Early life exposure to ambient air pollution and childhood asthma in China

Environmental Research 143 (2015) 83–92

Contents lists available at ScienceDirect

Environmental Research journal homepage: www.elsevier.com/locate/envres

Early life exposure to ambient air pollution and childhood asthma in China Qihong Deng a,b,n, Chan Lu a, Dan Norbäck c, Carl-Gustaf Bornehag d, Yinping Zhang e, Weiwei Liu a,b, Hong Yuan b,f, Jan Sundell a,e a

School of Energy Science and Engineering, Central South University, Changsha, China Institute of Environmental Health, Central South University, Changsha, China c Department of Medical Sciences/Occupational & Environmental Medicine, Uppsala University, Uppsala, Sweden d Department of Public Health Sciences, Karlstad University, Karlstad, Sweden e School of Architecture, Tsinghua University, Beijing, China f The Third Xiangya Hospital, Central South University, Changsha, China b

art ic l e i nf o

a b s t r a c t

Article history: Received 16 May 2015 Received in revised form 29 September 2015 Accepted 29 September 2015

Background: Early life is suggested to be a critical time in determining subsequent asthma development, but the extent to which the effect of early-life exposure to ambient air pollution on childhood asthma is unclear. Objectives: We investigated doctor-diagnosed asthma in preschool children due to exposure to ambient air pollution in utero and during the first year of life. Methods: In total 2490 children aged 3–6 years participated in a questionnaire study regarding doctordiagnosed asthma between September 2011 and January 2012 in China. Children’s exposure to critical air pollutants, sulfur dioxide (SO2) as proxy of industrial air pollution, nitrogen dioxide (NO2) as proxy of traffic pollution, and particulate matterr 10 mm in diameter (PM10) as a mixture, was estimated from the concentrations measured at the ambient air quality monitoring stations by using an inverse distance weighted (IDW) method. Logistic regression analysis was employed to determine the relationship between early-life exposure and childhood asthma in terms of odds ratio (OR) and 95% confidence interval (CI). Results: Association between early-life exposure to air pollutants and childhood asthma was observed. SO2 and NO2 had significant associations with adjusted OR (95% CI) of 1.45 (1.02–2.07) and 1.74 (1.15– 2.62) in utero and 1.62 (1.01–2.60) and 1.90 (1.20–3.00) during the first year for per 50 mg/m3 and 15 mg/m3 increase respectively. Exposure to the combined high level of SO2 and NO2 in China significantly elevated the asthmatic risk with adjusted OR (95% CI) of 1.76 (1.18–2.64) in utero and 1.85 (1.22–2.79) during the first year compared to the low level exposure. The associations were higher for males and the younger children aged 3–4 than females and the older children aged 5–6. Conclusions: Early-life exposure to ambient air pollution is associated with childhood asthma during which the level and source of air pollution play important roles. The high level and nature of combined industrial and traffic air pollution in China may contribute to the recent rapid increase of childhood asthma. & 2015 Elsevier Inc. All rights reserved.

Keywords: Asthma Children In utero Pregnancy First year of life Sulfur dioxide Nitrogen dioxide Particulate matter

1. Introduction Asthma is the most common chronic disease in children and the leading cause of pediatric hospitalization worldwide (Barnett and Nurmagambetov, 2011). The incidence and prevalence of

n Corresponding author at: School of Energy Science and Engineering, Central South University, Changsha 410083, Hunan, China. E-mail address: [email protected] (Q. Deng).

http://dx.doi.org/10.1016/j.envres.2015.09.032 0013-9351/& 2015 Elsevier Inc. All rights reserved.

childhood asthma increased markedly during the 20th century (Eder et al., 2006). Although the rising trend in developed countries may have recently plateaued due to better clinical treatment and public health efforts, the prevalence in developing countries is rapidly increasing (Ait-Khaled et al., 2007). In China, the prevalence of asthma among urban children has increased dramatically from 1.5% to 6.8% over the past decade (Zhang et al., 2013). The substantial increase implies a major public health problem given the huge population of China.

84

Q. Deng et al. / Environmental Research 143 (2015) 83–92

Fig. 1. Distributions of kindergartens and air pollution monitoring stations in Changsha.

Although it is largely accepted that gene-environment interactions are responsible for the development of asthma, environmental factors are likely responsible for the increase in the prevalence of asthma because population genetic variability does not change such rapidly (Kasznia-Kocot et al., 2010; London 2007). Despite the fact that pregnant women and children spend most of their time indoors, mounting evidence suggests that long-term exposure to ambient air pollution contributes to the development of asthma. However, it still remains controversial to ascertain the critical exposure window, the concentration level and specific sources of air pollution responsible for the observed effects (Anderson et al., 2013; Sarnat and Holguin, 2007; Trasande and Thurston, 2005). Developed countries focused attention on the prevailing traffic-related air pollution, represented by nitrogen dioxide (NO2) and fine particulate matter (PM2.5, aerodynamic diameterr2.5 mm), and numerous epidemiologic studies have reported its significant association with childhood asthma (Brunekreef and Sunyer, 2003; Gasana et al., 2012; Gauderman et al., 2005; Islam et al., 2007; Jerrett et al., 2008) although some studies did not find the association (Mölter et al., 2014, 2015). However, China is now still facing the world worst classic industry-related air pollution (Kan et al., 2012), represented by sulfur dioxide (SO2) and coarse particulate matter (PM10, aerodynamic diameterr10 mm), and the rapid increasing amount of vehicles in larger cities in recent years led to a combined classical industrial and modern traffic air pollution in China. Difference in the nature of air pollution, both level and source, between China and developed countries prompts the need for further investigation into the role of air pollution in the development of asthma. Asthma is a chronic inflammatory disorder of the airways and most cases are now considered to originate in early life (Duijts, 2012; Gluckman et al., 2008). According to Barker hypothesis (Barker and Osmond, 1986), prenatal and early postnatal exposures influence developmental plasticity and result in altered programming which leads to the development of a variety of complex diseases. Accordingly, exposures to air pollutants during

the prenatal period (growth of the airways) and shortly afterward, especially during the first year of life (expansion of alveoli) should be an important determinant in the later development of childhood asthma. However, very few studies have examined the effect of exposure of early infants to air pollution on their subsequent onset of asthma (Brauer et al., 2002; Clark et al., 2010; Gruzieva et al., 2013; Morgenstern et al., 2007; Nishimura et al., 2013) and only one study investigated the exposure of fetuses in utero (Clark et al., 2010). These scarce studies in developed countries also showed that early-life exposure to traffic-related air pollutants induced new onset of asthma. The evidence is so limited that the causal link between early-life exposure to air pollution and childhood asthma needs to be urgently explored in different populations and sites, especially in China where rapidly increasing prevalence of childhood asthma and different air pollution mixture were observed. We hypothesized that the rapid increase in the incidence of childhood asthma in China was associated with early life exposure to the high industrial and traffic air pollution mixture. To examine this, we engaged in a nationwide “China-Children-Homes-Health (CCHH)” study (Zhang et al., 2013) to investigate the relationship between air pollution exposure during in utero and the first year of life and doctor-diagnosed asthma in preschool children.

