International Journal of
Environmental Research
and Public Health
Article
Air Pollution and Dispensed Medications for Asthma,
and Possible Effect Modifiers Related to Mental
Health and Socio-Economy: A Longitudinal Cohort
Study of Swedish Children and Adolescents
Anna Oudin *, Lennart Bråbäck, Daniel Oudin Åström and Bertil Forsberg
Occupational and Environmental Medicine, Umeå University, 90187 Umeå, Sweden;
Lennart.Braback@umu.se (L.B.); Daniel.oudin_astrom@med.lu.se (D.O.Å.); bertil.forsberg@umu.se (B.F.)
* Correspondence: anna.oudin@umu.se
Received: 22 September 2017; Accepted: 11 November 2017; Published: 16 November 2017
Abstract: It has been suggested that children that are exposed to a stressful environment at home
have an increased susceptibility for air pollution-related asthma. The aim here was to investigate the
association between air pollution exposure and asthma, and effect modification by mental health and
by socio-economic status (as markers of a stressful environment). All individuals under 18 years of
age in four Swedish counties during 2007 to 2010 (1.2 million people) were included. The outcome
was defined as dispensing at least two asthma medications during follow up. We linked data on
NO2 from an empirical land use regression to data from national registers on outcome and potential
confounders. Data was analyzed with logistic regression. There was an odds ratio (OR) of 1.02
(95% Confidence Interval (CI: 1.01–1.03) for asthma associated with a 10 µg·m−3 increase in NO2 .
The association only seemed to be present in areas where NO2 was higher than 15 µg·m−3 with an
OR of 1.09 (95% CI: 1.07–1.12), and the association seemed stronger in children with parents with a
high education, OR = 1.05 (95% CI: 1.02–1.09) and OR = 1.04 (95% CI: 1.01–1.07) in children to mothers
and father with a high education, respectively. The association did not seem to depend on medication
history of psychiatric disorders. There was weak evidence for the association between air pollution
and asthma to be stronger in neighborhoods with higher education levels. In conclusion, air pollution
was associated with dispensed asthma medications, especially in areas with comparatively higher
levels of air pollution, and in children to parents with high education. We did not observe support
for our hypothesis that stressors linked to socio-economy or mental health problems would increase
susceptibility to the effects of air pollution on the development of asthma.
Keywords: asthma; childhood asthma; air pollution; stress; socio-economy; mental health
1. Introduction
Asthma is one of the most common chronic diseases in children. Despite decades of intense
research, the etiology and pathogenesis are still partly unknown, although environmental exposures
in infancy [1] and stress are of importance [1–7]. Negative life events can cause asthma attacks in
children, especially if they are exposed to chronic stress [8]. Furthermore, there is a strong association
between respiratory symptoms and psychological status [9], and parents’ mental health is associated
with childhood asthma in the offspring [10]. The causal relationship between asthma and mental
health/stress seem to point in both directions because asthma is strongly related to perceived life
quality but can also yield anxiety and fatigue. Childhood asthma has received a lot of attention as a
cause for mental health problems as it, for example, is a risk factor for neuropsychiatric disease [11].
Long-term exposure to air pollution seem to be a risk factor for asthma [2,12], reduced lung
function [13], and sensitization [14], although some large studies have been negative [15]. Interestingly,
Int. J. Environ. Res. Public Health 2017, 14, 1392; doi:10.3390/ijerph14111392
www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2017, 14, 1392
2 of 12
it has been suggested that children that are exposed to a stressful environment at home have an
increased susceptibility for air pollution-related asthma [16,17].
Physical health and mental health are thus intertwined, but air pollution can also have a direct
negative effect on the brain, for example cognitive development [18,19], cognitive decline and
dementia [20–22]. Experimental studies show that an association between air pollution exposure
and mental health is plausible [23]. Anxiety disorders and schizophrenia are more common in
urban areas [24,25], which can be attributed to urban social environment, but possibly thus also to
environmental factors. We recently showed that air pollution was associated with medications for
psychiatric disorders in Swedish children and adolescents [26], and there is support for air pollution
exposure during fetal life to increase risk of autism [27,28], and for air pollution to be linked to
behavioral problems in children [29,30].
In summary, the relationship between asthma, mental health, stress, socio-economy and air
pollution is likely highly complex (Figure 1). Our hypothesis is that mental health and socio-economy
modify the association between air pollution and pediatric asthma, with the theory that stressors linked
to socio-economy or mental health problems directly lead to deteriorated health, that susceptibility for
air pollution is increased through alterations in the immune system or other biological systems [31],
that city dwellers react differently to stress than persons residing in rural areas [32], or that air pollution
cause affective responses or impair cognition.
Figure 1. Possible causal pathways between exposure to air pollution, mental health, socioeconomic
status and asthma.
The aim of the present study was to investigate the association between air pollution exposure
and dispensed medications for asthma, and if that association was modified by mental health or by
socio-economic status, in a large cohort of Swedish children and adolescents, to somewhat disentangle
this complex relationship.
