What this study adds
In this large-scale study of 336,648 children, we report that NH3 and NH4+ exposure may be risk factors for the onset of asthma before their sixth birthday. Our findings support the need for the continuous monitoring of air pollution policies, given not only the potential detrimental health effects of NH3 and NH4+ exposure but also the fact that these pollutants contribute substantially to the formation of secondary fine particulate matter that is recognized as a significant burden of disease and mortality.
Asthma is the most prevalent disease in children worldwide.1 Concern is growing regarding the impact of outdoor ambient air pollution exposure on the risk of pediatric asthma. Several studies have shown that fine particulate matter less than 2.5 μm in diameter (PM2.5) is associated with respiratory symptoms, exacerbation, and new-onset asthma in children.2,3 Previous studies have focused almost exclusively on health effects associated with transport- and industrial-related PM2.5 exposure, whereas surprisingly little attention has been paid to agricultural air pollutants and the role of particulate precursors and components such as gaseous ammonia (NH3) and ammonium (NH4+), although NH4+ accounts for a substantial proportion of the PM2.5 mass.4 More information on the physical and chemical properties of NH3 and NH4+ is given in the online data supplement; http://links.lww.com/EE/A13.
Results of the few existing epidemiological studies showed an association of NH4+ concentrations and ambient ammoniumbisulphate exposure from lags of 0–4 days and increased respiratory hospitalizations in adults5,6 and increased emergency department visits due to upper respiratory infections in children.7 Furthermore, high ambient NH3 exposure during the last day has been linked with declining lung function, but not asthma symptoms,8 whereas annual average NH3 exposure has neither been associated with increased use of asthma medication nor asthma9 in 8- to 12-year-old children. A review of controlled human exposure studies on inorganic NH4+ salts concluded that there was no evidence of acute adverse pulmonary health effects from sulfate or nitrate salts at environmentally relevant levels, but at concentrations well above ambient levels, changes in pulmonary function might be occurring.10 Similarly, high NH3 and NH4+ concentrations in occupational settings and rodent experiments have been associated with airway irritation and increased inflammatory responses.11 However, so far, no studies have investigated the effect of long-term ambient NH3 and NH4+ exposure on the risk of asthma.
The present study differs from earlier studies in that (1) It applies atmospheric model calculations, enabling an investigation of the effect of long-term ambient exposure on a nation-wide basis while accounting for both spatial and temporal concentration variations on a regional and local level; (2) The use of Danish population registers provides an excellent opportunity to link environmental exposure via residential coordinates with information on personal characteristics and diagnosis history; (3) The focus is exclusively on children, who are possibly more susceptible than adults because their lungs are developing and they generally spend more time outdoors. The main objective of our study was to investigate the effect of gaseous NH3, particulate NH4+, the total concentration of NH3 and NH4+ (NHx), and PM2.5 on incident asthma in a large population of preschool-aged Danish children. We hypothesized that exposure to high levels of NH3, NH4+, NHx, and PM2.5 during the past 3 months increase the risk of asthma.
Information base: The Danish Civil Registration System
The Danish Civil Registration System12 is a register established in 1968 of all people alive and living in Denmark, currently 5.7 million persons. The system includes information on personal identification number, sex, date and place of birth, continuously updated information on vital status and individual and longitudinal information on place of residence. The personal identification number is used in all national registers, enabling accurate linkage between registers. The source population of the present study consisted of all persons who were born in Denmark from January 1, 2000, to December 31, 2011, and whose parents were both born in Denmark (642,075 persons). The residential longitudinal database obtained from the Danish Civil Registration System was linked with the Danish official standard addresses and coordinates to obtain exact information on the longitudinal geographical residential coordinates of all Danish residents from 1978 onwards.