2. Materials and methods 2.1. Study population Between September 2011 and January 2012, we conducted a survey for respiratory disease and symptoms in children in the kindergartens in Changsha, the capital city of Hunan Province in south-central China, having a population of 7.22 million and covering an area of 1909 km2 (Fig. 1). The study protocol were reviewed and approved by the Ethics Committee of Central South University and also by the health department and school board of

Q. Deng et al. / Environmental Research 143 (2015) 83–92

each kindergarten. A Chinese version of the questionnaire designed by the International Study of Asthma and Allergies in Childhood (ISAAC) was administered to collect information on health status of children and family members (Asher et al., 2006), with some changes to address housing and cultural characteristics in China. A total of 4988 questionnaires were randomly distributed to the children at 36 participating kindergartens, of which 21 kindergartens were located in downtown area and 15 kindergartens in suburban area (Fig. 1). Children were instructed to have the questionnaire completed by parents and to return it to kindergartens within one week. We received 3897 completed questionnaires and the overall response rate was 78%. We first excluded 745 children from kindergartens having a response rate lower than 50%, because low response rate is often correlated with low socioeconomic, educational, and health conditions (Nohr et al., 2006). These excluded kindergartens were basically distributed in the suburban areas where the children are mainly from the far rural areas and their parents are mainly rural migrant workers and thus the children’s exposure during pregnancy and the first year cannot be estimated by using the urban data. Then, we excluded 162 children with low birth weight ( o2.5 kg) and preterm birth (o 37 weeks of gestation), and 10 children with multiple births, as these conditions may confound the association between air pollution and asthma (Brauer et al., 2008; Clark et al., 2010). On the other hand, the children aged 3–6 were chosen in our study because there are few children4 6 years in kindergartens and diagnosis of asthma among childreno3 years is often confused. The children without information on the health outcome, covariates, and air pollution exposure were also excluded. Finally, the responses from 2490 valid questionnaires were entered into a database. 2.2. Exposure assessment We selected three pollutants, SO2, NO2, and PM10 to represent ambient air pollution; SO2 was used as an indicator of industryrelated air pollution, NO2 as an indicator of traffic-related air pollution, and PM10 as a surrogate of complex mixture of air pollutants (Kan et al., 2012). Daily 24-h average ambient concentrations of SO2, NO2 and PM10 were obtained from 7 municipal air quality monitoring stations during the period 2005–2010, that is, from the year when the oldest child was gestated to one year after the youngest child was born. Measurements at the monitoring stations strictly followed the standard methods set by the State Environmental Protection Agency of China: SO2 by ultraviolet fluorescent method (ML/EC9850, Ecotech, Australia), NO2 by the chemiluminescent method (ML/EC9841B, Ecotech, Australia), and PM10 by a tapered element oscillating microbalance (TEOM1400, Rupprecht & Patashnick, USA). We used an inverse distance weighted (IDW) method (Marshall et al., 2008; Bell, 2006) to estimate the air pollution concentrations at the kindergartens from the data at the monitoring stations. The concentration at the each kindergarten was interpolated by the concentrations at the closest four monitoring stations, with an inverse of the squared distance between the kindergarten and closest monitoring station (1/d2) used as the weighting function; the average distance (d) was 4.83 km. At first, we used the available daily average concentrations at 7 monitoring stations to obtain the daily mean concentrations at 36 kindergartens, and then the monthly mean concentrations of SO2, NO2 and PM10 at each kindergarten were computed as the averages of the daily mean concentrations at the kindergarten within each month. The exposure of children was calculated in terms of the air pollutant concentrations at their kindergartens, because parents usually enrolled their children in the kindergarten nearest to their home and hence exposure at home was considered to be the same

85

as that at the kindergarten. Exposure in utero was defined as the averages of the monthly mean concentrations of SO2, NO2 and PM10 during the full gestational months, and exposure during the first year were calculated by averaging the monthly mean concentrations from the month of birth to the 12th month after birth. Since only the information about the birth month, not the birth date, of each child was obtained from the questionnaire, we assumed that each child was born on the first day of the birth month, and then estimated the gestational months and the first year of life for each child. The children’s migration from one kindergarten to another could lead to misclassification of exposure, but we assume that children’s migration does not have impact on the results (please see Supplementary materials). One reason is that parents usually moved their houses within a small distance due to their work requirements, and on the other hand the errors in exposure assessment were likely to be random, which would not introduce positive bias into the associations. 2.3. Definition of health outcome Doctor-diagnosed asthma was identified on the basis of an affirmative response to the question: “Has a doctor ever diagnosed your child as having asthma?” 2.4. Covariates Potential confounding variables were obtained from the parent reported responses to the questionnaires. The considered covariates for child were sex, age, birth weight, breastfeeding, gestational age, and living area and covariates for parents were parental smoking during pregnancy, maternal age (the mother’s age at the time of child’s birth), parental atopy, and socioeconomic status. Children's living areas were divided into downtown and suburban areas according to the addresses of their kindergartens. Parental atopy is a measure of genetic predisposition to asthma and was defined as the father or mother of the index child ever having been diagnosed as having asthma or other allergic diseases. We used the house size and mother’s occupation to indicate the socioeconomic conditions of the parents (Nordling et al., 2008). It is worthy to note that nutrition, infection, and medication are also important covariates for childhood asthma, but were not obtained during the questionnaire survey and thus were not considered. 2.5. Statistical analysis We used logistic regression models to analyze the data. First, a simple univariate analysis was performed to assess the association between potential covariates and doctor-diagnosed asthma in children to obtain the significant covariates. Secondly, we employed the simple logistic regression analysis to estimate the crude effect of the exposure to each air pollutant on the childhood asthma. Thirdly, multiple logistic regression analysis was performed to evaluate the effects after adjustment for the significant covariates. Fourthly, the multiple logistic regression analysis was also performed in multi-pollutant model to evaluate the effects after adjustment for other air pollutants. The effect was presented as odds ratio (OR) and 95% confidence interval (CI). The p value less than 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (version 16.0, SPSS Inc., Chicago, USA). We assess the association between childhood asthma outcomes and air pollutants by both continuous and discrete models. In continuous model, ORs were estimated by an unit increase in the exposure level for each pollutant: one is by two standard deviations (SD) increase, i.e. 20 mg/m3 in PM10, 50 mg/m3 in SO2, and

86

Q. Deng et al. / Environmental Research 143 (2015) 83–92

Table 1 Cumulative incidence and odds ratios of childhood asthma by covariates. Responders (n¼ 2490)

Asthma n

Sex* Male Female Age (years)* 3

OR (95% CI) %

(95% CI)

p value o0.001

1337 1153

114 57

8.5 4.9

645

25

3.9

(7.0–10.0) (3.7–6.2)

1.79 1.00

(1.29–2.49)

0.023 1.00 (2.4–5.4)

4

916

77

8.4

2.28 (6.6–10.2)

5

769

57

7.4

6

160

12

7.5

(1.43–3.62) 1.99

(5.6–9.3)

(1.23–3.22) 2.01

(3.4–11.6) Birth weight (kg) 2.5–3.0

(0.99–4.10) 0.840

297

25

8.4

1.19

(0.75–1.90)

(5.2–11.6) 3.0–3.5

1144

82

7.2

1.00 (5.7–8.7)

