Global, national, and urban burdens of paediatric asthma incidence attributable to ambient NO2 pollution: estimates from global datasets – The Lancet
Health and Medical

Global, national, and urban burdens of paediatric asthma incidence attributable to ambient NO2 pollution: estimates from global datasets – The Lancet

Summary

Background

Paediatric asthma incidence is associated with exposure to traffic-related air pollution (TRAP), but the TRAP-attributable burden remains poorly quantified. Nitrogen dioxide (NO2) is a major component and common proxy of TRAP. In this study, we estimated the annual global number of new paediatric asthma cases attributable to NO2 exposure at a resolution sufficient to resolve intra-urban exposure gradients.

Methods

We obtained 2015 country-specific and age-group-specific asthma incidence rates from the Institute for Health Metrics and Evaluation for 194 countries and 2015 population counts at a spatial resolution of 250 × 250 m from the Global Human Settlement population grid. We used 2010–12 annual average surface NO2 concentrations derived from land-use regression at a resolution of 100 × 100 m, and we derived concentration-response functions from relative risk estimates reported in a multinational meta-analysis. We then estimated the NO2-attributable burden of asthma incidence in children aged 1–18 years in 194 countries and 125 major cities at a resolution of 250 × 250 m.

Findings

Globally, we estimated that 4·0 million (95% uncertainty interval [UI] 1·8–5·2) new paediatric asthma cases could be attributable to NO2 pollution annually; 64% of these occur in urban centres. This burden accounts for 13% (6–16) of global incidence. Regionally, the greatest burdens of new asthma cases associated with NO2 exposure per 100 000 children were estimated for Andean Latin America (340 cases per year, 95% UI 150–440), high-income North America (310, 140–400), and high-income Asia Pacific (300, 140–370). Within cities, the greatest burdens of new asthma cases associated with NO2 exposure per 100 000 children were estimated for Lima, Peru (690 cases per year, 95% UI 330–870); Shanghai, China (650, 340–770); and Bogota, Colombia (580, 270–730). Among 125 major cities, the percentage of new asthma cases attributable to NO2 pollution ranged from 5·6% (95% UI 2·4–7·4) in Orlu, Nigeria, to 48% (25–57) in Shanghai, China. This contribution exceeded 20% of new asthma cases in 92 cities. We estimated that about 92% of paediatric asthma incidence attributable to NO2 exposure occurred in areas with annual average NO2 concentrations lower than the WHO guideline of 21 parts per billion.

Interpretation

Efforts to reduce NO2 exposure could help prevent a substantial portion of new paediatric asthma cases in both developed and developing countries, and especially in urban areas. Traffic emissions should be a target for exposure-mitigation strategies. The adequacy of the WHO guideline for ambient NO2 concentrations might need to be revisited.

Funding

George Washington University.

Introduction

Global paediatric asthma prevalence has increased strikingly since the 1950s, and asthma is now the most commonly reported non-communicable disease among children worldwide.

Over the past decade, several epidemiological studies done in North America, Latin America, Europe, and east Asia have reported associations between traffic-related air pollution (TRAP) exposure and new-onset asthma in children, whereas such associations are less clear in adults.

  • Gasana J
  • Dillikar D
  • Mendy A
  • Forno E
  • Ramos Vieira E

Motor vehicle air pollution and asthma in children: a meta-analysis.

  • Anderson HR
  • Favarato G
  • Atkinson RW

Long-term exposure to air pollution and the incidence of asthma: meta-analysis of cohort studies.

  • Bowatte G
  • Lodge C
  • Lowe AJ
  • et al.

The influence of childhood traffic-related air pollution exposure on asthma, allergy and sensitization: a systematic review and a meta-analysis of birth cohort studies.

  • Khreis H
  • Kelly C
  • Tate J
  • Parslow R
  • Lucas K
  • Nieuwenhuijsen M

Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

  • Anenberg SC
  • Henze DK
  • Tinney V
  • et al.

Estimates of the global burden of ambient PM2·5, ozone, and NO

2

on asthma incidence and emergency room visits.

Corroborating evidence from toxicological and gene-environment studies have suggested that TRAP causes oxidative injury to the airways, leading to inflammation and remodelling that, in a genetically predisposed individual, could result in asthma development.

  • Guarnieri M
  • Balmes JR

Outdoor air pollution and asthma.

  • MacIntyre EA
  • Brauer M
  • Melen E
  • et al.

GSTP1 and TNF gene variants and associations between air pollution and incident childhood asthma: the Traffic, Asthma and Genetics (TAG) study.

Although the putative agent (or agents) within the TRAP mixture has yet to be identified,

  • Khreis H
  • Nieuwenhuijsen MJ

Traffic-related air pollution and childhood asthma: recent advances and remaining gaps in the exposure assessment methods.

epidemiological studies have most often relied on nitrogen dioxide (NO

2

) as a proxy, because NO

2

measurements are readily available in many countries, and the variability of TRAP mixture appears to be well characterised by NO

2

.

  • Beckerman B
  • Jerrett M
  • Brook JR
  • Verma DK
  • Arain MA
  • Finkelstein MM

Correlation of nitrogen dioxide with other traffic pollutants near a major expressway.

  • Levy I
  • Mihele C
  • Lu G
  • Narayan J
  • Brook JR

Evaluating multipollutant exposure and urban air quality: pollutant interrelationships, neighborhood variability, and nitrogen dioxide as a proxy pollutant.

Results from four meta-analyses of TRAP exposure and paediatric asthma incidence all indicated consistent associations with NO

2

, but results were mixed for associations with fine particulate matter (PM

2·5

;

appendix

).

  • Gasana J
  • Dillikar D
  • Mendy A
  • Forno E
  • Ramos Vieira E

Motor vehicle air pollution and asthma in children: a meta-analysis.

  • Anderson HR
  • Favarato G
  • Atkinson RW

Long-term exposure to air pollution and the incidence of asthma: meta-analysis of cohort studies.

  • Bowatte G
  • Lodge C
  • Lowe AJ
  • et al.

The influence of childhood traffic-related air pollution exposure on asthma, allergy and sensitization: a systematic review and a meta-analysis of birth cohort studies.

  • Khreis H
  • Kelly C
  • Tate J
  • Parslow R
  • Lucas K
  • Nieuwenhuijsen M

Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

Reviews

    Health Canada

    Human health risk assessment for ambient nitrogen dioxide.