2. Materials and Methods
2.1. Study Area and Cohort
The study area and cohort have been described in detail elsewhere [26], but briefly, we used
a prospective cohort design where all individuals under 18 years of age who at any time during
1 January 2007–31 December 2010 had a registered residential address in any of the four Swedish
counties of Stockholm, Västra Götaland, Skåne and Västerbotten (Figure 2).
Int. J. Environ. Res. Public Health 2017, 14, 1392
3 of 12
Figure 2. Concentrations of NO2 modelled by the Swedish Research Institute in the four counties of
the study area.
The four counties encompass more than half the Swedish population and are heterogeneous
in terms of geographical location, population size and population density but also with respect to
migration, socioeconomic characteristics, and urbanization and air pollution concentrations.
2.2. National Register Data
We used data from the Umeå SIMSAM lab [33]. Data in the lab comes from national registers
(which covers the entire Swedish population) and local registers, which were combined via the
personal identification number for the use of research. For the present study, we used data from
the Swedish National Board of Health and Welfare on a group of dispensed medication related to
asthma, both long acting and rescue medications, namely medications starting with the following
Int. J. Environ. Res. Public Health 2017, 14, 1392
4 of 12
Swedish ATC codes: R03AC (selective beta-2 stimulating medications), R03AK (adrenergic and other
medications for obstructive airway disease), R03BA (glucocorticoids), R03BC (anti allergic medications,
except corticosteroids), R03CC (selective beta-2-stimulating medications) and R03DC (leukotriene
receptor antagonists). We defined the outcome as dispensing at least two dispensed medications
with any of these codes. By defining the outcome as at least two dispensed asthma medications
during follow-up, our aim was to capture patients more likely to have asthma than patients who only
dispensed asthma medications once. Many children with respiratory symptoms, especially very young
children, are prescribed asthma medications to evaluate if their symptoms decrease by medication.
We thus wanted to avoid to classify children with unspecified respiratory symptoms as asthmatics.
We also used data on dispensed medications for psychiatric disorders, namely medications with an
ATC-code starting with N05 and N06, hereafter referred to as N05 and N06. N05 consists of neuroleptics
(antipsychotic medications), ataractics and sleeping pills (a broad group of sedative medications
including hydroxyzine and melatonin-based medications). N06 consist mainly of anti-depressants and
ADHD medications when prescribed to children. We defined two variables as dispensed N05 or N06
during baseline (when follow-up started). From the medical birth registry we used information on
maternal body mass index in early pregnancy (continuous variable) and maternal smoking during
early pregnancy (three categories) Furthermore, from Statistics Sweden we used data on age at the
start of follow-up (continuous variable), sex, maternal and paternal education level (four categories) at
start follow-up and yearly data on parental unemployment (yes/no). Based on data from Statistics
Sweden we defined a group-level (neighborhood) variable on socioeconomic status on Small Areas for
Market Statistics (SAMS), namely the proportion of the population in the SAMS area with three or
more years of undergraduate studies in the age category 25–65 years. SAMS are supposed to represent
homogeneous neighborhoods and there are 6016 SAMS areas in our study area. The quartiles of the
variable was <14%, 14% < 20%, 20% < 32% and ≥32% with three or more years of undergraduate
studies in the age category 25–65 years.
2.3. Air Pollution Exposure Assessment
The Swedish environmental research institute has developed an empirical Land Use Regression
model to estimate the urban contribution of NO2 added to the regional background level. The model is
based on the ratio of the urban content contribution, the meteorological parameters and the population
distribution. It takes into account that the levels are not evenly distributed across a city but related
to population density by including the spatial distribution of the urban contribution [34]. The base
year of the land use regression model was 2010 and the spatial resolution was 1 km2 . The model has
previously been showed to have fairly good accordance with a dispersion mode [35].
This model was used together with a model for the regional background levels built on monitoring
data to calculate exposure estimates of NO2 the entire Swedish population, for the year when the
individual was included in the study (any year between 2005 and 2010). We used this measure as a
marker of long-term exposure to air pollution with the underlying assumptions that the concentration
at study inclusion was a valid marker for long-term exposure, and that the spatial contrasts in exposure
during follow-up were fairly constant. The study was approved by the regional ethics board in Umeå
(Dnr 2010-157-31).
2.4. Statistical Analysis
The data was analyzed with logistic regression. The results are presented as Odds ratios (ORs) and
their 95% CIs. We studied NO2 as a continuous variables, and present ORs associated with pollutant
increases of 10 µg·m−3 . The initial cohort size was 1,294,290, but we excluded study persons with
any dispensed asthma medication during 2005 or 2006 to somewhat capturing incident cases, and the
cohort size was reduced to 1,215,754 individuals (Figure 3). It should be noted that we had no records
of dispensed medications before 2005, and therefore children denoted incident cases in our study may
be not be true new cases, but only new cases since 2005. Individuals with missing data on any of
Int. J. Environ. Res. Public Health 2017, 14, 1392
5 of 12
the variables included in the main adjusted model (age, sex, parental education, smoking and BMI
during early pregnancy) were excluded from the main analysis. The total size of the complete cohort
was 745,171 individuals. We used an open cohort approach. Individuals who moved into, or out of,
the study areas during follow-up were thus included or excluded in the cohort on December 31 (due to
register-technicalities) of the year they moved in or out. Children who were born or died during
follow-up were included/excluded at time of birth/death.
