Effects of personal particulate matter on peak expiratory flow
rate of asthmatic children
Chin-Sheng Tanga,⁎, Li-Te Changb, Hsien-Chi Leea, Chang-Chuan Chanc
aDepartment of Public Health, College of Medicine, Fu Jen Catholic University, Taipei, Taiwan, ROC
bDepartment of Environmental Engineering and Science, Feng Chia University, Taichung, Taiwan, ROC
cInstitute of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan University, Taipei, Taiwan, ROC
Received 2 December 2006; received in revised form 31 March 2007; accepted 10 April 2007
Available online 22 May 2007
Many researches have shown that the particulate matter (PM) of air pollution could affect the pulmonary functions, especially
for susceptible groups such as asthmatic children, where PM might decrease the lung function to different extents. To assess the
effects of PM on health, most studies use data from ambient air monitoring sites to represent personal exposure levels. However,
the data gathered from these fixed sites might introduce certain statistical uncertainties. The objectives of this study are to evaluate
the effects of various size ranges of PM on peak expiratory flow rate (PEFR) of asthmatic children, and to compare the model
performance of using different PM measurements (personal exposures versus fixed-site monitoring) in evaluation. Thirty asthmatic
children, aged 6 to 12 years, who live near the fixed monitoring site in Sin-Chung City, Taipei County, Taiwan, were recruited for
the study. Personal exposures to PM1, PM2.5, and PM10were measured continuously using a portable particle monitor (GRIMM
Mode 1.108, Germany). In addition, an activity diary and questionnaires were used to investigate possible confounding factors in
their home environments. The peak expiratory flow rate of each participant was monitored daily in the morning and in the evening
for two weeks. Results showed several trends, although not necessarily statistically significant, between personal PM exposures
and PEFR measurements in asthmatic children. In general, notable findings tend to implicate that not only fine particles (PM2.5) but
also coarse particles (PM2.5–10) are likely to contribute to the exacerbation of asthmatic conditions. Stronger lagged effect and
cumulative effect of PM on the decrements in morning PEFR were also found in the study. Finally, results of linear mixed-effect
model analysis suggested that personal PM data was more suitable for the assessment of change in children's PEFR than ambient
© 2007 Elsevier B.V. All rights reserved.
Keywords: Particulate matter; Coarse particle; Peak expiratory flow rate; Personal exposures; Asthmatic children
Numerous epidemiological researches have shown
that respiratory morbidity and mortality and declines in
lung function are associated with current levels of
particulate pollution in urban air (Dockery et al., 1993;
Pope et al., 1995; Vedal et al., 1998; Schwartz and Neas,
2000; Penttinen et al., 2001; Brunekreef and Holgate,
Science of the Total Environment 382 (2007) 43–51
⁎Corresponding author. Department of Public Health, College of
Medicine, Fu Jen Catholic University, 510 Chung Cheng Road,
Hsinchuang, Taipei County 24205, Taiwan, ROC. Tel.: +886 2
29053433; fax: +886 2 29056385.
E-mail address: email@example.com (C.-S. Tang).
0048-9697/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
2002). In these studies, particulate matter (PM) has
usually been measured as the mass of particles smaller
than 10 μm (PM10) or 2.5 μm (PM2.5) in diameter,with a
central monitoring site serving as a surrogate for
personal exposures. The relationship between particu-
late pollution and lung function, principally peak expi-
ratory flow rate (PEFR), is mostly consistent, despite
differences in definitions of outcome measurements and
statistical methods used to model the relationship be-
tween air pollution and health (Neas et al., 1995; Gielen
et al., 1997; Trenga et al., 2006; Bourotte et al., 2007).