      US Environmental Protection Agency

      Integrated science assessment for oxides of nitrogen—health criteria.

      done by the US Environmental Protection Agency and Health Canada, published in 2016, also concluded that the overall evidence indicates that a causal relationship probably exists between long-term NO

      2

      exposure and paediatric asthma development. NO

      2

      is a secondary pollutant formed mainly from fossil fuel combustion; traffic emissions can contribute up to 80% of ambient NO

      2

      in cities.

      • Levy I
      • Mihele C
      • Lu G
      • Narayan J
      • Brook JR

      Evaluating multipollutant exposure and urban air quality: pollutant interrelationships, neighborhood variability, and nitrogen dioxide as a proxy pollutant.

      • Sundvor I
      • Castell Balaguer N
      • Viana M
      • et al.

      Road traffic’s contribution to air quality in European cities—ETC/ACM technical paper 2012/14.

      Research in context

      Evidence before this study

      We searched MEDLINE, PubMed, and Google Scholar to identify meta-analyses of epidemiological studies on traffic-related air pollution (TRAP) exposure and paediatric asthma incidence, using a combination of the search terms “traffic-related air pollution”, “asthma incidence” or “asthma development”, “meta-analysis”, and “child*”. We limited the search to studies that pooled results from multiple countries, reported a numerical relative risk (RR) accompanied by an estimate of precision, were written in English, and were published up to March 1, 2018. Nitrogen dioxide (NO2) is a major component and commonly used proxy of the complex TRAP mixture. The most up-to-date and comprehensive multinational meta-analysis reported an RR of 1·26 (95% uncertainty interval 1·10–1·37) per 10 parts per billion (ppb) of TRAP-related NO2. The Global Burden of Disease studies have not included air pollution as a risk factor for asthma incidence or prevalence, and the only previous study to have estimated the global burden of paediatric asthma incidence attributable to ambient air pollution relied on exposure datasets with resolutions too coarse to resolve the intra-urban NO2 variability driven primarily by TRAP. A more recent study done at a resolution of 100 × 100 m for a major city in the UK estimated that 18–24% of all annual paediatric asthma incidence might be attributable to ambient NO2, suggesting that the global burden of asthma associated with NO2 exposure might be substantial.

      Added value of this study

      To our knowledge, this study provides the first global estimate of the annual burden of paediatric asthma incidence attributable to ambient NO2 pollution at a resolution sufficient to resolve intra-urban and near-roadway exposure gradients (250 × 250 m). We provide estimates for 194 countries and 125 major cities.

      Implications of all the available evidence

      Our findings suggest that a substantial portion of paediatric asthma incidence could be avoided by reducing NO2 pollution in both developed and developing countries, especially in urban areas. We estimate that about 97% of children lived and about 92% of new asthma cases attributable to NO2 occurred in areas with annual average NO2 concentrations lower than the WHO guideline of 21 ppb, suggesting that the adequacy of this guideline might need to be revisited.

      Despite the accumulating epidemiological evidence, the burden of paediatric asthma incidence attributable to TRAP exposure remains poorly quantified and has not been included in previous global burden of disease studies.

      • Cohen AJ
      • Brauer M
      • Burnett R
      • et al.

      Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015.

      • Prüss-Üstün A
      • Wolf J
      • Corvalán CF
      • Bos R
      • Neira MP

      Preventing disease through healthy environments: a global assessment of the burden of disease from environmental risks.

      In a study published in 2018,

      • Anenberg SC
      • Henze DK
      • Tinney V
      • et al.

      Estimates of the global burden of ambient PM2·5, ozone, and NO

      2

      on asthma incidence and emergency room visits.

      Anenberg and colleagues estimated that global asthma incidence attributable to anthropogenic NO

      2

      and PM

      2·5

      might cause about 1·7 million disability-adjusted life-years (DALYs, ie, years of healthy life lost) among children in 2015, with about 97% of this burden due to PM

      2·5

      . However, this assessment relied on satellite-derived NO

      2

      concentrations with a spatial resolution of approximately 10 × 10 km, which likely underestimated the effects of NO

      2

      for several reasons. First, the mid-afternoon satellite overpass time captures NO

      2

      concentration at a minimum in its diurnal cycle. Second, given that NO

      2

      concentrations decline to urban background levels at a distance greater than 300–500 m from major roadways and that intra-urban variation might be more strongly associated with exposure and asthma development than urban–rural variation, the spatial resolution of the satellite data was too coarse.

      • Anderson HR
      • Favarato G
      • Atkinson RW

      Long-term exposure to air pollution and the incidence of asthma: meta-analysis of cohort studies.

        Health Effects Institute

        Traffic-related air pollution: a critical review of the literature on emissions, exposure, and health effects.

        An assessment

        • Khreis H
        • de Hoogh K
        • Nieuwenhuijsen MJ

        Full-chain health impact assessment of traffic-related air pollution and childhood asthma.

        done with a spatial resolution of 100 × 100 m in a major city in the UK estimated that 18–24% of all annual paediatric asthma incidence might be attributable to ambient NO

        2

        . Because TRAP emissions are ubiquitous, the global burden is likely to be substantial, especially within densely populated urban areas.

        In this study, we combined global datasets of NO

        2

        concentrations, population, and asthma incidence with relative risks drawn from a multinational epidemiological meta-analysis

        • Khreis H
        • Kelly C
        • Tate J
        • Parslow R
        • Lucas K
        • Nieuwenhuijsen M

        Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

        to estimate the annual number of new paediatric asthma cases attributable to ambient NO

        2

        at a spatial resolution of 250 × 250 m. We focused on NO

        2

        because it is a widely used proxy of TRAP and because the evidence for association with paediatric asthma incidence is most robust for TRAP-derived NO

        2

        .

        Methods

         Population and demographics

        We used the 2015 Global Human Settlement population grid (GHS-POP) from the European Commission’s Joint Research Centre, which provides population counts for 1975, 1990, 2000, and 2015 at a spatial resolution of 250 × 250 m, the finest resolution available with global coverage.

          European Commission Joint Research CentreColumbia University Center for International Earth Science Information Network

          GHS population grid, derived from GPW4, multitemporal (1975, 1990, 2000, 2015).

          GHS-POP disaggregates residential population estimates from the Center for International Earth Science Information Network’s Gridded Population of the World, version 4 (GPWv4), by use of the distribution and density of built-up areas derived from satellite imaging. To estimate the paediatric population, we calculated age group fractions (1–4, 5–9, 10–14, and 15–18 years) by use of the gridded GPWv4 and the Basic Demographic Characteristics (version 4.10) for 2010, the latest year available.