Figure 3. Study flow-chart.
In an additional analysis, we adjusted the models for maternal and paternal income.
Further, we stratified our analyses on parental education, previous dispensed N05 or N06,
parental unemployment, and area SES. We also ran age-specific analyses for the age at baseline groups
0 < 5, 5 < 10, 10 < 15 and 15–18. We evaluated the presence of heterogeneity in the county-specific
results by including a cross-product term in the equation. Otherwise we refrained from calculating
p-values for effect modification since the size of the cohort would mean that most p-values were highly
statistically significant and instead assessed effect modification by stratifying analysis and visually
inspecting of the association estimates.
2.5. Sensitivity Analyses
To check if the associations was present in low-level areas, we ran a threshold analysis where
we stratified data on NO2 levels of 15 µg·m3 . Since asthma medications are prescribe very frequently
to small children we excluded children below the age of two in a sensitivity analysis. In another
analysis, we delayed start of follow-up one year, to 1 January 2008. We also restricted the analysis to
those children and adolescents who resided in the same address for at least two year from start of
follow-up. Finally, we investigated if patterns of missingness among our variables was differential
with respect to the outcome or level of exposure. We imputed missing observations in the covariates
using a Markov Chain Monte Carlo approach and reran the main analyses to investigate whether our
estimates changed. We also calculated a pooled estimate from the county-specific estimates. The main
analysis was run also with Cox regression, and mixed logistic regression to take into account the
multilevel nature of data (county-level). SAS V.9.2 (SAS, Stockhom, Schweden) software was used to
create data sets and run the analyses.
Int. J. Environ. Res. Public Health 2017, 14, 1392
6 of 12
3. Results
None of the background factors had a clear univariate association with medications for asthma,
except age, with the youngest children as expected being medicated to a higher extent (Table 1).
Furthermore, in the univariate analysis there were no marked heterogeneity in NO2 with respect to
any of the factors in the study (Table 1).
Table 1. Descriptive data for variables used in the study, stratified by asthma medication and mean
levels of NO2 (µg·m−3 ).
Variables
Categories
Asthma Medication *
Mean NO2
No
N
Yes
N (%)
Boys
566,571
50,809 (8)
10.2
Girls
558,711
39,663 (7)
10.3
<2
303,033
56,014 (16)
11.4
2<5
5 < 10
10 < 15
15–18
134,476
229,546
272,821
185,406
7466 (5)
11,037 (5)
12,576 (4)
3379 (2)
10.0
9.7
9.6
9.8
Missing
184,807
11,722 (6)
11.1
No
Yes <10 cig per day
Yes ≥10 cig per day
813,138
83,406
43,931
69,803 (8)
6262 (7)
2685 (6)
10.1
9.9
9.8
Age at baseline
Smoking early pregnancy
Mother’s education
Father’s education
(µg·m−3 )
Missing
95,790
6723 (7)
12.1
Elementary school
Only upper secondary school or
Post-secondary education < 2 years
Post-secondary education 2 ≤ 4 years
Post-secondary education ≥ 4 years
124,997
10,344 (8)
11.1
453,873
36,871 (8)
9.4
336,162
114,550
27,173 (7)
9361 (8)
10.3
11.1
Missing
109,437
7320 (6)
12.0
Elementary school
Only upper secondary school or
Post-secondary education < 2 years
Post-secondary education 2 ≤ 4 years
Post-secondary education ≥ 4 years
148,970
11,815 (7)
10.2
477,250
40,059 (8)
9.4
258,744
130,881
21,194 (8)
10,084 (7)
10.7
11.2
Group level education **
quartile 1 (<14%)
296,380
23,106 (7)
Group level education **
quartile 2 (14–<20%)
262,997
20,879 (7)
Group level education **
quartile 3 (20% < 32%)
292,250
24,230 (8)
9.0
Group level education **
quartile 4 (≥32%)
273,655
22,257 (8)
14.2
Mother unemployed at baseline
Father unemployed at baseline
Analeptika (N06) at baseline
Neuroleptika (N05) at baseline
8.9
No
1,033,154
85,288 (7)
10.2
Yes
92,128
7884 (8)
10.5
No
1,067,643
85,544 (7)
10.2
Yes
57,639
4928 (8)
11.5
10.3
No
1,115,524
90,028 (7)
Yes
9758
444 (4)
9.5
No
1,117,444
89,894 (7)
10.3
Yes
7838
578 (7)
Mean
10.1
Spearman rho
BMI early pregnancy
23.9
24.5
−0.04
Mother’s income baseline (SEK)
162,600
151,200
−0.03
Father’s income baseline (SEK)
280,000
267,800
−0.06
NO2 baseline (µg·m−3 )
10.2
10.5
1
* Defined as dispensing at least two medications with the ATC-codes R03AC, R03AK, R03BA, R03BC, R03CC and
R03DC during the study period (2007–2010); ** Proportion in the neighborhood (SAMS area) with three or more
years of undergraduate studies in the age category 25–65 years.