3.5–4.0

766

44

5.7

0.79

(0.54–1.15)

0.99

(0.59–1.64)

0.55 1.00

(0.35–0.87)

0.91 1.00 0.99

(0.54–1.54)

(4.1–7.4) 44.0

283

20

7.1 (4.1–10.1)

Breast-feeding* Yes No Gestational age (weeks) 37–39 39–41 441 Living area* Downtown Suburban Parental atopy* Yes No Parental smoking during pregnancy Yes No Maternal age (years) o25

0.009 2275 215

147 24

6.5 11.2

282 1206 1002

18 84 69

6.4 7.0 6.9

(5.5–7.5) (6.9–15.4) 0.766 (3.5–9.3) (5.5–8.4) (5.3–8.5)

(0.71–1.37)

0.042 1829 661

137 34

7.5 5.1

331 2159

64 107

19.3 5.0

1205 1285

87 84

7.2 6.5

411

25

6.1

(6.3–8.7) (3.5–6.8)

1.49 1.00

(1.01–2.20)

4.60 1.00

(3.29–6.43)

1.11 1.00

(0.82–1.52)

0.89

(0.56–1.40)

o0.001 (15.1–23.6) (4.0–5.9) 0.501 (5.8–8.7) (5.2–7.9) 0.850 (3.8–8.4)

25–30

1384

94

6.8

1.00 (5.5–8.1)

30–35

560

41

7.3

1.08

(0.74–1.59)

1.22

(0.64–2.34)

0.82

(0.53–1.28)

(5.2–9.5) 435

135

11

8.1 (3.5–12.8)

Family house size (m2) r60

0.354 430

28

6.5 (4.2–8.9)

61–100

1051

82

7.8

1.00 (6.2–9.4)

101–150

818

51

6.2

0.79

(0.55–1.13)

0.65

(0.33–1.28)

(4.6–7.9) 4150

191

10

5.2 (2.0–8.4)

Maternal occupation Unemployed

0.598 980

62

6.3

1.00 (4.8–7.9)

Unskilled blue-collar workers

160

11

6.9

1.09 (2.9–10.8)

Skilled blue-collar workers

37

4

10.8

Low level white-collar workers

674

53

7.9

High level white-collar workers

361

21

5.8

Others

278

20

7.2

(0.56–2.12) 1.80

(0.3–21.3)

(0.62–5.23) 1.26

(5.8–9.9)

(0.86–1.85) 0.92

(3.4–8.2) (4.1–10.3) *

Statistically significant difference in asthma stratified by covariates (po 0.05).

(0.55–1.52) 1.15 (0.68–1.94)

Q. Deng et al. / Environmental Research 143 (2015) 83–92

15 mg/m3 in NO2, so as to make the associations comparable between the air pollutants (Hertz-Picciotto et al., 2007), and the other is by a fixed increase, 10 mg/m3, for all pollutants in order to compare the associations between different groups. In discrete model, the exposure range was divided into four quartiles, and ORs were estimated for different exposure quartiles of air pollutants, in which the first (lowest) quartile was set as reference, so as to illustrate the risk difference between the exposure levels.

3. Results Of 2490 respondents, 171 (6.9%) reported doctor-diagnosed asthma. Associations between potential covariates and childhood asthma are shown in Table 1. The prevalence of asthma was significantly higher for males (8.5%), those without breast-feeding (11.2%), and those with parental atopy (19.3%) than the females (4.9%), those with breast-feeding (6.5%), and those without parental atopy (5.0%). Spatial and temporal disparities in the doctordiagnosed asthma were observed; the asthmatic risk for children living in downtown area was significantly higher than those living in suburban area (p ¼0.042) and the older children had significantly more doctor-diagnosed asthma than the younger (p ¼0.023). No significant difference was observed for the children with different birth weights and gestational ages after the children with low birth weight and preterm birth had been excluded in the present study. Also, no significant difference was found with respect to parental smoking during pregnancy (p ¼0.501). The doctor-diagnosed asthma was not associated with mother's age (p ¼0.850) and did not vary significantly among neighborhoods with different socioeconomic status. The significant covariates, i.e. sex, age, breast-feeding, living area, and parental atopy, were selected to be included in the statistical models. Table 2 summarizes the children's exposure levels. The average individual exposures (mean 7SD) to PM10, SO2 and NO2 were 110 711 mg/m3, 82 726 mg/m3 and 46 78 mg/m3 in utero and 10378 mg/m3, 68 724 mg/m3 and 48 77 mg/m3 during the first year, respectively. The levels of PM10 and SO2 were substantially decreased due to the effective measures taken by the government to improve the air quality but NO2 level was slightly increased due to the increase in the number of vehicles from the period in utero to the first year of life. Table 3 provides effect estimates of childhood asthma by continuous concentration increases in twice the SD of PM10 (20 mg/m3), SO2 (50 mg/m3), and NO2 (15 mg/m3) exposures. Early life exposure to traffic air pollutant NO2 had the highest and

87

Table 3 Odds ratios (95% CI) of childhood asthma by 2 SD increases in the pollutant concentrations.

In utero PM10 SO2 NO2 First year PM10 SO2 NO2 a

Increment (μg/m3)

Crude

Adjusteda

20 50 15

1.26 (0.96, 1.66) 1.68 (1.27, 2.21) 1.42 (1.06, 1.91)

1.04 (0.69, 1.58) 1.45 (1.02, 2.07) 1.74 (1.15, 2.62)

20 50 15

1.44 (1.01, 2.06) 1.80 (1.33, 2.43) 1.81 (1.28, 2.55)

1.28 (0.74, 2.23) 1.62 (1.01, 2.60) 1.90 (1.20, 3.00)

Adjusted for sex, age, breast-feeding, parental atopy, and living area.

significant association with ORs (95% CIs) of 1.74 (1.15–2.62) in utero and 1.90 (1.20–3.00) during the first year, respectively. SO2 had also significant association with ORs (95% CIs) of 1.45 (1.02– 2.07) in utero and 1.62 (1.01–2.60) during the first year. The association for PM10 was not significant and the lowest. Effect estimates appeared higher for infants’ direct exposure during the first year of life than those for fetuses’ indirect exposure in utero, but we cannot distinguish the effects in utero and during the first year of life because of a strong correlation between the exposures during the two periods (Table 4). Fig. 2 provides odds ratios for childhood asthma by exposure quartiles of air pollutants. The results were similar to those obtained by the continuously increasing model. It is clear that the upper quartiles of exposure for SO2 and NO2 increased the asthma risk with respect to the lowest quartile of exposure both in utero and during the first year; the risks were not significant at low exposure levels but became significant at high exposure levels. The ORs of PM10 for asthmatic risk were not significant at all exposure quartiles in utero, but during the first year the risk of PM10 may be significant at high exposure level although it was not significant from the view of the whole exposure domain (Table 3). To consider the impact of children’s migration, we analyzed the exposure level and its asthmatic risk for the 1671 (67.1%) children who did not move during pregnancy and the first year of life (please see Tables S1 and S2 in the Supplementary materials) since the moving patterns of the children were not known. The air pollutant exposure levels for the stayed children were nearly the same as those for the whole 2490 children. The odds ratios of asthma for the stayed children were slightly changed due to different number of children but the tendencies were consistent with those for the whole children. Therefore, the influence of the children’s migration in our present work was very limited.