            Center for International Earth Science Information NetworkColumbia University

            Gridded Population of the World, version 4 (GPWv4): basic demographic characteristics, revision 10.

            These outputs were then regridded from their original 30 arc-second resolution to 250 × 250 m and applied to the GHS-POP population estimates.

            Figure 1A

            shows national estimates of paediatric population size.

            Figure thumbnail gr1

            Figure 1National estimates of population aged 1–18 years (A), population-weighted annual asthma incidence rates (B), and population-weighted annual average nitrogen dioxide (NO2) concentrations (C)

            (B) Population-weighted averages were calculated from the country-specific and age-group-specific rates. (C) Concentrations are weighted by the size of the population aged 1–18 years of each country. ppb=parts per billion.

             Baseline asthma incidence

            We obtained 2015 country-specific and age-group-specific asthma incidence rates from the Institute for Health Metrics and Evaluation (IHME) for 194 countries.

              GBD 2015 Chronic Respiratory Disease Collaborators

              Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015.

              We found that most national asthma incidence rates did not change by more than 5% between 2010–12 and 2015 and, therefore, we chose 2015 to match the GHS-POP dataset. The definition of asthma was a reported diagnosis by a physician, with wheezing in the preceding 12 months.

              Figure 1B

              shows national asthma incidence rates standardised across the population aged 1–18 years. Additional details on the IHME incidence rates and methods for approximating DALYs are in the

              appendix

              .

               Concentration-response function

              Anenberg and colleagues

              • Anenberg SC
              • Henze DK
              • Tinney V
              • et al.

              Estimates of the global burden of ambient PM2·5, ozone, and NO

              2

              on asthma incidence and emergency room visits.

              did a literature review to identify meta-analyses of epidemiological studies on air pollution and asthma, four of which reported paediatric asthma incidence attributable to TRAP-related NO

              2

              (

              appendix

              ). These meta-analyses pooled results from individual studies in different countries to derive relative risk (RR) estimates that could be generally representative of the global population. We used the same search criteria as Anenberg and colleagues

              • Anenberg SC
              • Henze DK
              • Tinney V
              • et al.

              Estimates of the global burden of ambient PM2·5, ozone, and NO

              2

              on asthma incidence and emergency room visits.

              (

              appendix

              ) and found no additional meta-analyses published more recently. In this study, we derived concentration response functions (CRFs) from RRs reported by Khreis and colleagues,

              • Khreis H
              • Kelly C
              • Tate J
              • Parslow R
              • Lucas K
              • Nieuwenhuijsen M

              Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

              one of the meta-analyses identified in the search. By comparison with the other three meta-analyses, Khreis and colleagues reviewed the largest number of studies, which included all but one of those considered in the other meta-analyses: 20 studies from Europe, 11 from North America, five from Japan, three from China, one from South Korea, and one from Taiwan, totalling 41 studies; although there was variation in the RRs reported in these studies, no regional heterogeneity was apparent. The meta-analysis

              • Khreis H
              • Kelly C
              • Tate J
              • Parslow R
              • Lucas K
              • Nieuwenhuijsen M

              Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

              reported an overall random-effects RR of 1·26 per 10 parts per billion (ppb) TRAP-related NO

              2

              (95% uncertainty interval [UI] 1·10–1·37). This UI encompassed the central RR estimates reported in the other three meta-analyses (

              appendix

              ).

              Available evidence is insufficient to establish whether a low-concentration threshold exists, below which there is no risk of paediatric asthma development, and whether the association is linear or non-linear. In this study, we assumed a log-linear relationship between NO

              2

              concentration and outcome, as Khreis and colleagues did.

              • Khreis H
              • Kelly C
              • Tate J
              • Parslow R
              • Lucas K
              • Nieuwenhuijsen M

              Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

              We applied a counterfactual concentration of 2 ppb, the 5th percentile of the minimum exposure concentrations reported in the individual studies considered by Khreis and colleagues.

              • Khreis H
              • Kelly C
              • Tate J
              • Parslow R
              • Lucas K
              • Nieuwenhuijsen M

              Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

              This approach assumes that long-term exposure to NO

              2

              less than 2 ppb confers no excess risk of incident asthma in children. As a sensitivity test, we also assumed counterfactual concentrations of 0 ppb and 5 ppb, the 25th percentile of the minimum exposure concentrations reported (

              appendix

              ).

               Ambient NO2 concentrations

              We used a dataset of surface NO

              2

              concentrations developed by Larkin and colleagues,

              • Larkin A
              • Geddes JA
              • Martin R V
              • et al.

              Global land use regression model for nitrogen dioxide air pollution.

              which is, to date, the only global dataset able to resolve near-roadway concentrations at a spatial resolution of 100 × 100 m, for 2010–12 annual average concentrations. Larkin and colleagues used land-use regression modelling, an empirical method that combines ground-based air pollution measurements (from 5220 air monitors in 58 countries) with satellite-derived pollutant estimates and other geospatial variables to predict surface concentrations. The strongest predictors of these were major roads within 100 m and satellite-derived NO

              2

              concentrations. The land-use regression model captured 54% of global NO

              2

              variation with a mean absolute error of 3·7 ppb. On a continental level, this model varied between capturing 31% (Africa) to 63% (South America) of regional variation, with mean absolute errors ranging from 2·3 ppb (Africa) to 4·4 ppb (North America). The model also generated within-city spatial patterns and concentrations similar to other land-use regression models created for specific cities. Concentrations ranged between 0 ppb and 72 ppb (we estimated that about 0·5% of the global population lived in areas with 0 ppb). Urban–rural and intra-urban concentration gradients for three major cities are shown in the

              appendix

              . By comparison, the surface NO

              2

              concentration dataset used by Anenberg and colleagues

              • Anenberg SC
              • Henze DK
              • Tinney V
              • et al.

              Estimates of the global burden of ambient PM2·5, ozone, and NO

              2

              on asthma incidence and emergency room visits.

              had a spatial resolution of 0·1° × 0·1° (roughly 10 × 10 km), with values ranging between 0 ppb and 17·5 ppb. This dataset was derived from column retrievals from the Ozone Monitoring Instrument on board the US National Aeronautics and Space Administration’s Aura satellite.

              • Lamsal LN
              • Martin R V
              • van Donkelaar A
              • et al.

              Ground-level nitrogen dioxide concentrations inferred from the satellite-borne Ozone Monitoring Instrument.

              We regridded and reprojected the output from Larkin and colleagues

              • Larkin A
              • Geddes JA
              • Martin R V
              • et al.