However, in young children (<2 years at study entry), there was an association between smoking
during early pregnancy and asthma (data not shown). There was an association between medications
Int. J. Environ. Res. Public Health 2017, 14, 1392
7 of 12
for asthma and NO2 when adjusted for age with an Odds ratio (OR) of 1.05 (95% Confidence Interval
(CI: 1.04–1.06) associated with a 10 µg·m−3 increase in NO2 (Table 2).
When adjusted for sex, parental education, smoking during pregnancy and BMI in early
pregnancy, the OR was attenuated to 1.02 (95% CI: 1.01–1.03; Table 2). The estimate that was
pooled from the county-specific estimates was 1.06 (95% CI: 1.02–1.11). The analysis where data
was stratified on NO2 above and below 15 (µg·m−3 ) indicates that the association between NO2 and
dispensed asthma medication only seem to be present in areas with higher levels of air pollution,
above 15 (µg·m−3 ) in this study (Table 2). When excluding children under the age of two at study entry,
the OR in the adjusted analysis was 1.06 (95% CI: 1.04–1.09; Table 2), suggesting that the association
between air pollution and asthma medications was stronger after early childhood.
Table 2. Odds ratios (ORs) and their 95% Confidence Intervals (CIs) for asthma medication in
association with a 10 (µg·m−3 ) increase in NO2 for all cohort members, and in sub-groups of the cohort.
OR
95% CI
Adjusted 4
Age-Adjusted
All
1.05
1.04–1.06
1.02
1.01–1.03
Mother low education 1
1.01
0.98–1.05
0.99
(0.95–1.02)
Mother high education 2
1.04
1.01–1.06
1.05
1.02–1.09
Low father education
1
1.03
0.997–1.06
0.99
0.96–1.027
Father high education
2
Sub–groups
1.03
1.00–1.05
1.04
1.01–1.07
Unemployment mother baseline
1.03
1.00–1.07
1.00
0.97–1.04
Unemployment father baseline
1.06
1.02–1.10
1.02
0.97–1.06
<2
2≤5
5 ≤ 10
10 ≤ 15
15–18
1.06
1.05
1.12
1.14
1.10
1.04–1.07
1.01–1.08
1.08–1.15
1.11–1.17
1.04–1.15
1.02
1.03
1.07
1.07
1.00
1.01–1.04
0.99–1.08
1.03–1.11
1.03–1.11
0.83–1.20
Dispensed neuroleptika (N05) baseline year
1.07
0.95–1.21
1.03
0.87–1.21
Dispensed analeptika (N06) baseline year
1.05
0.92–1.21
1.12
0.90–1.40
Age baseline
3
1.01
0.99–1.03
0.93
0.90–0.96
Neighborhoodl education quartile 2 (14–<20%) 3
0.96
0.94–0.99
0.92
(0.89–0.96)
Neighborhood education quartile 3 (20% < 32%) 3
1.04
1.01–1.07
1.01
0.98–1.04
1.05
1.03–1.06
1.05
1.03–1.07
1.06
1.04–1.07
1.02
1.01–1.04
Neighborhoodeducation quartile 1 (<14%)
Neighborhood education quartile 4 (≥32%)
Children < 2 years of age
1
3
Children ≥ 2 years of age
1.11
1.09–1.13
1.06
1.04–1.09
Stockholm
1.04
1.02–1.06
1.02
1.00–1.04
Skåne
1.09
1.07–1.12
1.07
1.05–1.10
Västra Götaland
1.02
1.01–1.04
0.99
0.98–1.01
Västerbotten
1.04
0.98–1.11
1.02
0.95–1.10
NO2 <15
(µg·m−3 )
0.97
0.95–0.99
0.93
0.91–0.95
NO2 ≥15
(µg·m−3 )
1.11
1.09–1.14
1.09
1.07–1.12
2
3
Elementary school only; Post-secondary education ≥ 4 years; The proportion of the population in the
neighborhood (SAMS-area) with three or more years of undergraduate studies in the age category 25–65 years;
4 Adjusted for age at baseline, sex, four categories of parental (maternal and paternal) education and for three
categories of smoking and a continuous measure of BMI during early pregnancy.
The estimates were not very precise (wide confidence intervals), but there was no clear evidence
for medication of psychiatric disorders to influence the association between air pollution and asthma
(Table 2). The association between air pollution and asthma medications seemed stronger in children
Int. J. Environ. Res. Public Health 2017, 14, 1392
8 of 12
with parents with high education than in children with parents with low education (Table 2). There was
weak evidence for levels of socio-economy (education) in the neighborhood to have an impact on
the association between NO2 and dispensed asthma medication, with an adjusted OR in areas with
highest quartile of education level of 1.05 (95% CI: 1.03–1.07). In the first three quartiles, there was no
evidence for an association, with ORs close to 1 (Table 2). From Table 1, it is evident that the levels
of air pollution were higher in area with the highest quartile of education level, so these results are
consistent with the analysis showing an association mainly where NO2 concentrations are higher.