Table 2 Descriptive statistics of exposure levels for children aged 3–6 years. Mean 7 SD Total In utero PM10

Minimum

25th percentile

50th percentile

75th percentile

Maximum

86

103

108

115

157

35

62

75

98

163

28

40

45

52

66

85

96

103

108

138

28

51

64

79

148

31

42

46

53

62

1107 11 SO2 82 726 NO2 46 78 First-year PM10 1037 8 SO2 687 24 NO2 487 7 Abbreviations: PM10 (μg/m3), particulate matterr 10 μm in aerodynamic diameter; SO2 (μg/m3), sulfur dioxide; NO2 (μg/m3), nitrogen dioxide; SD, standard deviation.

88

Q. Deng et al. / Environmental Research 143 (2015) 83–92

Table 4 Pearson correlation between air pollutants during different exposure periods. In utero

In utero PM10 SO2 NO2 First-year PM10 SO2 NO2 a

Table 5 Adjusted odds ratios (95% CI) of childhood asthma by multiple pollutant model. Increment (μg/m3)

First-year

PM10

SO2

NO2

PM10

SO2

NO2

1

0.27a 1

 0.32a 0.48a 1

0.53a 0.23a  0.34a

0.58a 0.82a 0.13a

 0.25a 0.68a 0.86a

1

0.50a 1

 0.10a 0.42a 1

In utero PM10

20

SO2

50

NO2

15

First-year PM10

20

Multi-pollutant modela

Correlation is significant at the 0.01 level (2-tailed).

We examined the effect of interaction between air pollutants from different sources. The correlations between pollutants during pregnancy and the first year were shown in Table 4. Weak or moderate correlations were observed between different pollutants within the same exposure period (from 0.32 to 0.50), and thus a multi–pollutant model was performed (Table 5). The asthmatic risk of PM10 was not affected by the other pollutants, as we found its ORs (95% CIs) were basically kept constant when taking other pollutants into account, such as: 1.04 (0.69–1.58) for PM10 alone (Table 3), 1.07 (0.71–1.60) for adding SO2, 1.02 (0.68–1.52) for adding NO2, and 1.02 (0.68–1.53) for adding both SO2 and NO2 in utero. Meanwhile, the asthmatic risks of SO2 and NO2 were only slightly changed after adjusting for PM10. However, due to the moderate correlation between the traffic-related and industryrelated pollutant, the asthmatic risks of NO2 and SO2 were substantially changed when adjusted for each other, as we found that ORs (95% CIs) of SO2 and NO2 were changed from 1.45 (1.02–2.07) and 1.74 (1.15–2.62) to 1.00 (0.57–1.78) and 1.73 (0.90–3.35) in utero and from 1.62 (1.01–2.60) and 1.90 (1.20–3.00) to 1.05 (0.54– 2.04) and 1.84 (0.98–3.44) during the first year, respectively. The air pollutants exposures between in utero and the first-year were highly correlated, such as 0.82 for SO2 and 0.86 for NO2, and hence the mutual model was not valid, and therefore we cannot distinguish the effects of the air pollutant exposures between in utero and the first year. Due to the correlation between industrial and traffic air pollutants, we investigated the combined asthmatic risk of SO2 and NO2. Table 6 shows the risk of childhood asthma for the combined SO2 and NO2 at different categories of exposure level. We found that the asthmatic risk of the combinedexposure to high SO2 and NO2 was highly significant (p o0.01) with adjusted OR (95% CI) of

SO2

50

NO2

15

PM10 þ SO2

PM10 þ NO2

SO2 þ NO2

PM10 þSO2 þNO2

1.07 (0.71, 1.60) 1.46 (1.03, 2.08) –

1.02 (0.68, 1.52) –

–

1.02 (0.68, 1.53)

1.36 (0.79, 2.33) 1.68 (1.04, 2.70) –

1.29 (0.76, 2.21) –

1.74 (1.15, 2.62)

1.91 (1.21, 3.02)

1.00 (0.57, 1.01 (0.56, 1.80) 1.78) 1.73 (0.90, 1.73 (0.89, 3.36) 3.35) –

1.31 (0.76, 2.24)

1.05 (0.54, 1.10 (0.56, 2.16) 2.04) 1.84 (0.98, 1.79 (0.96, 3.36) 3.44)

a Adjusted for sex, age, breast-feeding, parental atopy, living area, and other one or two air pollutants within the same exposure period.

Table 6 Adjusted odds ratios (95% CI) of childhood asthma for the combined SO2 and NO2. Combined exposure SO2

NO2

Low High Low High

Low Low High High

In utero

First year

1.00 1.57 (0.94, 2.65) 1.16 (0.64, 2.11) 1.76 (1.18, 2.64)

1.00 1.02 (0.56, 1.86) 1.63 (0.93, 2.84) 1.85 (1.22, 2.79)

Low and high exposure of air pollutant is indicated by the personal exposureo and Z median of each pollutant exposure. Adjusted for sex, age, breastfeeding, parental atopy, and living area.

1.76 (1.18–2.64) in utero and 1.85 (1.22–2.79) during the first year. Table 7 compares the associations between asthma and air pollutants for younger and older children at different early life exposure levels shown in Fig. 3. Effect estimates appeared larger for the young children aged 3–4 than the older aged 5–6, although confidence intervals overlapped between groups. Only associations for SO2 were significant for older children, but associations for both SO2 and NO2 were significant for the younger children in utero and during the first year of life. The association for PM10 was also significant for the younger children during the first year exposure.

Fig. 2. Adjusted odds ratios of childhood asthma during in utero and first year by exposure quartiles relative to 1st quartile. The exposure levels are divided into four quartiles according to Table 2: 1st quartile ¼ below the 25th percentile; 2nd quartile¼ between the 25th and 50th percentiles; 3rd quartile ¼between the 50th and 75th percentiles; 4th quartile ¼ above the 75th percentile. Adjusted for sex, age, breast-feeding, parental atopy, and living area.

Q. Deng et al. / Environmental Research 143 (2015) 83–92

Table 7 Adjusted odds ratios (95% CI) of childhood asthma for air pollutants stratified by children's age.

In utero PM10 SO2 NO2 First year PM10 SO2 NO2

Increment (μg/m3)

3–4 years old

5–6 year old

10 10 10

1.27 (0.89, 1.83) 1.13 (1.02, 1.24) 1.24 (1.04, 1.64)

1.13 (0.89, 1.44) 1.10 (1.01, 1.20) 1.43 (0.87, 2.05)

10 10 10

1.81 (1.31, 2.50) 1.21 (1.06, 1.39) 1.45 (1.06, 1.98)

0.89 (0.64, 1.24) 1.13 (1.02, 1.26) 1.40 (0.83, 2.04)

89

Table 8 Adjusted odds ratios (95% CI) of childhood asthma for air pollutants stratified by sex.