              Global land use regression model for nitrogen dioxide air pollution.

              to match the resolution and projection of the population dataset at 250 × 250 m, by use of the Geospatial Data Abstraction Library. Although some very fine-scale features were lost at this coarser scale, intra-city gradients such as those occurring near major roadways could still be captured at 250 m (

              appendix

              ).

              Figure 1C

              shows national population-weighted NO

              2

              concentrations.

               Health impact function

              Following the example of Anenberg and colleagues,

              • Anenberg SC
              • Henze DK
              • Tinney V
              • et al.

              Estimates of the global burden of ambient PM2·5, ozone, and NO

              2

              on asthma incidence and emergency room visits.

              we applied methods typically used to assess premature mortality attributable to air pollution to estimate the number of new paediatric asthma cases attributable to NO

              2

              pollution annually,

              • Cohen AJ
              • Brauer M
              • Burnett R
              • et al.

              Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015.

              • Anenberg SC
              • Horowitz LW
              • Tong DQ
              • West JJ

              An estimate of the global burden of anthropogenic ozone and fine particulate matter on premature human mortality using atmospheric modeling.

              by use of NO

              2

              data (2010–12 annual average) and population data (2015) that were as recent and as close in time to each other as possible, and spatially resolved below 250 m. The health impact function is:

              Math Eq

              where

              Inc

              is the asthma incidence for country

              c

              and age group

              a

              (1–4, 5–9, 10–14, and 15–18 years),

              Pop

              is the gridded population for age group

              a

              , β is the concentration-response factor (from the meta-analysis of epidemiological studies), and

              X

              is the gridded annual average NO

              2

              concentration. We applied

              Math Eq

              in each 250 × 250 m grid cell (

              i, j

              ) globally before summing by national or urban extents. We then multiplied these aggregated sums by the relevant national asthma incidence rates, and provided results for 21 regions, 194 countries, and 125 major cities.

               Region, country, and city aggregation

              We grouped countries into 21 regions or seven super-regions, according to IHME specifications based on epidemiological similarity and geographical closeness (

              appendix

              ). Results were aggregated by country by use of the GPWv4 National Identifier Grid.

                Center for International Earth Science Information NetworkColumbia University

                Gridded Population of the World, version 4 (GPWv4): national identifier grid, revision 10.

                Defining city boundaries is challenging because no universal definition exists.

                • Uchiyama Y
                • Mori K

                Methods for specifying spatial boundaries of cities in the world: the impacts of delineation methods on city sustainability indices.

                We used a globally consistent definition of urbanisation from the Global Human Settlement 1 × 1 km grid (GHS-SMOD), for 2015. Urban centres were defined as any contiguous cells with 50 000 or more people and a population density of 1500 or more inhabitants per km

                2

                or a density of built-up higher than 50% per grid cell.

                • Pesaresi M
                • Freire S

                GHS settlement grid following the REGIO model 2014 in application to GHSL Landsat and CIESIN GPW v4-multitemporal (1975–1990–2000–2015).

                By combining GHS-SMOD with GHS-POP, we selected the 125 most populous urban centres (population size >3·5 million people). Given that these urban centres do not necessarily match administrative boundaries and sometimes group together multiple neighbouring cities (

                appendix

                ), we reported city-level results in units of cases per 100 000 children and percentage of total incidence.

                 Uncertainties in input variables of the health impact function

                Although each input variable is uncertain, our results are accompanied by UIs derived from those of the RR, given that this error is well constrained. We also did a sensitivity analysis to compare the influence of uncertainties in the RR, NO

                2

                concentrations, and asthma incidence rates (uncertainties in the GHS-POP population dataset were not available;

                appendix

                ).

                 Role of the funding source

                The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. PA and SCA had full access to all the data in the study and had final responsibility for the decision to submit for publication.

                Results

                Globally, we estimated that ambient NO

                2

                might be associated with 4·0 million (95% UI 1·8–5·2) new paediatric asthma cases annually, accounting for 13% (95% UI 5·8–16) of global incidence (

                table

                ). The mean burden of new cases associated with ambient NO

                2

                translates to approximately 730 000 DALYs per year (

                appendix

                ). We estimated that about 97% of children lived and about 92% of NO

                2

                -attributable paediatric asthma incidence occurred in areas with annual average NO

                2

                concentrations lower than the WHO guideline of 21 ppb.

                TableAnnual number and percentage of new asthma cases attributable to ambient nitrogen dioxide (NO2) exposure for children aged 1–18 years

                95% uncertainty intervals (UI) reflect uncertainties in the relative risk estimates of childhood asthma incidence attributable to traffic-related NO

                2

                pollution. Countries are grouped into regions according to the Institute for Health Metrics and Evaluation specification (

                appendix

                ). Numbers are rounded to two significant figures.

                Regionally, the largest burden of new asthma cases associated with NO

                2

                exposure per 100 000 children per year was estimated for Andean Latin America, followed by high-income North America and high-income Asia Pacific (

                table

                ;

                appendix

                ). The largest percentage of regional paediatric asthma incidence attributable to NO

                2

                pollution was estimated for high-income Asia Pacific, followed by east Asia and high-income North America, southern Latin America, and western Europe (

                table

                ). Eastern sub-Saharan Africa was estimated to have the lowest incidence of new asthma cases attributable to NO

                2

                exposure per 100 000 children, whereas Oceania had the lowest percentage of incidence attributable to NO

                2

                (

                table

                ).

                The largest national burdens of new asthma cases attributable to NO

                2

                exposure were estimated for China (760 000 new cases per year, 95% UI 350 000–970 000), India (350 000, 160 000–450 000), the USA (240 000, 110 000–300 000), Indonesia (160 000, 72 000–210 000), and Brazil (140 000, 63 000–180 000;

                figure 2A

                ,

                appendix

                ). Of all countries, China had the second largest paediatric population and third highest population-weighted NO

                2

                concentrations (

                figure 1

                ). India, despite having the lowest paediatric asthma incidence rate, had the largest paediatric population. Although the USA, Indonesia, and Brazil had similar paediatric population sizes, the USA had the highest population-weighted NO

                2

                concentrations and Indonesia had the highest asthma incidence rate among these three countries. The largest national burden per 100 000 children per year was estimated for Kuwait (550 cases per year, 95% UI 260–690), followed by the United Arab Emirates (460, 220–580), Canada (450, 210–580), Taiwan (420, 190–540), and Qatar (410, 200–520;

                figure 2B

                ) due to high asthma incidence rates, high population-weighted urban NO

                2

                concentrations, or both. Of 194 countries assessed, Canada had the 34th highest paediatric asthma incidence rate—the highest among high-income nations. Regarding the percentage of asthma incidence attributable to NO

                2

                exposure, we estimated values exceeding 15% for 48 countries (

                figure 2C

                ). The highest percentages of asthma incidence attributable to NO

                2

                were estimated for South Korea (31%, 95% UI 15–38), United Arab Emirates (30%, 14–38), Kuwait (30%, 14–38), and Qatar (30%, 14–37).