The dispersion was also highest in quartile 4, with a standard deviation of 11 µg·m−3 , in quartile
1–3, 6.3, 5.3 and 5.1 µg·m−3 . The delay of start of follow-up and restricting the analysis to those who
resided in the same address at least two years after follow-up did only marginally affect the estimates
(data not shown). Adjusting for parental income only marginally affected the estimates (data not
shown). There was some evidence of heterogeneity between counties (p for effect modification = 0.004),
but the results with mixed logistic regression with county as a level was similar to the ordinary logistic
regression. The results with Cox regression was similar to the results generated with logistic regression
(data not shown).
4. Discussion
In this large prospective cohort consisting of more than half of all Swedish children and
adolescents, we observed evidence for air pollution to be associated with dispensed medications
for asthma, especially in areas with comparatively higher levels of air pollution. We observed no clear
evidence for our hypothesis, that stressors linked to low socio-economy or mental health problems
would increase susceptibility to the effects of air pollution on the development of asthma On the
contrary, the association between air pollution and asthma seemed stronger in children to parents with
high education than with low education.
Our hypothesis derived from two earlier studies, where children exposed to a stressful
environment at home seemed to have an increased susceptibility for air pollution-related asthma [16,17].
There are several potential explanations for the differences in findings that should be discussed. First of
all, the two previous studies were an interview study of 73 children with asthma [16], and a prospective
cohort study of 2497 children based on questionnaires [17], whereas we used data from national
registers on a cohort of more than half of all Swedish children and adolescents. Using data from national
registers has a major advantages in its longitudinal design and the high-quality data, for example
selection bias is not a problem. However, the other two studies used more exact measures of stress,
namely interviews of life stress [16] and parental stress from a perceived stress scale [17]. We used
data on socio-economy and mental health, which although they are very crude indicators of stress,
are relevant potential effect modifiers nevertheless. Furthermore, life-style factors may not have been
fully accounted for in the analysis since we were restricted to variables that were present in the registers,
which could have resulted in bias residual confounding. The potential confounders we had data on did
not seem to have strong associations with neither outcome nor exposure however, although the effect
estimates were somewhat attenuated when adjusting for them (Tables 1 and 2). The estimates were
quite stable when adjusting for parental income and group-level socio-economy, which may indicate
that residual confounding due to socio-economy is not likely. Secondly, another difference compared
to the two previous studies was how we defined the outcome; namely as dispensing at least two
asthma medications during follow-up, and we did not use information on diagnosis (doctor-diagnosed
new onset asthma during 3 years of follow-up in one study [17] and biologic and clinical outcomes
in children with asthma in the other study [16]). Medication use to describe health is increasingly
used in the Nordic countries, where national registers provide the opportunity to do so [26,36,37].
Socioeconomic status could influence the probability to dispense medication, but that would also be
true if using diagnosis as an outcome. It is well known that health care seeking behavior is highly
dependent on socioeconomic status, also in Sweden which is one of the most equal countries in the
world with a Gini coefficient of around 0.30 [38]. However, health care is free for every child younger
Int. J. Environ. Res. Public Health 2017, 14, 1392
9 of 12
than 18 years of age in Sweden, and medications are heavily subsidized example with respect to
income differences. Our results may thus not be generalized to, or compare well with, countries with
less access to welfare and health care. Thirdly, our study was conducted in Sweden, where air pollution
are generally quite low (the two other studies were done in California [17] and Vancouver [16]), and we
cannot rule out that mental health or socio-economic status would modify the association between air
pollution and asthma in areas with higher levels of air pollution.
In the present study we used modeled levels of NO2 , a marker of traffic-related air pollution,
as exposure measure. Exposure misclassification must be considered in any study on long-term
exposure to air pollution. We used exposure data from 2010 and assumed that differences (contrasts)
in exposure would be similar back in time over the follow-up period (2005–2010), which is probably
a reasonable assumption given the comparable short time period. An obvious source of exposure
misclassification however is the fact that ambient exposure was modeled outdoors at the home address
and that exposure at school, day-care or indoor exposure were not taken into account. We know from
previous work that actual exposure may correlate only mildly with modelled outdoor exposure [39].
Furthermore, we used data on air pollution and start of follow-up as a marker for long-term exposure
to air pollution. The validity of that assumption could of course be questioned, but we used a similar
approach in the European Study of Cohorts for Air Pollution Effects (ESCAPE), and for that study
we showed that when people changed residential address the air pollution concentrations were often
similar at the new and old address. Exposure misclassification could theoretically bias the estimates
both away and towards the null, if the misclassification is differential with respect to the outcome.
Such misclassification is easy to imagine, for example if cohort members with respiratory symptoms
and low socio-economy stay indoors more than cohort members with respiratory symptoms and high
socio-economy, but it is difficult to speculate in size and direction of bias from such differential exposure
misclassification. We have used the same approach as in many other air pollution epidemiology studies
where we have been able to observe associations. For example, we observed strong associations
between dispensed medications for psychiatric disorders and air pollution using almost exactly the
same cohort and exposure measure as in the present study [26]. We believe, therefore, that exposure
misclassification is an unlikely explanation for the results of the present study, at least if any true
causal effect modification from any of the variables investigated on the association between traffic air
pollution and asthma would be strong.