In utero PM10 SO2 NO2 First year PM10 SO2 NO2

Increment (μg/m3)

Male

Female

10 10 10

1.10 (0.85, 1.42) 1.10 (1.01, 1.20) 1.48 (1.06, 2.06)

0.87 (0.61, 1.26) 1.03 (0.91, 1.17) 1.40 (0.87, 2.26)

10 10 10

0.88 (0.63, 1.22) 1.14 (1.02, 1.28) 1.57 (1.08, 2.26)

2.23 (1.30, 3.85) 1.02 (0.85, 1.21) 1.51 (0.87, 2.61)

# Adjusted for sex, breast-feeding, parental atopy and living area.

# Adjusted for age, breast-feeding, parental atopy and living area.

Table 8 provides associations with air pollution stratified by sex. The boys were significantly affected by industrial and traffic air pollutants, SO2 and NO2, not only in utero with ORs (95% CI)¼ 1.10 (1.01–1.20) and 1.48 (1.06–2.06) respectively but also during the first year with ORs (95% CI)¼1.14 (1.02–1.28) and 1.57 (1.08– 2.26), but girls were only significantly affected by PM10 during the first year of life with OR (95% CI)¼2.23 (1.30–3.85).

pollution are of greater significance than later exposure due to the susceptibility of target organs and systems during developmental periods of life (Gehring et al., 2002; Mortimer et al., 2008; Nordling et al., 2008). However, very few studies examined the effect of early life exposure to air pollution on the prevalence or incidence of asthma in children. A study in the Netherlands first observed that traffic-related air pollution was associated with childhood asthma at 2 years of age and found stronger associations before 1 year of age (Brauer et al., 2002). Later, another birth cohort study from France further observed that the asthma epidemic was significantly associated with exposure to traffic-related pollutants before age of 3 but not associated with lifelong exposure (Zmirou et al., 2004). Subsequently, the significant impact of the exposure to air pollution during the first year of life on the later development of asthma received more attentions (Carlsten et al., 2011; Clark et al., 2010; Gruzieva et al., 2013; Morgenstern et al., 2007; Nishimura et al., 2013). So far, only one study in Canada (Clark et al., 2010) focused on the exposure during pregnancy and found that the asthmatic risk for children aged 3–4 years were respectively 1.10 (1.05–1.15), 1.09 (1.05–1.13), 1.03 (1.02–1.05) for 10, 1, and 1 mg/m3 increase in exposures of NO2, PM10, and SO2 in utero. Above studies in developed countries have demonstrated that early life exposure to the governing traffic-related pollutants was significantly associated with the later development of asthma, which was confirmed in our study. However, this evidence is not sufficient. Our study also illustrated significant effects of early life exposure to the prevailing industrial and particulate air pollution in China.

4. Discussion 4.1. Main findings of this study This study is, to the best of our knowledge, the first in China and one of the few to examine the effect of early life exposure to ambient air pollution in utero and during the first year on the risk of pediatric asthma. The novelty of this study is the examination of a population with much higher overall exposure than has been examined previously, especially to industrial pollution. We found that the doctor-diagnosed childhood asthma was significantly associated with the early life exposure to ambient criteria air pollutants PM10, SO2 and NO2. The associations were stronger for males and the younger children aged 3–4 than females and the older children aged 5–6. Exposure level and source of the air pollution played an important roles in the course of development of childhood asthma. We found the asthmatic risk of exposure to high industrial air pollution (SO2) in China was significant and cannot be overlooked. As the traffic air pollution increases in China, its risk for childhood asthma was increased and becomes significant. The combination of exposure to high industrial and traffic air pollutants significantly increased the risk. 4.2. Importance of early life exposure in the development of childhood asthma It was surmised that prenatal and early life exposures to air

4.3. The role of source and level of air pollution For the traffic air pollutant NO2, we found that the early life exposure was a significant risk factor for childhood asthma, which is consistent with the results reported in developed countries where the traffic air pollution is high (Brauer et al., 2002, 2007; Clark et al., 2010; Gehring et al., 2010; Gruzieva et al., 2013;

Fig. 3. Exposure levels of air pollutants for children aged 3–4 and 5–6 during early life exposure.

90

Q. Deng et al. / Environmental Research 143 (2015) 83–92

Morgenstern et al., 2007, 2008; Nishimura et al., 2013; Zmirou et al., 2004). However, our discrete model further demonstrated that the effect of traffic air pollutant NO2 was closely related to its exposure level, i.e. it was significant at high exposure level but not significant at low exposure level. Some studies also showed the effect of traffic air pollution at different exposure levels (Clark et al., 2010; Studnicka et al., 1997; Zmirou et al., 2004) and a recent meta-analysis (Gasana et al., 2012) concluded that living or attending schools near high traffic density roads with higher levels of motor vehicle air pollutants increased the incidence and prevalence of childhood asthma. For industrial air pollutant SO2, we observed significant correlation between early life exposure and childhood asthma in China. Two possible explanations for this trend are: firstly the SO2 level in China was shown to be the highest in the world (Watts, 2006) due to China’s surging economy along with a coal-dominated energy structure, and secondly the effect of SO2 was dependent of its exposure level and was significant at high exposure level. Recently, the effect of long-term exposure to outdoor SO2 on childhood asthma was observed, although it has been ignored in the developed countries due to its low concentration level (Dong et al., 2011; Pénard-Morand et al., 2010; Pikhart et al., 2001). Similar to our work, Clark et al. (2010) found that the asthmatic risk for children aged 3–4 years were significantly associated SO2 exposure in utero and during the first year of life. Several studies also observed the risks of SO2 at high concentrations near sources (Pénard-Morand et al., 2010; Smargiassi et al., 2009). Therefore, the role of early life exposure to industrial air pollutant SO2 in Chinese cities cannot be overlooked due to its high exposure level, which is different from the developed countries. For the particulate air pollutant PM10, we found that it was significant at high exposure level during the first year (Fig. 2), although it was not significant in the continuous model from the view of the whole exposure domain (Table 3). Some studies reported that childhood asthma was not related to PM10 (Akinbami et al., 2010; Gruzieva et al., 2013; Hwang et al., 2005; McConnell et al., 2010; Zhang et al., 2002), but several studies observed the effects were significant (Pénard-Morand et al., 2010; Pikhart et al., 2001). The inconsistence may be due to the different exposure levels. We found the asthma risk of PM10 was not significant for the older children aged 5–6 but was significant for the younger children aged 3–4. This result is consistent with the two related studies in China: an early study found that childhood asthma was not related to PM10 (Zhang et al., 2002) but a recent study observed that the asthma risk of PM10 was significant (Dong et al., 2011). 4.4. Rapidly increasing asthma prevalence in China related to the nature of air pollution The characteristics of air pollution in China are twofold. One is that the air pollution level in China is much higher than that in the developed countries and the international guidelines. The average individual exposure to PM10, SO2 and NO2 (mean7 SD) in our study was high, 110 711 mg/m3, 82 726 mg/m3 and 46 78 mg/m3 in utero and 103 78 mg/m3, 68 724 mg/m3 and 48 77 mg/m3 during the first year, respectively. The other is that the air pollution in China is now changing from the classic industrial type to the combined industrial and traffic type. High level of industrial and traffic air pollution was associated with childhood asthma. During the past decade, the rapid urbanization in China has greatly worsened the air pollution in larger cities (Chan and Yao, 2008; Han et al., 2014). This should be the inherent reason that the prevalence of asthma among urban children has increased dramatically over the past decade (Zhang et al., 2013). On the other hand, we demonstrated exposure to the