                Figure thumbnail gr2

                Figure 2National annual estimates of number of new asthma cases (A), number of new asthma cases per 100 000 children (B), and percentage of new asthma cases attributable to ambient nitrogen dioxide (NO2) exposure for children aged 1–18 years

                The values shown are derived using the central mean relative risk estimate and assuming a counterfactual NO2 concentration of 2 parts per billion.

                We estimated that 64% of paediatric asthma incidence attributable to NO

                2

                occurred in urban centres globally. This percentage increased to 90% if surrounding suburban areas were also considered (GHS-SMOD’s so-called low-density clusters). On a national level, the percentage of paediatric asthma incidence attributable to NO

                2

                occurring in urban centres exceeded 50% in 108 countries (

                figure 3

                ,

                appendix

                ), reflecting the distribution of children living in urban areas and the prevalence of high NO

                2

                concentrations found in urban centres. Generally, national asthma incidence rates were highest in the youngest age group (1–4 years) and decreased with increasing age. This effect dwarfed the age group variations in population size, resulting in the largest asthma burdens attributable to NO

                2

                being estimated for children aged 1–4 years, both in total and per 100 000 children, in all but one country (Moldova;

                appendix

                ). Regarding the influence of uncertainties in the input variables, we found that asthma incidence rates had the least influence, given that their UIs were small. UIs resulting from those of RRs and from NO

                2

                concentrations were generally within the same order of magnitude (

                appendix

                ).

                Figure thumbnail gr3

                Figure 3National annual estimates of percentage of nitrogen dioxide (NO2)-attributable paediatric asthma incidence occurring in urban centres (A), percentage of children living in urban centres (B), and ratio of population-weighted NO2 concentration in urban centres relative to non-urban areas (C)

                Urban centres are defined as any contiguous cells with 50 000 or more people and a population density of 1500 or more inhabitants per km2 or a density of built-up greater than 50% per grid cell, following the Global Human Settlement classification.

                Of the 125 major cities we investigated, the largest burdens of paediatric asthma incidence attributable to NO

                2

                exposure per 100 000 children were estimated for Lima, Peru (690 cases per year, 95% UI 330–870); Shanghai, China (650, 340–770); Bogota, Colombia (580, 270–730); Beijing, China (560, 280–680); and Toronto, Canada (550, 250–700;

                figure 4

                ,

                appendix

                ). The large burdens estimated for Chinese cities were primarily driven by high population-weighted NO

                2

                concentrations (28–33 ppb). By contrast, Lima, Bogota, and Toronto had medium NO

                2

                concentrations (14–17 ppb), but had among the highest national paediatric asthma incidence rates, especially for children aged 1–4 years, in the 54 countries where the 125 cities are located. The lowest population-standardised burdens were found in Orlu, Nigeria (83 cases per year, 95% UI 36–110), and in four cities in India (Varanasi, 88 [40–110]; Kanpur, 92 [42–120]; Lucknow, 99 [45–130]; and Kolkata, 100 [47–130];

                figure 4

                ,

                appendix

                ). India had the lowest asthma incidence rate among all countries, whereas Orlu had the lowest population-weighted NO

                2

                concentration (5 ppb). The percentage of new asthma cases attributable to NO

                2

                pollution by city ranged from 5·6% (95% UI 2·4–7·4) in Orlu, Nigeria, to 48% (25–57) in Shanghai, China, and primarily reflected the variations in NO

                2

                exposures within each region (

                figure 5

                ). These percentages exceeded 20% in 92 cities and exceeded 30% in 32 cities (

                figure 5

                ,

                appendix

                ). The ten largest contributions were estimated for eight cities in China (Shanghai, Tianjin, Beijing, Shenyang, Xi’an, Taiyuan, Zhengzhou, and Harbin), for Moscow (Russia), and for Seoul (South Korea;

                figure 5

                ,

                appendix

                ). All these cities had high urban NO

                2

                concentrations. The meta-analysis from which the RRs were drawn included three epidemiological studies done in China.

                Figure thumbnail gr4

                Figure 4Annual number of new asthma cases per 100 000 children attributable to ambient nitrogen dioxide (NO2) exposure for children aged 1–18 years

                Number of new paediatric asthma cases attributable to NO

                2

                exposure in 125 major cities, grouped into seven super-regions: high income (1); Latin America and Caribbean (2); sub-Saharan Africa (3); north Africa and Middle East (4); south Asia (5); southeast Asia, east Asia, and Oceania (6); and central Europe, eastern Europe, and central Asia (7). Within a super-region, cities were ranked in decreasing order of paediatric population-weighted NO

                2

                concentrations and coloured by country. The values shown were derived with use of the central mean relative risk estimate and assuming a counterfactual NO

                2

                concentration of 2 parts per billion. In some cases, an urban centre identified in the Global Human Settlement 1 × 1 km grid spans multiple cities (see

                for full urban cluster names).

                Figure thumbnail gr5

                Figure 5Annual percentage of new asthma cases attributable to ambient nitrogen dioxide (NO2) exposure for children aged 1–18 years

                Percentage of new paediatric asthma cases attributable to NO

                2

                exposure in 125 major cities, grouped into seven super-regions: high income (1); Latin America and Caribbean (2); sub-Saharan Africa (3); north Africa and Middle East (4); south Asia (5); southeast Asia, east Asia, and Oceania (6); and central Europe, eastern Europe, and central Asia (7). Within a super-region, cities were ranked in decreasing order of paediatric population-weighted NO

                2

                concentrations and coloured by country. The values shown were derived with use of the central mean relative risk estimate and assuming a counterfactual NO

                2

                concentration of 2 parts per billion. In some cases, an urban centre identified in the Global Human Settlement 1 × 1 km grid spans multiple cities (see

                for full urban cluster names).