Furthermore, the associations cannot be explained by heterogeneity in prevalence or exposure
across Sweden, as the pooled estimate (from county-specific estimates) seemed slightly higher than
the non-pooled estimate. It is somewhat surprising that smoking during pregnancy did not seem to be
clearly associated with asthma in the children in our study, since smoking during pregnancy seem
to be a risk factor for asthma in children and adolescent [40,41], particularly for asthma or asthmatic
symptoms during the first years of life [42–44]. An association between smoking during pregnancy
and asthma in young children is supported in our data, where there was a univariate association
between smoking during early pregnancy and the outcome, but only in children which were very
young (<2 years at study entry).
5. Conclusions
In conclusion, we observed associations between dispensed asthma medications and levels
of air pollution at the home address, in areas where levels were comparatively high (NO2 annual
mean ≥15 (µg·m−3 ). We observed evidence for the association to be stronger in children to parents
with high education, but we did not observe support for our hypothesis that stressors linked to
socio-economy or mental health problems would increase susceptibility to the effects of air pollution
on the development of asthma.
Acknowledgments: The Umeå SIMSAM Laboratory data infrastructure used in this study was developed with
support from the Swedish Research Council (2008-7491) and with strategic support from Umeå University.
The work was funded by Vårdalstiftelsen with the Dnr VÅ 2011-25/430 (AO).
Int. J. Environ. Res. Public Health 2017, 14, 1392
10 of 12
Author Contributions: All authors contributed to the conception or design of the work, or the acquisition, analysis,
or interpretation of data for the work, and drafted the work or revised it critically for important intellectual
content and gave their final approval of the version to be published and agreed to be accountable for all aspects of
the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately
investigated and resolved. Anna Oudin, Bertil Forsberg and Lennart Bråbäck conceived and designed the study.
Anna Oudin acquired and analysed the data and drafted the manuscript. Daniel Oudin Åström helped interpret
the data for the work and with the statistical analysis of the data.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Bråbäck, L.; Vogt, H.; Hjern, A. Migration and asthma medication in international adoptees and immigrant
families in Sweden. Clin. Exp. Allergy 2011, 41, 1108–1115. [CrossRef] [PubMed]
Bråbäck, L.; Forsberg, B. Does traffic exhaust contribute to the development of asthma and allergic
sensitization in children: Findings from recent cohort studies. Environ. Health 2009, 8, 17. [CrossRef]
[PubMed]
Kozyrskyj, A.L.; Bahreinian, S.; Azad, M.B. Early life exposures: Impact on asthma and allergic disease.
Curr. Opin. Allergy Clin. Immunol. 2011, 11, 400–406. [CrossRef] [PubMed]
Chen, E.; Fisher, E.B.; Bacharier, L.B.; Strunk, R.C. Socioeconomic Status, Stress, and Immune Markers in
Adolescents with Asthma. Psychosom. Med. 2003, 65, 984–992. [CrossRef] [PubMed]
Chen, E.; Hanson, M.D.; Paterson, L.Q.; Griffin, M.J.; Walker, H.A.; Miller, G.E. Socioeconomic status and
inflammatory processes in childhood asthma: The role of psychological stress. J. Allergy Clin. Immunol. 2006,
117, 1014–1020. [CrossRef] [PubMed]
Douwes, J.; Brooks, C.; Pearce, N. Stress and asthma: Hippocrates revisited. J. Epidemiol. Commun. Health
2010, 64, 561–562. [CrossRef] [PubMed]
Gregory, E.M.; Edith, C. Life Stress and Diminished Expression of Genes Encoding Glucocorticoid Receptor
and β2 -Adrenergic Receptor in Children with Asthma. Proc. Natl. Acad. Sci. USA 2006, 103, 5496–5501.
Sandberg, S.; Paton, J.Y.; Ahola, S.; McCann, D.C.; McGuinness, D.; Hillary, C.R.; Oja, H. The role of acute
and chronic stress in asthma attacks in children. Lancet 2000, 356, 982–987. [CrossRef]
Leander, M.; Lampa, E.; Rask-Andersen, A.; Franklin, K.; Gislason, T.; Oudin, A.; Svanes, C.; Torén, K.;
Janson, C. Impact of anxiety and depression on respiratory symptoms. Respir. Med. 2014, 108, 1594–1600.
[CrossRef] [PubMed]
Weil, C.M.; Wade, S.L.; Bauman, L.J.; Lynn, H.; Mitchell, H.; Lavigne, J. The relationship between psychosocial
factors and asthma morbidity in inner-city children with asthma. Pediatrics 1999, 104, 1274–1281. [CrossRef]
[PubMed]
Mogensen, N.; Larsson, H.; Lundholm, C.; Almqvist, C. Association between childhood asthma and ADHD
symptoms in adolescence—A prospective population-based twin study. Eur. J. Allergy Clin. Immunol. 2011,
66, 1224–1230. [CrossRef] [PubMed]
Bowatte, G.; Lodge, C.; Lowe, A.; Erbas, B.; Perret, J.; Abramson, M.; Matheson, M.; Dharmage, S.
