developed in several countries, estimates of daily nitrogen dioxide (NO2) and ozone (O3) concentrations at high spatial resolution are lacking, and no such models have been developed in Sweden. We collected data on daily air pollutant concentrations from routine monitoring networks over the period 2005–2016 and matched them with satellite data, dispersion models, meteorological parameters, and land-use variables. We developed a machine-learning approach, the random forest (RF), to estimate daily concentrations of PM10 (PM<10 microns), PM2.5 (PM<2.5 microns), PM2.5–10 (PM between 2.5 and 10 microns), NO2, and O3 for each squared kilometer of Sweden over the period 2005–2016. Our models were able to describe between 64% (PM10) and 78% (O3) of air pollutant variability in held-out observations, and between 37% (NO2) and 61% (O3) in held-out monitors, with no major differences across years and seasons and better performance in larger cities such as Stockholm. These estimates will allow to investigate air pollution effects across the whole of Sweden, including suburban and rural areas, previously neglected by epidemiological investigations.

show abstract

Section: Comparison With Previous Studiesmentioning

confidence: 75%

Section: Comparison With Previous Studiesmentioning

confidence: 99%

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…• First, we repeat all the analyses using alternative methods to estimate exposure to PM 2.5 . 26 • Second, because our study relies on observational data, our results could be sensitive to modeling choices (e.g., distributional assumptions or assumptions of linearity). We evaluate sensitivity to such choices by conducting analyses: Results for the sensitivity analyses are shown in Supplementary Materials.…”

Section: Sensitivity Analysismentioning

confidence: 99%

Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study

Nethery

Sabath

et al. 2020

Preprint

942

904

View full text Add to dashboard Cite

Background: United States government scientists estimate that COVID-19 may kill between 100,000 and 240,000 Americans. The majority of the pre-existing conditions that increase the risk of death for COVID-19 are the same diseases that are affected by long-term exposure to air pollution. We investigate whether long-term average exposure to fine particulate matter (PM 2.5 ) increases the risk of COVID-19 deaths in the United States.Methods: Data was collected for approximately 3,000 counties in the United States (98% of the population) up to April 04, 2020. We fit zero-inflated negative binomial mixed models using county level COVID-19 deaths as the outcome and county level long-term average of PM 2.5 as the exposure. We adjust by population size, hospital beds, number of individuals tested, weather, and socioeconomic and behavioral variables including, but not limited to obesity and smoking. We include a random intercept by state to account for potential correlation in counties within the same state. Results:We found that an increase of only 1 ߤ g/m 3 in PM 2.5 is associated with a 15% increase in the COVID-19 death rate, 95% confidence interval (CI) (5%, 25%). Results are statistically significant and robust to secondary and sensitivity analyses. Conclusions:A small increase in long-term exposure to PM 2.5 leads to a large increase in COVID-19 death rate, with the magnitude of increase 20 times that observed for PM 2.5 and allcause mortality. The study results underscore the importance of continuing to enforce existing air pollution regulations to protect human health both during and after the COVID-19 crisis. The data and code are publicly available.

show abstract

“…Ensemble modelling is another common statistical approach [64][65][66][67], that incorporates predictions from multiple base learners, which allows combining their predictive power and creating a final model that outperforms each base learner. A recent model developed for the contiguous US used generalized additive model that accounted for geographic difference to combine PM 2.5 estimates from neural network, random forest, and gradient boosting.…”

Section: Remote Sensingmentioning

confidence: 99%

Novel Approaches to Air Pollution Exposure and Clinical Outcomes Assessment in Environmental Health Studies

et al. 2020

Self Cite

View full text Add to dashboard Cite

An accurate assessment of pollutants’ exposure and precise evaluation of the clinical outcomes pose two major challenges to the contemporary environmental health research. The common methods for exposure assessment are based on residential addresses and are prone to many biases. Pollution levels are defined based on monitoring stations that are sparsely distributed and frequently distanced far from residential addresses. In addition, the degree of an association between outdoor and indoor air pollution levels is not fully elucidated, making the exposure assessment all the more inaccurate. Clinical outcomes’ assessment, on the other hand, mostly relies on the access to medical records from hospital admissions and outpatients’ visits in clinics. This method differentiates by health care seeking behavior and is therefore, problematic in evaluation of an onset, duration, and severity of an outcome. In the current paper, we review a number of novel solutions aimed to mitigate the aforementioned biases. First, a hybrid satellite-based modeling approach provides daily continuous spatiotemporal estimations with improved spatial resolution of 1 × 1 km2 and 200 × 200 m2 grid, and thus allows a more accurate exposure assessment. Utilizing low-cost air pollution sensors allowing a direct measurement of indoor air pollution levels can further validate these models. Furthermore, the real temporal-spatial activity can be assessed by GPS tracking devices within the individuals’ smartphones. A widespread use of smart devices can help with obtaining objective measurements of some of the clinical outcomes such as vital signs and glucose levels. Finally, human biomonitoring can be efficiently done at a population level, providing accurate estimates of in-vivo absorbed pollutants and allowing for the evaluation of body responses, by biomarkers examination. We suggest that the adoption of these novel methods will change the research paradigm heavily relying on ecological methodology and support development of the new clinical practices preventing adverse environmental effects on human health.

show abstract

An ensemble-based model of PM2.5 concentration across the contiguous United States with high spatiotemporal resolution

Cited by 469 publications

References 69 publications

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study

Novel Approaches to Air Pollution Exposure and Clinical Outcomes Assessment in Environmental Health Studies

Contact Info

Product

Resources

About