                Discussion

                This study provides the first global estimate of the burden of paediatric asthma incidence attributable to ambient NO

                2

                at a spatial resolution fine enough to resolve intra-urban and near-roadway exposure gradients, for 194 countries and 125 major cities. We estimated that, each year, 4·0 million (95% UI 1·8–5·2) new cases of paediatric asthma might be attributable to NO

                2

                pollution, accounting for 13% (5·8–16) of global incidence; 48 of 194 countries had percentages greater than 15% of national incidence. The city-level percentages ranged between 6% and 48%, with 92 of 125 cities exceeding 20%. The 2015 Global Burden of Disease study

                  GBD 2015 Chronic Respiratory Disease Collaborators

                  Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015.

                  estimated that smoking and occupational asthmagens could account for 16·5% of the global asthma prevalence for all ages in 2015. Although the burden of premature mortality attributable to ambient PM

                  2·5

                  pollution is generally greater in low-income and middle-income countries because of high exposures,

                  • Cohen AJ
                  • Brauer M
                  • Burnett R
                  • et al.

                  Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015.

                  the effects of NO

                  2

                  pollution on paediatric asthma incidence appear to be less differentiated by level of development. Many high-income countries have high NO

                  2

                  exposures, especially those in North America, western Europe, and Asia Pacific. In Canada, this problem is also exacerbated by high paediatric asthma incidence rates (

                  appendix

                  ).

                  Because transportation can contribute up to 80% of ambient NO

                  2

                  in cities and is an important driver of increasing greenhouse gas emissions globally,

                  • Sundvor I
                  • Castell Balaguer N
                  • Viana M
                  • et al.

                  Road traffic’s contribution to air quality in European cities—ETC/ACM technical paper 2012/14.

                    IPCC

                    Drivers, trends and mitigation.

                    we also explored whether there were associations between carbon dioxide (CO

                    2

                    ) emissions and NO

                    2

                    exposure, to highlight potential co-benefits of exposure-mitigation policies. We found that, in general, countries and cities with higher CO

                    2

                    emissions tended to have higher paediatric NO

                    2

                    exposures and larger burdens of asthma incidence attributable to ambient NO

                    2

                    (

                    appendix

                    ).

                    Anenberg and colleagues

                    • Anenberg SC
                    • Henze DK
                    • Tinney V
                    • et al.

                    Estimates of the global burden of ambient PM2·5, ozone, and NO

                    2

                    on asthma incidence and emergency room visits.

                    have estimated the global burden of asthma incidence attributable to anthropogenic NO

                    2

                    for children aged 0–17 years, relying on satellite-derived NO

                    2

                    concentrations, from the Ozone Monitoring Instrument, at a resolution of roughly 10 × 10 km. The authors suggested that they likely underestimated actual exposures because the satellite’s mid-afternoon overpass time coincides with the minimum concentration in diurnal NO

                    2

                    variation and because near-roadway concentrations were not resolved. Indeed, with use of the same RR drawn from Khreis and colleagues,

                    • Khreis H
                    • Kelly C
                    • Tate J
                    • Parslow R
                    • Lucas K
                    • Nieuwenhuijsen M

                    Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

                    but with different population data, their estimated global burden of 1·3 million new asthma cases per year attributable to NO

                    2

                    is about 3·1 times smaller than our estimate assuming a counterfactual concentration of 2 ppb (and about 1·7 times smaller than our estimate assuming a counterfactual concentration of 5 ppb). In another study,

                    • Khreis H
                    • de Hoogh K
                    • Nieuwenhuijsen MJ

                    Full-chain health impact assessment of traffic-related air pollution and childhood asthma.

                    published in 2018, Khreis and colleagues quantified the annual paediatric asthma incidence attributable to air pollution in Bradford, UK, by use of concentration estimates at 100 × 100 m resolution and assuming no low concentration threshold. By use of ambient NO

                    2

                    concentrations modelled by land-use regression, Khreis and colleagues estimated that 24% of annual paediatric asthma cases might be attributable to NO

                    2

                    exposure in 2009 (with an annual mean NO

                    2

                    concentration of 12 ppb). This matches our equivalent estimate (assuming a counterfactual of 0 ppb) of 28% for the Leeds–Bradford metropolitan area, which had annual mean NO

                    2

                    concentrations of 13 ppb during 2010–12. Estimates assuming a counterfactual concentration of 2 ppb were 23%, whereas those assuming 5 ppb were 17% (data not shown).

                    Several important limitations and uncertainties exist in our approach. First, the epidemiological evidence considered by Khreis and colleagues

                    • Khreis H
                    • Kelly C
                    • Tate J
                    • Parslow R
                    • Lucas K
                    • Nieuwenhuijsen M

                    Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

                    and the ground-based NO

                    2

                    monitoring data used in the global land-use regression model for NO

                    2

                    developed by Larkin and colleagues

                    • Larkin A
                    • Geddes JA
                    • Martin R V
                    • et al.

                    Global land use regression model for nitrogen dioxide air pollution.

                    are mainly available only for North America, Europe, and east Asia. These limitations in the global representation of the outcome measure and of the exposure characterisation upon which we relied consequently apply to our estimates. Additionally, because most ground-based NO

                    2

                    monitors are in urban areas, the global land-use regression model might overestimate concentrations in rural areas. However, this is unlikely to add appreciable bias because rural populations are sparse. Available epidemiological studies have also been done within a range (about 1–50 ppb) that approximately spans global exposures (0–72 ppb).

                    A second limitation is that NO

                    2

                    concentrations used in this study represent the 2010–12 annual average, whereas the population and asthma incidence rates are 2015 estimates. Duncan and colleagues

                    • Duncan BN
                    • Lamsal LN
                    • Thompson AM
                    • et al.

                    A space-based, high-resolution view of notable changes in urban NOx pollution around the world (2005–2014).

                    found that between 2005 and 2014, there have been substantial trends in urban NO

                    2

                    concentrations worldwide, with general decreases in US and European cities and increases in Asia, although complex spatial heterogeneity exists. Future assessments could apply these trends to the NO

                    2

                    land-use regression dataset to estimate effects in other years.

                    Third, we relied on national paediatric asthma incidence rates in our analysis, but subnational variations most likely exist, especially between urban and rural populations. Additionally, although the IHME incidence rates were derived from prevalence data covering 121 countries and all 21 regions,

                      GBD 2015 Chronic Respiratory Disease Collaborators

                      Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015.

                      the definition of asthma based on physician diagnosis might underestimate the effects of vehicular emissions on asthma in low-income and middle-income countries because of low health-care access or the neglect of subclinical symptoms that might not manifest in a diagnosis. Fourth, uncertainties remain in the shape of the CRF of paediatric asthma incidence and NO

                      2

                      exposure. In this study, we assumed a log-linear relationship and a low-concentration threshold of 2 ppb. We also assumed a causal relationship between NO

                      2

                      exposure and asthma incidence and used a CRF derived from single-pollutant epidemiological models (because multipollutant models were not available). The evidence so far suggests that associations with asthma incidence are more robust for NO

                      2

                      than for other air pollutants and that NO

                      2

                      is better than PM

                      2·5

                      for representing near-roadway variability of the TRAP mixture.