Modeling of NH3, NH4+, and PM2.5 concentrations
Concentrations of pollutants were modeled on a daily basis with a spatial resolution of 5.56 km × 5.56 km for all of Denmark covering the period January 1, 2005, to December 31, 2012. Primary pollutants of interest comprised NH3, NH4+, total inorganic ammonia (NHx), and PM2.5. Secondary pollutants comprised O3, NO2, and SO2. Concentrations were shown for Denmark’s five administratively and geographically disjoint regions (north Denmark, central Denmark, south Denmark, capital region, Zealand). Location and characteristics of the Danish regions are shown in Figure e1; http://links.lww.com/EE/A13 (see supplementary information). All were obtained from the Danish Eulerian Hemispheric Model (DEHM),13 which is part of the Danish Ammonia Model System (DAMOS) previously described in detail.14,15 Briefly, the coupling of DEHM and DAMOS compose a three-dimensional Eulerian atmospheric chemistry transport model (CTM) with a horizontal domain covering the Northern Hemisphere. This allows extracts of hourly average concentrations to grid cells covering Denmark based on the regional and local contribution.14
Assessment of asthma
Asthma cases within the cohort were identified from the Danish National Patient Register.16 This register contains data on all admissions to public hospitals in Denmark from 1977 onwards. All Danish residents are entitled to free national health care, and the Danish National Patient Register has virtually complete data on hospital visits. Information on outpatient visits was included from 1995 and onwards. The diagnostic system used was the International Classification of Diseases, 10th revision (ICD-10-CM). Cohort members were classified with asthma (ICD-10: J45 or J46) if they had been admitted to a public hospital or had been treated as an outpatient and had received their first diagnosis of asthma as the final diagnosis of the visit (release diagnosis). The date of onset was defined as the first day of the first visit (in- or outpatient) when the diagnosis of asthma was made. Childhood asthma was defined as children who were first diagnosed with asthma between the ages of one and six. The reason for choosing this age group is that early onset asthma is largely based on inflammation and to a minor degree allergy, and we wanted to study this in relation to air pollution since air pollution may induce inflammatory changes in the airways.
Study design and statistical analyses
We used survival analyses techniques to follow all children from their first to their sixth birthday between 2006 and 2012 for the risk of asthma, using exposure to air pollutions during the past 3 months as the exposure of interest. From the outset of the study, we hypothesized that past exposure to air pollution during the past 3 months influences the risk of asthma. The rationale behind the choice of exposure time window was based on an assumption that asthma develops over time and possibly not due to a single peak in exposure. Incidence rate ratios were estimated using an individually time-matched case–control design. For each child first diagnosed with asthma, we selected at random 25 control children who shared the same sex and birthday, who were alive, and who were not diagnosed with asthma when the case was included (n = 322,694). For cases and their individually matched controls, we calculated the average exposure during the past 3 months before the case was diagnosed, accounting for residential changes for both cases and controls. Average pollutant concentrations were first categorized into deciles using the lowest decile as the reference category. Secondary analyses treated the average pollutant concentrations as a continuous variable measuring the risk for children exposed to the highest decile compared to children exposed to the lowest decile, while utilizing information on all intermediate deciles. Incidence rate ratios were estimated using conditional logistic regression, with each case–control set forming separate strata. The following adjustment scenarios were used: (1) age, sex, date of birth, calendar year (base adjustment), (2) base adjustment + socioconomic status (SES), (3) base adjustment + region, and (4) base adjustment + region + parental SES. The latter was measured by maternal educational level and paternal income that may be risk factors for asthma in children. The rationale for including region is several: regions are administrative responsible for registration of asthma diagnosing, regions may differ in meteorological factors and unknown factors that affect the air quality overall, and region and residential area can serve as a proxy for SES. We performed sensitivity analyses, considering the past 6- and 12-month exposure time-windows, considering different age groups (1, 2–3, and 4–5 years), and considering consistency of the results in the five geographical regions by performing a log likelihood test for interaction between the effect of each pollutant and region. Statistical analysis was conducted using SAS statistical software (SAS Institute Inc., Cary, NC).
Study population characteristics
During the period 2006–2012, 12,935 persons born in Denmark were diagnosed with asthma for the first time in their life before their sixth birthday. Table e1; http://links.lww.com/EE/A13 (see supplementary information) shows characteristics of the cases and their individually matched controls (recorded at the time each case was first diagnosed) comprising 335,629 in total. By design, cases and controls were matched by age and sex. Study participants ranged in age from 1 to 6 years; the majority of cases occurred among children less than 3 years old, and 61% of the children were boys (Table e1; http://links.lww.com/EE/A13). The asthma incidence was lower in the capital region compared with the other Danish regions, with the region of south Denmark having the highest hazard. High SES was associated with a lower hazard (Table e1; http://links.lww.com/EE/A13).