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 2015, 70, 245–256. [CrossRef]
[PubMed]
Gehring, U.; Gruzieva, O.; Agius, R.M.; Beelen, R.; Custovic, A.; Cyrys, J.; Eeftens, M.; Flexeder, C.;
Fuertes, E.; Heinrich, J.; et al. Air pollution exposure and lung function in children: The ESCAPE project.
Environ. Health Perspect. 2013, 121, 1357. [CrossRef] [PubMed]
Nordling, E.; Berglind, N.; Melén, E.; Emenius, G.; Hallberg, J.; Nyberg, F.; Pershagen, G.; Svartengren, M.;
Wickman, M.; Bellander, T. Traffic-related air pollution and childhood respiratory symptoms, function and
allergies. Epidemiology 2008, 19, 401–408. [CrossRef] [PubMed]
Mölter, A.; Simpson, A.; Berdel, D.; Brunekreef, B.; Custovic, A.; Cyrys, J.; de Jongste, J.; De Vocht, F.;
Fuertes, E.; Gehring, U.; et al. A multicentre study of air pollution exposure and childhood asthma prevalence:
The ESCAPE project. Eur. Respir. J. 2015, 45, 610–624.
Chen, E.; Schreier, H.; Strunk, R.; Brauer, M. Chronic traffic-related air pollution and stress interact to predict
biologic and clinical outcomes in asthma. EHP 2008, 116, 970–975. [CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2017, 14, 1392
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
11 of 12
Shankardass, K.; McConnell, R.; Jerrett, M.; Milam, J.; Richardson, J.; Berhane, K. Parental stress increases
the effect of traffic-related air pollution on childhood asthma incidence. Proc. Natl. Acad. Sci. USA 2009, 106,
12406–12411. [CrossRef] [PubMed]
Guxens, M.; Sunyer, J. A review of epidemiological studies on neuropsychological effects of air pollution.
Swiss Med. Wkly. 2012, 141, w13322. [PubMed]
Sunyer, J.; Esnaola, M.; Alvarez-Pedrerol, M.; Forns, J.; Rivas, I.; López-Vicente, M.; Suades-González, E.;
Foraster, M.; Garcia-Esteban, R.; Basagaña, X.; et al. Association between traffic-related air pollution in
schools and cognitive development in primary school children: A prospective cohort study. PLoS Med. 2015,
12, e1001792. [CrossRef] [PubMed]
Chen, H.; Kwong, J.C.; Copes, R.; Tu, K.; Villeneuve, P.J.; van Donkelaar, A.; Hystad, P.; Martin, R.V.;
Murray, B.J.; Jessiman, B.; et al. Living near major roads and the incidence of dementia, Parkinson’s disease,
and multiple sclerosis: A population-based cohort study. Lancet 2017, 389, 718–726. [CrossRef]
Oudin, A.; Forsberg, B.; Adolfsson, A.N.; Lind, N.; Modig, L.; Nordin, M.; Nordin, S.; Adolfsson, R.;
Nilsson, L.-G. Traffic-related air pollution and dementia incidence in Northern Sweden: A longitudinal
study. Environ. Health Perspect. 2016, 124, 306. [CrossRef] [PubMed]
Power, M.C.; Adar, S.D.; Yanosky, J.D.; Weuve, J. Exposure to air pollution as a potential contributor to
cognitive function, cognitive decline, brain imaging, and dementia: A systematic review of epidemiologic
research. Neurotoxicology 2016, 56, 235–253. [CrossRef] [PubMed]
Genc, S.; Zadeoglulari, Z.; Fuss, S.H.; Genc, K. The adverse effects of air pollution on the nervous system.
J. Toxicol. 2012, 2012. [CrossRef] [PubMed]
Peen, J.; Schoevers, R.A.; Beekman, A.T.; Dekker, J. The current status of urban-rural differences in psychiatric
disorders. Acta Psychiatr. Scand. 2010, 121, 84–93. [CrossRef] [PubMed]
Krabbendam, L.; van Os, J. Schizophrenia and Urbanicity: A Major Environmental Influence—Conditional
on Genetic Risk. Schizophr. Bull. 2005, 31, 795–799. [CrossRef] [PubMed]
Oudin, A.; Bråbäck, L.; Åström, D.O.; Strömgren, M.; Forsberg, B. Association between neighbourhood air
pollution concentrations and dispensed medication for psychiatric disorders in a large longitudinal cohort
of Swedish children and adolescents. BMJ Open 2016, 6, e010004. [CrossRef] [PubMed]
Becerra, T.A.; Wilhelm, M.; Olsen, J.; Cockburn, M.; Ritz, B. Ambient air pollution and autism in Los Angeles
county, California. Environ. Health Perspect. 2013, 121, 380–386. [CrossRef] [PubMed]
Volk, H.E.; Lurmann, F.; Penfold, B.; Hertz-Picciotto, I.; McConnell, R. Traffic-related air pollution, particulate
matter, and autism. JAMA Psychiatry 2013, 70, 71–77. [CrossRef] [PubMed]
Newman, N.C.; Ryan, P.; LeMasters, G.; Levin, L.; Bernstein, D.; Hershey, G.K.K.; Lockey, J.E.; Villareal, M.;
Reponen, T.; Grinshpun, S.; et al. Traffic-related air pollution exposure in the first year of life and behavioral
scores at 7 years of age. Environ. Health Perspect. 2013, 121, 731. [CrossRef] [PubMed]
Forns, J.; Dadvand, P.; Foraster, M.; Alvarez-Pedrerol, M.; Rivas, I.; López-Vicente, M.; Suades-Gonzalez, E.;
Garcia-Esteban, R.; Esnaola, M.; Cirach, M.; et al. Traffic-related air pollution, noise at school, and behavioral
problems in Barcelona schoolchildren: A cross-sectional study. Environ. Health Perspect. 2016, 124, 529.