                      • Karner AA
                      • Eisinger DS
                      • Niemeier DA

                      Near-roadway air quality: synthesizing the findings from real-world data.

                      • Cooper MJ
                      • Martin R V
                      • van Donkelaar A
                      • Lamsal L
                      • Brauer M
                      • Brook JR

                      A satellite-based multi-pollutant index of global air quality.

                      However, it is possible that NO

                      2

                      is a proxy for other putative agents in the TRAP mixture, that the reported associations are sensitive to control for co-pollutants, or both. All these uncertainties mean that our NO

                      2

                      -attributable asthma estimates might be overestimated or underestimated.

                      Our analysis and characterisation of uncertainties can consequently be improved by future studies aiming to resolve the limitations described. The availability of global NO

                      2

                      concentration and population datasets that are up to date and have even finer temporal and spatial resolutions than those of the datasets used in this study will be particularly valuable. In this study, we did not isolate traffic-specific contributions to ambient NO

                      2

                      concentrations, but this approach is more consistent with existing epidemiological studies, the majority of which also relied on NO

                      2

                      exposures derived from land-use regression without isolating the contribution of traffic emissions.

                      • Khreis H
                      • Kelly C
                      • Tate J
                      • Parslow R
                      • Lucas K
                      • Nieuwenhuijsen M

                      Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

                      Isolating the role of TRAP can help to constrain estimates of the portion of total asthma incidence that could be avoided by targeting this emission source. Additionally, it would be useful for future epidemiological studies to identify the putative agents within the TRAP mixture and target understudied populations, especially in Africa and central and southeast Asia.

                      Despite substantial decreases in NO

                      2

                      concentrations over the past decade in large areas of the USA and western Europe,

                      • Duncan BN
                      • Lamsal LN
                      • Thompson AM
                      • et al.

                      A space-based, high-resolution view of notable changes in urban NOx pollution around the world (2005–2014).

                      our findings suggest that existing levels of ambient NO

                      2

                      pollution are a substantial risk factor for paediatric asthma incidence in both developed and developing countries, especially in urban areas. In fact, our findings suggest that the adequacy of the WHO guideline for annual average NO

                      2

                      concentrations might need to be revisited. We also found that countries and cities with higher CO

                      2

                      emissions from fossil fuel combustion tend to have higher NO

                      2

                      exposures, providing further support that alignment of policy initiatives to mitigate air pollution and climate change can have multiple public health benefits.

                      • Anenberg SC
                      • Miller J
                      • Minjares R
                      • et al.

                      Impacts and mitigation of excess diesel-related NOx emissions in 11 major vehicle markets.

                      • West JJ
                      • Smith SJ
                      • Silva RA
                      • et al.

                      Co-benefits of mitigating global greenhouse gas emissions for future air quality and human health.

                      Traffic emissions should be a target for exposure-mitigation strategies.

                      Contributors

                      SCA and MB conceived the project idea. SCA and PA designed the analysis methods. PH and MB contributed to the development of the land-use regression NO2 dataset used in this analysis. PA did the analysis and had primary responsibility for writing the manuscript, to which all authors contributed.

                      Declaration of interests

                      SCA provided research consulting services on separate projects for the International Council on Clean Transportation and the Health Effects Institute within the 36 months during and before the work reported in this article. PA, MB, and PH declare no competing interests.

                      Acknowledgments

                      Funding was provided by the George Washington University. We thank all the data providers for making their data publicly available. Baseline disease rates were provided by the Institute of Health Metrics and Evaluation. Global Human Settlement (GHS) population and urban extents were provided by the Joint Research Centre of the European Commission. Open-source Data Inventory for Anthropogenic CO2 (ODIAC) emissions dataset was provided by Tomohiro Oda of Colorado State University (Fort Collins, CO, USA) and Global Monitoring Division, US National Oceanic and Atmospheric Administration’s Earth System Research Laboratory (Boulder, CO, USA). The ODIAC project is supported by Greenhouse Gas Observing Satellite project, National Institute for Environmental Studies, Japan. We thank Andrew Larkin for providing the NO2 dataset and technical advice, Haneen Khreis for helpful discussions on the TRAP-asthma epidemiology, Aneta J Florczyk and Sergio Freire for providing insights into the GHS datasets, and Casey Kalman for research assistance on matching GHS-SMOD urban centre identifiers to city names using ArcGIS.

                      Supplementary Material

                      References

                        • WHO

                        Asthma fact sheet.

                        • Gasana J
                        • Dillikar D
                        • Mendy A
                        • Forno E
                        • Ramos Vieira E

                        Motor vehicle air pollution and asthma in children: a meta-analysis.

                        Env Res. 2012; 117: 36-45

                        • Anderson HR
                        • Favarato G
                        • Atkinson RW

                        Long-term exposure to air pollution and the incidence of asthma: meta-analysis of cohort studies.

                        Air Qual Atmos Heal. 2013; 6: 47-56

                        • Bowatte G
                        • Lodge C
                        • Lowe AJ
                        • et al.

                        The influence of childhood traffic-related air pollution exposure on asthma, allergy and sensitization: a systematic review and a meta-analysis of birth cohort studies.

                        Allergy. 2014; 70: 245-256

                        • Khreis H
                        • Kelly C
                        • Tate J
                        • Parslow R
                        • Lucas K
                        • Nieuwenhuijsen M

                        Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis.

                        Environ Int. 2017; 100: 1-31

                        • Anenberg SC
                        • Henze DK
                        • Tinney V
                        • et al.

                        Estimates of the global burden of ambient PM2·5, ozone, and NO2 on asthma incidence and emergency room visits.

                        Environ Health Perspect. 2018; 126: 107004

                        • Guarnieri M
                        • Balmes JR

                        Outdoor air pollution and asthma.

                        Lancet. 2014; 383: 1581-1592

                        • MacIntyre EA
                        • Brauer M
                        • Melen E
                        • et al.

                        GSTP1 and TNF gene variants and associations between air pollution and incident childhood asthma: the Traffic, Asthma and Genetics (TAG) study.