Concentrations and distribution of ammonia and ammonia-related pollutants
Figure e2; http://links.lww.com/EE/A13 maps pollutant concentrations averaged over 2008 in the study area. The annual mean (SD) was 1.19 ppb (0.39) for NH4+, 1.99 pbb (1.35) for NH3, 3.19 pbb (1.40) for NHx, and 7.10 μg/m3 (1.66) for PM2.5. Concentrations of NH3 were highest in the southwestern part of Denmark, generally higher over the mainland of Denmark (Jutland) and lowest over Zealand. Concentrations of NH4+ showed a strong south-west north-east gradient ranging from 0.30 to 0.77 ppb. PM2.5 concentrations ranged from 7.20 to 10.00 μg/m3 and were slightly elevated over the largest cities, especially over the capital city of Copenhagen. While highly positive correlations were found between concentrations of NH3 and NHx (r = 0.96) and NH4+ and PM2.5 (r = 0.91), a moderate correlation was shown between NH4+ and NHx (r = 0.28), low positive correlations were found between NHx and PM2.5 (r = 0.09), whereas low negative correlations were found between NH3 and PM2.5 (r = −0.17). No correlation was found between NH3 and NH4+ (r = −0.002).
NH3, NH4+, NHx, and PM2.5 concentrations and asthma morbidity
Figure 1 depicts effects of NH3, NH4+, NHx, and PM2.5 concentrations and risk of asthma. A clear positive exposure–response association between the level of NH3 and a later risk of asthma is shown. The children exposed to the highest level of NH3 had a 1.74 (95% confidence interval = 1.60, 1.89)-fold increased risk of asthma compared to children exposed to the lowest level (Table e1; http://links.lww.com/EE/A13). NH4+ and NHx showed nearly identical patterns as shown in Figure 1 and Table 1. On the other hand, PM2.5 exposure was not associated with asthma (adjusted hazard ratio; 95% confidence interval 0.96; 0.86–1.06; Table 1). Similar results were found for NH3, NH4+, NHx, and PM2.5 when using 6- and 12-month exposure time windows (Table e2; http://links.lww.com/EE/A13) and consistently across different age groups (Table e3–e5; http://links.lww.com/EE/A13).
When we adjusted effects sizes for SES, the positive associations were slightly attenuated (Table 1), but the associations for NH3, NH4+, NHx, and PM2.5 with asthma incidence were negative when adjusting for region. Correspondingly, a similar pattern of association was shown when we performed the analyses stratified on geographical regions (Table e6; http://links.lww.com/EE/A13). Finally, we performed all analyses adjusting for base characteristics (sex, date of birth, age, and calendar year), region, and SES and additionally for any other pollutant by including one pollutant simultaneously in each logistic regression analysis in a two-pollutant model (Figure 2). The negative association for PM2.5, NH3, and NHx with asthma remained, whereas the positive association persisted for NH4+ when adjusting for PM2.5.
The present study is unique as it allowed us to potentially link concentrations of specific components of particulate air pollution with incident cases of asthma, thus, to investigate whether gaseous NH3, particulate NH4+, and PM2.5 are risk factors for asthma. The basic adjusted analyses revealed a positive association between NH3, NH4+, and the sum NHx exposure and asthma diagnosis but showed no association between PM2.5 and asthma. In a second step, the positive associations of NH3 and NH4+ exposure with asthma were slightly attenuated, when adjusting for SES, whereas the associations disappeared when adjusting for region in a third step. Finally, in a fourth step, we adjusted additionally for any other pollutant in a two-pollutant model, where the positive association remained for NH4+ when adjusting for PM2.5, which was made possible by the virtue of the study size. These findings indicate that early life exposure to NH4+ may be a risk factor for onset asthma, but we cannot rule out confounding by region or other factors covarying with NH3, NH4+, and PM2.5. If regional confounding exists, this means that the regions either differ in diagnostic practises, population characteristics, concentrations of other pollutants, or in some other unknown factor(s) related to the region that may be the underlying cause of asthma and hereby introduced ecological fallacy. Additionally, it can be argued that stratification for and adjustment for region could have caused an over-adjustment due to the evident regional distribution of pollutants causing a lack of exposure contrast within regions. Furthermore, although PM2.5 followed the regional sections, associations of PM2.5 with asthma were generally inconsistent and negative when adjusting for region, not supporting earlier findings showing a positive association for PM2.5 with childhood asthma.2,3,17 The high correlation between NH4+ and PM2.5 (r = 0.91) and uneven distribution of the pollutants may thus be the reason for the change in association between NH4+ and asthma when adjusting for region. The adjustment for PM2.5 can be viewed as an adjustment for traffic and heating of houses by wood burning in the cities, unmasking the effect of NH4+ to be observed in more urban areas, as can be seen in Figure e2; http://links.lww.com/EE/A13.