combination of SO2 and NO2 heightened its effect compared to SO2 and NO2 alone (Peden, 1997), and therefore the heavily mixed classic/traffic nature of air pollution in China may contribute to the rapid increase in childhood asthma. 4.5. Pathogenesis and biological plausibility Our observed association between early life exposure to ambient air pollutants and childhood asthma is biologically plausible. The period of in utero and the first year of life is critical in the development of the immune and respiratory systems, and potential harmful effects of toxic pollutants during this period might result in long-lasting impaired capacity to fight infections and increased risk of allergic manifestations later in life (Gascon et al., 2015; Kozyrskyj et al., 2011; Warner 2004). Asthma is characterized by T-helper cell 2 (Th2) inflammation, leading to airway hyperresponsiveness and tissue remodeling. There is in vitro and in vivo evidence that air pollution may increase allergic inflammation and airway hyperresponsiveness through up-regulation of Th2 and down-regulation of T-helper cell 1 (Th1) that leads to a shift in the Th1/Th2 balance (Busses and Lemanske, 2001; Nadeau et al., 2010; Salvi et al., 2001; Takano et al., 1997; Zmirou et al., 2004). Recently, air pollutants have been considered to act as an adjuvant for allergic sensitization to inhaled allergens by increasing immunoglobulin E (IgE) antibodies. IgE can stimulate the immune system, which raises the possibility that long-term exposures may lead to increased prevalence of asthma and allergic diseases (Gould and Sutton, 2008; Li et al., 2003). On the other hand, asthma is a chronic inflammatory disorder of the airways (Gowers et al., 2012; Kim et al., 2013). Both SO2 and NO2 have direct pro-inflammatory effects on the sensitive airways of children (Lee et al., 2002; Peden, 1997). PM10 can produce a large quantity of reactive oxygen species which trigger the inflammatory processes in the respiratory tract (Li et al., 2003). The airways of an affected individual are unusually responsive to a wide range of stimuli and contract too much and too easily that was known as bronchoconstriction (Peden, 1997). Therefore, early life exposure to the air pollutants could cause chronic inflammatory disorder of the airways that caused childhood asthma. Recently, epigenetic regulation provides another mechanistic explanation for linking early-life exposure with subsequent development of asthma (Patel and Miller, 2009). Exposures to air pollutants during the prenatal period (growth of the airways) and the first year of life (expansion of alveoli) could cause altered programming, which can disrupt development and alter the gene transcription and expression without changes in the DNA sequence. These so-called epigenetic changes can create a phenotype of increased sensitivity to allergens or irritants, hyperresponsiveness and a skewed Th-2 response that caused a lasting vulnerability to asthma in later life (Martino and Prescott, 2011; Miller and Ho, 2008). Our results show a gender disparity in the patterns of doctordiagnosed asthma and their associations with the ambient air pollution. The different susceptibility to asthma could be explained by differences in airway size (De Marco et al., 2000). The airway size is on average smaller in boys than in girls in infancy (Tepper et al., 1986). Therefore, during childhood girls had a significantly lower risk of developing asthma than did boys. 4.6. Biases and limitations Some possible biases could exist in our study. First is the impact of moving pattern on exposure misclassification. We assumed and also validated that children's migration between kindergartens does not have significant impact on the results. However, this should be interpreted carefully because the moving effect may be

Q. Deng et al. / Environmental Research 143 (2015) 83–92

much larger for larger cities. Second is about parental smoking. Mother's smoking during pregnancy has been demonstrated as one of the most important influencing factors for childhood asthma. But in the present study, only 8 mothers (0.3%) smoked during pregnancy, and thus its significant effect was not observed. Third is the sample size. We used only 2490 (63.9%) returned questionnaires. The exclusion of the subjects not fitting our definition and with missing data, to some extent, may cause selection bias in the results. The last is the recall bias due to the retrospective questionnaire study. There are some limitations to this study. Firstly, given that the present study focused on ambient air pollutants, we have not considered the effect of exposures to indoor air pollution and allergens, such as dampness dampness/and moulds and pets, which can be associated with childhood asthma (Heinrich, 2011; Lau et al., 2000). Since no association between indoor exposures to air pollution and allergens and outdoor exposures to air pollutants, their effects on childhood asthma are independent to each other (Deng et al., 2015), but the potential interaction between outdoor air pollution and indoor air pollution and allergens may be important and will be further investigated in our next work. Secondly, we only considered three criteria air pollutants, PM10, SO2 and NO2, to represent the ambient air pollution. In China, these pollutants have been paid great attention and have been routinely monitored and the levels have been published to the public (Kan et al., 2012). However, it is important to further investigate the roles of other pollutants, particularly PM2.5, which was a better proxy for traffic air pollution (Gehring et al., 2002; Hertz-Picciotto et al., 2007). Thirdly, we assumed the same exposure to air pollution between kindergarten and home by considering the situation in China that parents usually enrolled their children in the kindergarten nearest to their home. However, the traffic pollution near the kindergartens is somewhat different from that at home (McConnell et al., 2010). Fourthly, exposure to air pollution was modeled by using an inverse distance weighted (IDW) method based on the data from ambient air quality monitoring stations. This modeling method may lead to bias of exposure because the number of monitoring stations was few and the sources of air pollution and the land use conditions were ignored. Fifthly, some covariates such as nutrition, infection, and medication are very important factors for childhood asthma, but were not considered in our study. Finally, as the first step of the nationwide project (CCHH), our study only included one city. Additional research in other cities is needed to confirm the persistence and generalizability of the results, given that the concentration levels of the air pollutants in our study were comparable to those recorded in a number of cities in China (Chan and Yao, 2008; Kan et al., 2012).

Conflicts of interest We declare that we have no conflicts of interest.

Acknowledgments This work was supported by the Natural Science Foundation of China (No. 51178466) and by the Doctoral Program of Higher Education of China (No. 20120162110011). The authors gratefully acknowledge all the children, their parents/guardians and the kindergartens for their time and enthusiastic participation. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.envres.2015.09.032.