Asthma is a priority regarding child health care in
Taiwan. While the mortality rate related to asthma in
Taiwan has slowly decreased over the past two decades,
the prevalence of asthma under children has increased
from 1.3% in 1974 to 10.8% in 1994 (Hsieh and Shen,
1988; Kuo et al., 2003; Jan et al., 2004). A recent study
further suggested a slight increase in asthma prevalence
for elementary school children in Taiwan, compared to
the results reported 10 years ago (Chang et al., 2006).
health, but uncontrolled asthma, which could result in
higher demands on emergency services and hospitaliza-
tion, is also a significant financial burden on health care
systems (Barnes et al., 1996).
Despite the progress that has been made to date, com-
paratively few studies have directly assessed the relative
of lung functions in asthmatic children. This is due, in
for personal exposures (Howard-Reed et al., 2000). To
address this issue, we investigated the effects of particu-
children with asthma. The relationships between particles
of different sizes (1 to 10 μm) and PEFR were examined.
In addition, results of model performance using personal
monitoring for particulate exposures were compared to
those using data from a central ambient monitoring site.
2.1. Study design
This panel study was conducted to monitor changes
in personal PM exposures and PEFR simultaneously for
the subjects during December 2003 to February 2005.
Thirty asthmatic children in Taipei County were recruit-
ed, all of whom were enrolled in the same elementary
school and live within 2 km from a stationary moni-
toring site (Taiwan Particle Supersite), which is operated
by Taiwan Environmental Protection Administration
(Taiwan EPA). In each 2-week session, each subject
completed both the continuous PM exposure assessment
in the first five days and the PEFR monitoring procedure
throughout the 14 session days. Asthma was diagnosed
by a physician and was thus defined according to the
criteria of the American Thoracic Society. At the begin-
ning of each sampling session, field staff obtained data
from each subject, including sex, age, height, weight,
symptoms in the past 12 months, and medical history.
Furthermore, household information such as presence of
cockroaches, dust mites, mold, furry pets, carpeting,
plants inside the house, home dampness, environmental
tobacco smoke (ETS), gas cooking appliances, mosqui-
to repellant, and incense burning were collected through
walk-through survey by a technician. Field staff also
collected information regarding environmental factors
of physical nature such as whether the house contained
an air cleaner or air conditioner, as well as the presence
of outdoor traffic, industries, and temple pollution. The
review board of the Environmental Protection Bureau of
Taipei County approved the research protocol, and a
written consent was obtained from each participant's
parents before the study was launched.
2.2. PEFR monitoring
During the study, each subject performed PEFR
maneuvers in the morning after awaking and in the
evening near bedtime every day of the 2-week session.
The subjects were instructed to take the measurements
before taking any medication. The highest of three
values from consecutive PEFR maneuvers in the
morning, as well as in the evening, were retained for
further analysis. Subjects were trained by field staff to
handle the electronic PEFR monitor (Asthma Monitor
PF-100, Microlife, Taiwan) properly and were given a
standard operation procedure for self-measuring. PEFR
measurements were recorded in the monitor automati-
cally and were later downloaded by field staff.
2.3. PM monitoring
Personal exposures to different sizes of particles were
measured continuously using a portable particle monitor
(DUSTcheck Portable Dust Monitor, model 1.108,
GRIMM Labortechnik Ltd., Germany). Mass concentra-
tions of PM1, PM2.5, and PM10, as well as ambient
temperature and relative humidity, were measured and
recorded forone-minute periods. The raw data were then
summarized to one-hour segments for statistical analy-
sis. Mass concentrations of PM2.5–10were obtained by
subtracting the PM2.5fraction from the concurrent PM10
levels. A similar approach was applied to derive the data
44 C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
for PM1–2.5concentrations. Before and after field sam-
the manufacturer with reference methods.
To measure the participants' PM exposures during
DUSTcheck monitor was assigned to each subject from
07:00 to 21:00 daily. Detailed information on household
activities and time-activity patterns of the monitored
individual were noted in personal diaries. During each
day, subjects reported various microenvironments they
visited, including living room, bedroom, classroom,
cram school, other non-residential indoor areas, and
outdoor places. The information of diary entries was
reconfirmed by the field staff member immediately after
daily sampling. Personal PM exposures were only calcu-
lated if the DUSTcheck monitor was with the subject at
least 75% of the time during sampling.