[CrossRef] [PubMed]
McEwen, B.S.; Flier, J.S.; Underhill, L.H. Protective and Damaging Effects of Stress Mediators. N. Engl. J. Med.
1998, 338, 171–172. [CrossRef] [PubMed]
Lederbogen, F.; Kirsch, P.; Haddad, L.; Streit, F.; Tost, H.; Schuch, P.; Wust, S.; Pruessner, J.; Rietschel, M.;
Deuschle, M.; et al. City living and urban upbringing affect neural social stress processing in humans. Nature
2011, 474, 498–501. [CrossRef] [PubMed]
Lindgren, U.; Nilsson, K.; de Luna, X.; Ivarsson, A. Data resource profile: Swedish microdata research from
childhood into lifelong health and welfare (Umeå SIMSAM Lab). Int. J. Epidemiol. 2016, 45, 1075–1075g.
[CrossRef] [PubMed]
Gustafsson, M.; Forsberg, B.; Orru, H.; Åström, S.; Tekie, H.; Sjöberg, K. Quantification of Population Exposure
to NO2 , PM2.5 and PM10 and Estimated Health Impacts in Sweden 2010; B 2197; IVL: Gothenburg, Sweden, 2014.
Sjöberg, K.; Haeger-Eugensson, M.; Forsberg, B.; Åström, S.; Hellsten, S.; Tang, L. Quantification of Population
Exposure to Nitrogen Dioxide in Sweden 2005; Report B 1749; IVL: Falaise, France, 2007.
Damgaard, A.L.; Hansen, B.M.; Mathiasen, R.; Buchvald, F.; Lange, T.; Greisen, G. Prematurity and
prescription asthma medication from childhood to young adulthood: A Danish national cohort study.
PLoS ONE 2015, 10, e0117253. [CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2017, 14, 1392
37.
38.
39.
40.
41.
42.
43.
44.
12 of 12
Carlsen, H.K.; Zoëga, H.; Valdimarsdóttir, U.; Gíslason, T.; Hrafnkelsson, B. Hydrogen sulfide and particle
matter levels associated with increased dispensing of anti-asthma drugs in Iceland’s capital. Environ. Res.
2012, 113, 33–39. [CrossRef] [PubMed]
Popham, F.; Dibben, C.; Bambra, C. Are health inequalities really not the smallest in the Nordic welfare
states? A comparison of mortality inequality in 37 countries. J. Epidemiol. Commun. Health 2013. [CrossRef]
[PubMed]
Stroh, E.; Rittner, R.; Oudin, A.; Ardo, J.; Jakobsson, K.; Bjork, J.; Tinnerberg, H. Measured and modeled
personal and environmental NO2 exposure. Popul. Health Metr. 2012, 10, 10. [CrossRef] [PubMed]
Harju, M.; Keski-Nisula, L.; Georgiadis, L.; Heinonen, S. Parental smoking and cessation during pregnancy
and the risk of childhood asthma. BMC Public Health 2016, 16, 428. [CrossRef] [PubMed]
Hollams, E.M.; De Klerk, N.H.; Holt, P.G.; Sly, P.D. Persistent effects of maternal smoking during pregnancy
on lung function and asthma in adolescents. Am. J. Respir. Crit. Care Med. 2014, 189, 401–407. [CrossRef]
[PubMed]
Bråbäck, L.; Björ, O.; Nordahl, G. Early determinants of first hospital admissions for asthma and acute
bronchitis among Swedish children. Acta Paediatr. 2003, 92, 27–33. [CrossRef] [PubMed]
Silvestri, M.; Franchi, S.; Pistorio, A.; Petecchia, L.; Rusconi, F. Smoke exposure, wheezing, and asthma
development: A systematic review and meta-analysis in unselected birth cohorts. Pediatr. Pulmonol. 2015, 50,
353–362. [CrossRef] [PubMed]
Burke, H.; Leonardi-Bee, J.; Hashim, A.; Pine-Abata, H.; Chen, Y.; Cook, D.G.; Britton, J.R.; McKeever, T.M.
Prenatal and passive smoke exposure and incidence of asthma and wheeze: Systematic review and
meta-analysis. Pediatrics 2012, 129, 735–744. [CrossRef] [PubMed]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).