Comparison with other studies
To our knowledge, this large-scale study is the first to explore ambient NH3 and NH4+ exposure and the risk of developing pediatric asthma from a nation-wide perspective. Among the studies performed with children, results showed no association with self-reported asthma symptoms and medication use,8,9 whereas an inverse association was found between NH3 concentrations with forced expiratory volume in 1 second.8 Moreover, some studies have suggested an increased risk of respiratory hospitalizations linked with high NH4+ concentrations in adults in Atlanta (USA)5 and New York (USA).6 Similarly, one study found an association of NH4+ exposure with increased emergency department visits due to upper respiratory infections, but not due to pneumonia or bronchiolitis, in a large population of 1- to 4-year-olds in Atlanta (USA).7 Generalizations of these results are problematic as the studies explored the respiratory health effects related to NH3 and NH4+ exposures either in small restricted areas, or during short time-periods, or in small-scale study populations, which increases the risk of bias affecting the results.
The positive associations between NH3, NH4+, and asthma may be causal relations or, alternatively, it may be that NH3 and NH4+ serve as markers for other farming exposures associated with asthma—for example, organic dust exposure that has previously been linked to respiratory symptoms and asthma,18 also in combination with NH.3,19 Other studies on respiratory morbidity relied on proxies as “living on or near a farm” due to the known large agricultural contribution of local and regional emissions of NH3 and NH4+. Some studies have indicated detrimental health effects in adults in terms of declined lung function,20 increased respiratory symptoms,18 and doctor-diagnosed asthma.21 Beneficial effects of farm contact have also been reported in terms of a lower prevalence of sensitization and atopic asthma.22 Studies have shown that children exposed to farming environments suffer more from asthma and respiratory symptoms23,24 and less of allergy1 than their counterparts. However, most farming studies use “general farming exposures” as the exposure variable, and therefore these findings cannot be specifically related to either NH3 or NH4+ exposures.
Although our findings indicated that only NH4+ is a risk factor for onset asthma early in life, emissions of NH3 are essential as NH3 is a precursor to NH4+ that is formed in a reaction between NH3 and acid aerosols. The spatial pattern of NH3 concentrations showed a south-west, north-east gradient and NH4+ showed a south-west north-east gradient. A validation study on the DEHM and DAMOS models concluded that these gradients are caused partly by atmospheric transport of primarily PM2.5 from northern Europe and by local and regional ammonia emissions south of Denmark,15 thus emphasizing transboundary movement of pollutants. In a previous study on European NH3 emissions, the NH3 emissions varied considerably with the month of the year, in which Danish emissions were low to moderate in February but moderate to high in April due to manure application. However, we did not observe an increase in asthma diagnoses in the 3-month period following April (data not shown).
Strengths and limitations
A major strength of our study is the large sample size that was possible owing to the register-based design. This design secures high completeness of data, prospective data collection and eliminates selection bias. Therefore, the study is representative of preschool children of parents born in Denmark, regardless of their place of residence, SES, sex, and age, but precludes children and parents that are emigrants to avoid handling the complexicity of different cultures and ethnicities that are known to influence on asthma. Additionally, as the exposure time window was the same for cases and their individually matched controls, the study design and the adjustment for sex, date of birth, age, and calendar year preclude confounding by time-invariant characteristics of participants. In our study, some covariates were found to be risk factors of onset asthma, including children’s mothers having a short education compared to a longer education and fathers having a low income compared to a higher income. A shortcoming of our study, however, is the lack of information on other potential confounders that might have played a role in our study, especially smoking and dampness in the children’s home, two factors known to greatly influence the incidence of childhood asthma.25 Information on smoking is of particular importance because tobacco smoke, on the one hand, is a strong predictor of asthma, and on the other hand, NH3 is an endogenous part of tobacco smoke chemistry.26,27 Furthermore, it would have been informative to know whether the children lived on or near a farm and spent time in enclosed livestock facilities to compare the risk of highly exposed children with low-exposed children.