91

References Ait-Khaled, N., Enarson, D.A., Bissell, K., Billo, N.E., 2007. Access to inhaled corticosteroids is key to improving quality of care for asthma in developing countries. Allergy 62, 230–236. Akinbami, L.J., Lynch, C.D., Parker, J.D., Woodruff, T.J., 2010. The association between childhood asthma prevalence and monitored air pollutants in metropolitan areas, United States, 2001–2004. Environ. Res. 110, 294–301. Anderson, H.R., Favarato, G., Atkinson, R.W., 2013. Long-term exposure to air pollution and the incidence of asthma: meta-analysis of cohort studies. Air Qual. Atmos. Health 6, 47–56. Asher, M., Montefort, S., Bjorksten, B., Lai, C.K.W., Strachan, D.P., Weiland, S.K., et al., 2006. Worldwide time trends in the prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and eczema in childhood: ISAAC phases one and three repeat multicountry cross-sectional surveys. Lancet 368, 733–743. Barker, D.J., Osmond, C., 1986. Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet 327, 1077–1081. Barnett, S.B.L., Nurmagambetov, T.A., 2011. Costs of asthma in the United States: 2002–2007. J. Allergy Clin. Immunol. 127, 145–152. Bell, M.L., 2006. The use of ambient air quality modeling to estimate individual and population exposure for human health research: a case study of ozone in the Northern Georgia Region of the United States. Env. Int. 32, 586–593. Brauer, M., Hoek, G., Smit, H.A., de Jongste, J.C., Gerritsen, J., Postma, D.S., et al., 2007. Air pollution and development of asthma, allergy and infections in a birth cohort. Eur. Respir. J. 29, 879–888. Brauer, M., Hoek, G., van Vliet, P., Meliefste, K., Fischer, P.H., Wijga, A., et al., 2002. Air Pollution from traffic and the development of respiratory infections and asthmatic and allergic symptoms in children. Am. J. Respir. Crit. Care Med. 166, 1092–1098. Brauer, M., Lencar, C., Tamburic, L., Koehoorn, M., Demers, P., Karr, C., 2008. A cohort study of traffic-related air pollution impacts on birth outcomes. Environ. Health Perspect. 116, 680–686. Brunekreef, B., Sunyer, J., 2003. Asthma, rhinitis and air pollution: is traffic to blame? Eur. Respir. J. 21, 913–915. Busses, W.W., Lemanske, R.F., 2001. Asthma. N Engl. J. Med. 344, 350–352. Carlsten, C., Dybuncio, A., Becker, A., Chan-Yeung, M., Brauer, M., 2011. Traffic-related air pollution and incident asthma in a high-risk birth cohort. Occup. Environ. Med. 68, 291–295. Chan, C.K., Yao, X., 2008. Air pollution in mega cities in China. Atmos. Environ. 42, 1–42. Clark, N.A., Demers, P.A., Karr, C.J., Koehoorn, M., Lencar, C., Tamburic, L., et al., 2010. Effect of early life exposure to air pollution on development of childhood asthma. Env. Health Perspect. 118, 284–290. De Marco, R., Locatelli, F., Sunyer, J., Burney, P., 2000. Differences in incidence of reported asthma related to age in men and women: a retrospective analysis of the data of the European Respiratory Health Survey. Am. J. Respir. Crit. Care Med. 162, 68–74. Deng, Q., Lu, C., Ou, C., Liu, W., 2015. Effects of early life exposure to outdoor air pollution and indoor renovation on childhood asthma in China. Build. Environ. 93, 84–91. Dong, G.H., Chen, T., Liu, M.M., Wang, D., Ma, Y.N., Ren, W.H., et al., 2011. Gender differences and effect of air pollution on asthma in children with and without allergic predisposition: northeast Chinese children health study. PLoS One 6 (7), e22470. Duijts, L., 2012. Fetal and infant origins of asthma. Eur. J. Epidemiol. 27, 5–14. Eder, W., Ege, M.J., von Mutius, E., 2006. The asthma epidemic. N Engl. J. Med. 355, 2226–2235. Gasana, J., Dillikar, D., Mendy, A., Forno, E., Vieira, E.R., 2012. Motor vehicle air pollution and asthma in children: a meta-analysis. Environ. Res. 117, 36–45. Gascon, M., Casas, M., Morales, E., Valvi, D., Ballesteros-Gómez, A., Luque, N., et al., 2015. Prenatal exposure to bisphenol A and phthalates and childhood respiratory tract infections and allergy. J. Allergy Clin. Immunol. 135, 370–378. Gauderman, W.J., Avol, E., Lurmann, F., Kuenzli, N., Gilliland, F., Peters, J., et al., 2005. Childhood asthma and exposure to traffic and nitrogen dioxide. Epidemiology 16, 737–743. Gehring, U., Cyrys, J., Sedlmeir, G., Brunekreef, B., Bellander, T., Fischer, P., et al., 2002. Traffic-related air pollution and respiratory health during the first 2 yrs of life. Eur. Respir. J. 19, 690–698. Gehring, U., Wijga, A.H., Brauer, M., Fischer, P., de Jongste, J.C., Kerkhof, M., et al., 2010. Traffic-related air pollution and the development of asthma and allergies during the first 8 years of life. Am. J. Respir. Crit. Care Med. 181, 596–603. Gluckman, P.D., Hanson, M.A., Cooper, C., Thornburg, K.L., 2008. Effect of in utero and early-life conditions on adult health and disease. N Engl. J. Med. 359, 61–73. Gould, H.J., Sutton, B.J., 2008. IgE in allergy and asthma today. Nat. Rev. Immunol. 8, 205–217. Gowers, A.M., Cullinan, P., Ayres, J.G., Anderson, H.R., Strachan, D.P., Holgate, S.T., et al., 2012. Does outdoor air pollution induce new cases of asthma? Biological plausibility and evidence: a review. Respirology 17, 887–898. Gruzieva, O., Bergström, A., Hulchiy, O., Kull, I., Lind, T., Melén, E., et al., 2013. Exposure to air pollution from traffic and childhood asthma until 12 years of age. Epidemiology 24, 54–61. Han, L., Zhou, W., Li, W., Li, L., 2014. Impact of urbanization level on urban air quality: a case of fine particles (PM2.5) in Chinese cities. Environ. Pollut. 194, 163–170.