                        Environ Health Perspect. 2014; 122: 418-424

                        • Khreis H
                        • Nieuwenhuijsen MJ

                        Traffic-related air pollution and childhood asthma: recent advances and remaining gaps in the exposure assessment methods.

                        Int J Environ Res Public Health. 2017; 14: 312

                        • Beckerman B
                        • Jerrett M
                        • Brook JR
                        • Verma DK
                        • Arain MA
                        • Finkelstein MM

                        Correlation of nitrogen dioxide with other traffic pollutants near a major expressway.

                        Atmos Environ. 2008; 42: 275-290

                        • Levy I
                        • Mihele C
                        • Lu G
                        • Narayan J
                        • Brook JR

                        Evaluating multipollutant exposure and urban air quality: pollutant interrelationships, neighborhood variability, and nitrogen dioxide as a proxy pollutant.

                        Environ Health Perspect. 2014; 122: 65-72

                        • Health Canada

                        Human health risk assessment for ambient nitrogen dioxide.

                        Health Canada,
                        Ottawa; 2016

                        • US Environmental Protection Agency

                        Integrated science assessment for oxides of nitrogen—health criteria.

                        US Environmental Protection Agency,
                        Research Triangle Park; 2016

                        • Sundvor I
                        • Castell Balaguer N
                        • Viana M
                        • et al.

                        Road traffic’s contribution to air quality in European cities—ETC/ACM technical paper 2012/14.

                        European Topic Centre for Air Pollution and Climate Change Mitigation,
                        Copenhagen; 2013

                        • Cohen AJ
                        • Brauer M
                        • Burnett R
                        • et al.

                        Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015.

                        Lancet. 2017; 389: 1907-1918

                        • Prüss-Üstün A
                        • Wolf J
                        • Corvalán CF
                        • Bos R
                        • Neira MP

                        Preventing disease through healthy environments: a global assessment of the burden of disease from environmental risks.

                        World Health Organization,
                        Geneva; 2016

                        • Health Effects Institute

                        Traffic-related air pollution: a critical review of the literature on emissions, exposure, and health effects.

                        Health Effects Institute,
                        Boston; 2010

                        • Khreis H
                        • de Hoogh K
                        • Nieuwenhuijsen MJ

                        Full-chain health impact assessment of traffic-related air pollution and childhood asthma.

                        Environ Int. 2018; 114: 365-375

                        • European Commission Joint Research Centre
                        • Columbia University Center for International Earth Science Information Network

                        GHS population grid, derived from GPW4, multitemporal (1975, 1990, 2000, 2015).

                        • Center for International Earth Science Information Network
                        • Columbia University

                        Gridded Population of the World, version 4 (GPWv4): basic demographic characteristics, revision 10.

                        • GBD 2015 Chronic Respiratory Disease Collaborators

                        Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015.

                        Lancet Respir Med. 2017; 5: 691-706

                        • Larkin A
                        • Geddes JA
                        • Martin R V
                        • et al.

                        Global land use regression model for nitrogen dioxide air pollution.

                        Environ Sci Technol. 2017; 51: 6957-6964

                        • Lamsal LN
                        • Martin R V
                        • van Donkelaar A
                        • et al.

                        Ground-level nitrogen dioxide concentrations inferred from the satellite-borne Ozone Monitoring Instrument.

                        J Geophys Res Atmos. 2008; 113: D16

                        • Anenberg SC
                        • Horowitz LW
                        • Tong DQ
                        • West JJ

                        An estimate of the global burden of anthropogenic ozone and fine particulate matter on premature human mortality using atmospheric modeling.

                        Environ Health Perspect. 2010; 118: 1189-1195

                        • Center for International Earth Science Information Network
                        • Columbia University

                        Gridded Population of the World, version 4 (GPWv4): national identifier grid, revision 10.

                        • Uchiyama Y
                        • Mori K

                        Methods for specifying spatial boundaries of cities in the world: the impacts of delineation methods on city sustainability indices.

                        Sci Total Environ. 2017; 592: 345-356

                        • Pesaresi M
                        • Freire S

                        GHS settlement grid following the REGIO model 2014 in application to GHSL Landsat and CIESIN GPW v4-multitemporal (1975–1990–2000–2015).

                        • IPCC

                        Drivers, trends and mitigation.

                        in: Intergovernmental panel on climate change Climate change 2014: mitigation of climate change—working group III contribution to the IPCC fifth assessment report. Cambridge University Press,
                        Cambridge; 2015: 351-412

                        • Duncan BN
                        • Lamsal LN
                        • Thompson AM
                        • et al.

                        A space-based, high-resolution view of notable changes in urban NOx pollution around the world (2005–2014).

                        J Geophys Res Atmos. 2015; 121: 976-996

                        • Karner AA
                        • Eisinger DS
                        • Niemeier DA

                        Near-roadway air quality: synthesizing the findings from real-world data.

                        Environ Sci Technol. 2010; 44: 5334-5344

                        • Cooper MJ
                        • Martin R V
                        • van Donkelaar A
                        • Lamsal L
                        • Brauer M
                        • Brook JR

                        A satellite-based multi-pollutant index of global air quality.

                        Environ Sci Technol. 2012; 46: 8523-8524

                        • Anenberg SC
                        • Miller J
                        • Minjares R
                        • et al.

                        Impacts and mitigation of excess diesel-related NOx emissions in 11 major vehicle markets.

                        Nature. 2017; 545: 467

                        • West JJ
                        • Smith SJ
                        • Silva RA
                        • et al.

                        Co-benefits of mitigating global greenhouse gas emissions for future air quality and human health.

                        Nat Clim Chang. 2013; 3: 885

                      Article Info

                      Publication History

                      Published: April 10, 2019

                      Identification

                      DOI: https://doi.org/10.1016/S2542-5196(19)30046-4

                      Copyright

                      © 2019 The Author(s). Published by Elsevier Ltd.

                      User License

                      Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) |

                      How you can reuse Information Icon

                      Permitted

                      For non-commercial purposes:

                      • Read, print & download
                      • Redistribute or republish the final article
                      • Text & data mine
                      • Translate the article (private use only, not for distribution)
                      • Reuse portions or extracts from the article in other works

                      Not Permitted

                      • Sell or re-use for commercial purposes
                      • Distribute translations or adaptations of the article


                      Elsevier’s open access license policy

                      ScienceDirect

                      Access this article on ScienceDirect

                      Linked Articles

                      Read More