Another strength of this study was the nationwide estimation of exposure that provides information on modeled NH3 and NH4+ exposure during the past year for all participants, comprising both long-term exposures with a high grid resolution and providing satisfying exposure contrasts. Although the validation of the DAMOS model has shown satisfying agreement between modeled and measured concentrations of NH3,28 a limitation of the model appears to be related to the performance of the wet deposition of NH3, which can potentially be improved by better estimation of precipitation.14 A clear advantage of this study is that it contained enough spatial variation in pollutants to enable us to explore the change in effects when adjusting for any other pollutant. Still, the complexity of pollutant mixtures like PM2.5 makes it difficult to identify causal components and study interactions in those mixtures. Also the high correlations between pollutants often makes it difficult to disentangle their individual effects on health. This may also explain why previous studies on PM2.5 constituents and health effects have been inconsistent in their results and PM2.5 mass persists as the best predictor of adverse health outcomes.29
We used modeled pollutant concentrations linked to each child’s residential address as proxy for each child’s exposure. Besides being at home, most Danish children attend daycare centers that most often are located in the same area as their residential address. Therefore, we assumed that the children remain in the same area of their home address the majority of their time.
An alternative to exposure modeling by residential address would be the measurement of exposures at residential addresses or personal air monitoring. However, although these exposure assessments are favored, it is impractical in a large study population over time. Validation of the DAMOS model has shown satisfying agreement between modeled and measured concentrations of NH3 (correlation coefficients for five monitoring stations, range 0.51–0.78) based on the daily mean values averaged over measurement stations as time series and annual mean values, as well as the daily maximum values as scatter plots.28
In epidemiological studies, various definitions of pediatric asthma are used, and therefore outcome misclassification is often a concern, especially in children with less distinctive symptoms. It is a strength of our study that we used information on doctor-diagnosed asthma derived from the Danish National Patient Register, which limits misclassification since diagnosis is based on validated criteria. However, the register most likely only reflects cases of severe asthma and leaves out children with mild to moderate asthma (or easy treatable asthma) diagnosed from general practitioners. Most children in our study were diagnosed with asthma before 3 years of age. These children may be defined as early wheezers as in early ages, it can be unclear whether respiratory symptoms are part of asthma development or attributable to respiratory virus infections. As a result, effect estimates may therefore bias associations towards the null. Our stratified analyses in different age groups revealed, however, similar results, although the findings were most consistent in the 2- to 3 year-olds. It should be noted though that asthma is a heterogeneous disease, and the time of diagnosis and progress may reflect different asthma phenotypes as recently suggested.30 Unfortunately, we had no information on atopy, lung function, inflammatory markers, and disease progress and prognosis, and therefore, we cannot distinguish further between phenotypes.
Our results indicate that high levels of NH4+ may be associated with an increased risk of developing asthma in preschool children causing an important public health problem affecting a significant number of children. Therefore, there is a continuous need to pay attention to air pollution policies and highly exposed populations of children to protect these children from developing asthma that may affect them and society throughout their lives. Further exploration in areas with a high agricultural activity is needed to confirm our result.
Conflict of interest statement
The authors declare that they have no conflicts of interest with regard to the content of this report.
Source of funding
Supported by dNmark: Danish Nitrogen Mitigation Assessment: Research and Know-how for a sustainable, low-nitrogen food production by the Danish Council for Strategic Research, the Danish Council for Strategic Research, the Centre for Integrated Register-Based Research at Aarhus University (CIRRAU) Denmark, and NordForsk under the Nordic Programme on Health and Welfare (project 75007).
Data availability Health and Welfare, NordicWelfAir: Data are not available due to a strict data-sharing agreement. Data on individual characteristics were requested through Statistics Denmark from the Danish Civil Registration System and data on asthma from the Danish National Patient Register. Environmental data can be requested for payment from the Department of Environmental Sciences, Aarhus University. The overall link of information on residential coordinates with exposure assessment, personal characteristics, and diagnosis history was performed within Statistics Denmark that requires all users to have permission to access data and carry out analyses.