92

Q. Deng et al. / Environmental Research 143 (2015) 83–92

Heinrich, J., 2011. Influence of indoor factors in dwellings on the development of childhood asthma. Int. J. Hyg. Environ. Health 214, 1–25. Hwang, B.F., Lee, Y.L., Lin, Y.C., Jaakkola, J.J.K., Guo, Y.L., 2005. Traffic related air pollution as a determinant of asthma among Taiwanese school children. Thorax 60, 467–473. Hertz–Picciotto, I., Baker, R.J., Yap, P.S., Dostál, M., Joad, J.P., Lipsett, M., et al., 2007. Early childhood lower respiratory illness and air pollution. Environ. Health Perspect. 115, 1510–1518. Islam, T., Gauderman, W.J., Berhane, K., McConnell, R., Avol, E., Peters, J.M., et al., 2007. Relationship between air pollution, lung function and asthma in adolescents. Thorax 62, 957–963. Jerrett, M., Shankardass, K., Berhane, K., Gauderman, W.J., Künzli, N., Avol, E., et al., 2008. Traffic-related air pollution and asthma onset in children: a prospective cohort study with individual exposure measurement. Environ. Health Perspect. 116, 1433–1438. Kan, H., Chen, R., Tong, S., 2012. Ambient air pollution, climate change, and population health in China. Environ. Int. 42, 10–19. Kasznia-Kocot, J., Kowalska, M., Górny, R.L., Niesler, A., Wypych-Ślusarska, A., 2010. Environmental risk factors for respiratory symptoms and childhood asthma. Ann. Agric. Environ. Med. 17, 221–229. Kim, K.H., Jahan, S.A., Kabir, E., 2013. A review on human health perspective of air pollution with respect to allergies and asthma. Environ. Int. 59, 41–52. Kozyrskyj, A.L., Bahreinian, S., Azad, M.B., 2011. Early life exposures: impact on asthma and allergic disease. Curr. Opin. Allergy Clin. Immunol. 11, 400–406. Lau, S., Illi, S., Sommerfeld, C., Niggemann, B., Bergmann, R., von Mutius, E., et al., 2000. Early exposure to house-dust mite and cat allergens and development of childhood asthma: a cohort study. Lancet 356, 1392–1397. Lee, J.T., Kim, H., Song, H., Hong, Y.C., Cho, Y.S., Shin, S.Y., et al., 2002. Air pollution and asthma among children in Seoul, Korea. Epidemiology 13, 481–484. Li, N., Hao, M., Phalen, R.F., Hinds, W.C., Nel, A.E., 2003. Particulate air pollutants and asthma: a paradigm for the role of oxidative stress in PM-induced adverse health effects. Clin. Immunol. 109, 250–265. London, S.J., 2007. Gene-air pollution interactions in asthma. Proc. Am. Thorac. Soc. 4, 217–220. Marshall, J.D., Nethery, E., Brauer, M., 2008. Within-urban variability in ambient air pollution: comparison of estimation methods. Atmos. Environ. 42, 1359–1369. Martino, D., Prescott, S., 2011. Epigenetics and prenatal influences on asthma and allergic airways disease. Chest 139, 640–647. McConnell, R., Islam, T., Shankardass, K., Jerrett, M., Lurmann, F., Gilliland, F., et al., 2010. Childhood incident asthma and traffic-related air pollution at home and school. Environ. Health Perspect. 118, 1021–1026. Miller, R.L., Ho, S.M., 2008. Environmental epigenetics and asthma: current concepts and call for studies. Am. J. Respir. Crit. Care Med. 177, 567–573. Mölter, A., Agius, R., de Vocht, F., Lindley, S., Gerrard, W., Custovic, A., et al., 2014. Effects of long-term exposure to PM10 and NO2 on asthma and wheeze in a prospective birth cohort. J. Epidemiol. Community Health 68, 21–28. Mölter, A., Simpson, A., Berdel, D., Brunekreef, B., Custovic, A., Cyrys, J., et al., 2015. A multicentre study of air pollution exposure and childhood asthma prevalence: the escape project. Eur. Respir. J. 45, 610–624. Morgenstern, V., Zutavern, A., Cyrys, J., Brockow, I., Gehring, U., Koletzko, S., et al., 2007. Respiratory health and individual estimated exposure to traffic-related air pollutants in a cohort of young children. Occup. Environ. Med. 64, 8–16. Morgenstern, V., Zutavern, A., Cyrys, J., Brockow, I., Koletzko, S., Kramer, U., et al., 2008. Atopic diseases, allergic sensitization, and exposure to traffic-related air pollution in children. Am. J. Respir. Crit. Care Med. 177, 1331–1337. Mortimer, K., Neugebauer, R., Lurmann, F., Alcorn, S., Balmes, J., Tager, I., 2008. Air pollution and pulmonary function in asthmatic children: effects of prenatal and

lifetime exposures. Epidemiology 19, 550–557. Nadeau, K., McDonald-Hyman, C., Noth, E.M., Pratt, B., Hammond, S.K., Balmes, J., et al., 2010. Ambient air pollution impairs regulatory T-cell function in asthma. J. Allergy Clin. Immunol. 126, 845–852. Nishimura, K.K., Galanter, J.M., Roth, L.A., Oh, S.S., Thakur, N., Nguyen, E.A., et al., 2013. Early-life air pollution and asthma risk in minority children. The GALA II and SAGE II studies. Am. J. Respir. Crit. Care Med. 188, 309–318. Nohr, E.A., Frydenberg, M., Henriksen, T.B., Olsen, J., 2006. Does low participation in cohort studies induce bias? Epidemiology 17, 413–418. Nordling, E., Berglind, N., Melén, E., Emenius, G., Hallberg, J., Nyberg, F., et al., 2008. Traffic-related air pollution and childhood respiratory symptoms, function and allergies. Epidemiology 19, 401–408. Patel, M.M., Miller, R.L., 2009. Air pollution and childhood asthma: recent advances and future directions. Curr. Opin. Pediatr. 21, 235–242. Peden, D.B., 1997. Mechanisms of pollution‐induced airway disease: in vivo studies. Allergy 52 (S38), 37–44. Pénard-Morand, C., Raherison, C., Charpin, D., Kopferschmitt, C., Lavaud, F., Caillaud, D., et al., 2010. Long-term exposure to close-proximity air pollution and asthma and allergies in urban children. Eur. Respir. J. 36, 33–40. Pikhart, H., Bobak, M., Gorynski, P., Wojtyniak, B., Danova, J., Celko, M.A., et al., 2001. Outdoor sulphur dioxide and respiratory symptoms in Czech and Polish school children: a small-area study (SAVIAH). Int. Arch. Occup. Environ. Health 74, 574–578. Salvi, S.S., Babu, K.S., Holgate, S.T., 2001. Is asthma really due to a polarized T cell response toward a helper T cell type 2 phenotype? Am. J. Respir. Crit. Care Med. 164, 1343–1346. Sarnat, J.A., Holguin, F., 2007. Asthma and air quality. Curr. Opin. Pulm. Med. 13, 63–66. Smargiassi, A., Kosatsky, T., Hicks, J., Plante, C., Armstrong, B., Villeneuve, P.J., et al., 2009. Risk of asthmatic episodes in children exposed to sulfur dioxide stack emissions from a refinery point source in Montreal, Canada. Environ. Health Perspect. 117, 653–659. Studnicka, M., Hackl, E., Pischinger, J., Fangmeyer, C., Haschke, N., Kuhr, J., et al., 1997. Traffic-related NO2 and the prevalence of asthma and respiratory symptoms in seven year olds. Eur. Respir. J. 10, 2275–2278. Takano, H., Yoshikawa, T., Ichinose, T., Miyabara, Y., Imaoka, K., Sagai, M., 1997. Diesel exhaust particles enhance antigen-induced airway inflammation and local cytokine expression in mice. Am. J. Respir. Crit. Care Med. 156, 36–42. Tepper, R.S., Morgan, W.J., Cota, K., Wright, A., Taussig, L.M., Pediatricians, G., 1986. Physiologic growth and development of the lung during the first year of life. Am. Rev. Respir. Dis. 134, 513–519. Trasande, L., Thurston, G.D., 2005. The role of air pollution in asthma and other pediatric morbidities. J. Allergy Clin. Immunol. 115, 689–699. Watts, J., 2006. Doctors blame air pollution for China's asthma increases. Lancet 368, 719–720. Warner, J.O., 2004. The early life origins of asthma and related allergic disorders. Arch. Dis. Child. 89, 97–102. Zhang, J.J., Hu, W., Wei, F., Wu, G., Korn, L.R., Chapman, R.S., 2002. Children's respiratory morbidity prevalence in relation to air pollution in four Chinese cities. Environ Health Perspect. 110, 961–967. Zhang, Y.P., Li, B.Z., Huang, C., Yang, X., Qian, H., Deng, Q.H., et al., 2013. Ten cities cross-sectional questionnaire survey of children asthma and other allergies in China. Chin. Sci. Bull. 58, 4182–4189. Zmirou, D., Gauvin, S., Pin, I., Momas, I., Sahraoui, F., Just, J., et al., 2004. Traffic related air pollution and incidence of childhood asthma: results of the Vesta case-control study. J. Epidemiol. Community Heal. 58, 18–23.