1. Eder W, Ege MJ, von ME. The asthma
epidemic.N Engl J Med200621222635
2. Gehring U, Wijga AH, Brauer M, et al. Traffic-related air pollution
and the development of asthma
and allergies during the first 8 years of life.Am J Respir Crit Care Med2010181596603
3. Gasana J, Dillikar D, Mendy A, et al. Motor vehicle air pollution
in children: a meta-analysis.Environ Res20121173645
4. Bressi M, Sciare J, Ghersi V, et al. A one-year comprehensive chemical characterisation of fine aerosol (PM 2.5) at urban, suburban and rural background sites in the region of Paris (France).Atmos Chem Phys20131578257844
5. Sarnat JA, Marmur A, Klein M, et al. Fine particle sources and cardiorespiratory morbidity : an application of chemical mass balance and factor analytical source-apportionment methods.Environ Health Perspect2008116459466
6. Jones RR, Hogrefe C, Fitzgerald EF, et al. Respiratory hospitalizations in association with fine PM and its components in New York State Respiratory hospitalizations in association with fine PM and its components in New York State.J Air Waste Manage Assoc201565559569
7. Darrow LA, Klein M, Dana Flanders W, et al. Air pollution
and acute respiratory infections among children 0–4 years of age: An 18-year time-series study.Am J Epidemiol2014180968977
8. Loftus C, Yost M, Sampson P, et al. Ambient ammonia
exposures in an agricultural community and pediatric asthma
9. Dockery DW, Cunningham J, Damokosh AI, et al. Health effects of acid aerosols on North American children: respiratory symptoms.Environ Health Perspect1996104500505
10. Schlesinger RB, Cassee F. Atmospheric secondary inorganic particulate matter: the toxicological perspective as a basis for health effects risk assessment.Inhal Toxicol20029197235
11. Roney N; Department of Health & Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, AtlantaToxicological Profile for Ammonia
12. Pedersen CB, Gotzsche H, Moller JO MP. The Danish Civil Registration System. A cohort of eight million persons.Dan Med Bul200639441449
13. Christensen JH. The Danish Eulerian hemispheric model—A three-dimensional air pollution
model used for the Arctic.Atmos Environ19973141694191
14. Brandt J, Silver JD, Frohn LM, et al. An integrated model study for Europe and North America using the Danish Eulerian Hemispheric Model with focus on intercontinental transport of air pollution
15. Geels C, Andersen HV, Skjøth CA, et al. Improved modelling of atmospheric ammonia
over Denmark using the coupled modelling system DAMOS.Biogeosciences2012726252647
16. Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient Register.Scand J Public Heal2011393033
17. Guarnieri M, Balmes JR. Outdoor air pollution
18. Smit LAM, Heederik D, Doekes G, et al. Occupational endotoxin exposure reduces the risk of atopic sensitization but increases the risk of bronchial hyperresponsiveness.Int Arch Allergy Immunol2010152151158
19. Donham KJ, Cumro D, Reynolds S. Synergistic effects of dust and ammonia
on the occupational health effects of poultry production workers.J Agromedicine201585776
20. Radon K, Schulze A, Ehrenstein V, et al. Environmental exposure to confined animal feeding operations and respiratory health of neighboring residents.Epidemiology20073300308
21. Sigurdarson ST, Kline JN. School proximity to concentrated animal feeding operations and prevalence of asthma
22. Elholm G, Schlünssen V, Doekes G, et al. Become a farmer and avoid new allergic sensitization: adult farming exposures protect against new-onset atopic sensitization.J Allergy Clin Immunol201313212391241
23. Mirabelli MC, Wing S, Marshall SW, Wilcosky TC. Asthma
symptoms among adolescents who attend public schools that are located near confined swine feeding operations.Pediatrics20061e66e75
24. Pavilonis BT, Sanderson WT, Merchant JA. Relative exposure to swine animal feeding operations and childhood asthma
prevalence in an agricultural cohort.Environ Res20131227480
25. Grabenhenrich LB, Gough H, Reich A, et al. Early-life determinants of asthma
from birth to age 20 years: A German birth cohort study.J Allergy Clin Immunol20144979988
26. Seeman JI, Carchman RA. The possible role of ammonia
toxicity on the exposure, deposition, retention, and the bioavailability of nicotine during smoking.Food Chem Toxicol2008618631881
27. Willems EW, Rambali B, Vleeming W, et al. Significance of ammonium
compounds on nicotine exposure to cigarette smokers.Food Chem Toxicol20065678688
28. Skjøth CA, Geels C, Berge H, et al. Physics Spatial and temporal variations in ammonia
emissions – a freely accessible model code for Europe.Atmos Chem Phys20111152215236
29. Adams K, Greenbaum DS, Shaikh R, et al. Particulate matter components, sources, and health: Systematic approaches to testing effects.J Air Waste Manag Assoc2015655445458
30. Collins SA, Pike KC, Inskip HM, et al. Validation of novel wheeze phenotypes using longitudinal airway function and atopic sensitization data in the first 6 years of life: evidence from the Southampton Women’s survey.Pediatr Pulmonol20137683692