In addition to personal exposure monitoring, ambient
PM2.5and PM10concentrations were measured at the
Taiwan Particle Supersite using tapered-element oscillat-
ing microbalance (TEOM) monitors (model 1400a,
Thermo Electron Corp., USA), an inertial instrument
that measures particle mass in real time on an exchange-
able filter cartridge by monitoring frequency changes of a
tapered element. Results of collocating the DUSTcheck
monitors simultaneously, withR2values of 0.90 and 0.91
for PM10and PM2.5, respectively (Chuang et al., 2005).
2.4. Statistical analysis
To estimate the time course of PM effects on PEFR,
we calculated the PM exposure levels as the mean
24-h concentration for the day on which the effect on
lung function was measured (lag 0), and for the 2 pre-
ceding days (lag 1 and lag 2). Also, cumulative par-
ticulate exposures were defined as the 2-day mean
concentration of lag 0 and lag 1 (2-day mean), and the
3-day mean concentration of lag 0 to lag 2 preceding that
day (3-day mean).
Linear mixed-effect models were used to estimate the
effect of particulate exposures on PEFR, adjusted for
personal and meteorological variables. Such mixed-
effect models have the advantage of adjustment for
invariant variables by fixed-effect models and account-
ing for individual differences by random-effect models
(Diggle et al., 2002; Chan and Wu, 2005; Bourotte et al.,
2007). To control the potential confounding variables of
subject characteristics and environmental factors, we
treated each subject's gender, age, body mass index
(BMI), history of respiratory or atopic disease in the
family, ETS, acute asthmatic exacerbation in the past
12 months, ambient temperature and relative humidity,
presence of indoor pollutants (yes/no), and presence of
outdoor pollutants (yes/no) as fixed effects, and each
subject as a random effect in the mixed-effect model.
Particulate levels with different exposure metrics (lags
the time course of particulate effects. To standardize the
association between PMexposure andPEFR,the current
study expresses results as the change in PEFR for an
interquartile range (IQR) in PM levels, i.e. difference
between the 25th and 75th percentiles (Lipfert and
ambient PM levels with personal PM exposures for
PEFR change assessment, the PM variable in linear
mixed-effect models was included using either personal
measurements or ambient concentrations measured at
the central monitoring site. Statistical analyses were
performed using the MIXED procedure in the S-Plus
2000 program (MathSoft Inc., Cambridge, MA, USA).
Model selection was based on minimizing Akaike's
Basic characteristics of 30 asthmatic children participating in the study
Gender (n) (male/female)
Acute exacerbation of asthma in past 12 months
0 [n (%)]
1 [n (%)]
2 [n (%)]
3 [n (%)]
Anti-inflammatory medication use
Yes [n (%)]
No [n (%)]
Respiratory or atopic disease in family
Yes [n (%)]
No [n (%)]
Presence of indoor air pollutants
ETS exposure [n (%)]
Mold [n (%)]
Cockroaches [n (%)]
Incense burning [n (%)]
Presence of outdoor air pollutants
Industries [n (%)]
Traffic exposure [n (%)]
Temples [n (%)]
Morning PEFR (L/min) (n=404)
Evening PEFR (L/min) (n=517)
Values are mean±SD (range) unless otherwise noted.
45C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
information criterion (AIC) (Akaike, 1974; Delfino et al.,
2004; Bourotte et al., 2007).
The demographic data are summarized in Table 1.
The study population consisted of 20 boys and 10 girls,
with an average age of 10 years. Twelve of the subjects
(40%) had had at least one episode of acute asthma in
the 12 months preceding the study, while only 11 chil-
dren (37%) were using anti-inflammatory medication.
Among the participants, 9 children (30%) were being
exposed to ETS, and 13 (43.3%) to residential incense
burning. Finally, the average PEFR for the children was
240 L/min, ranging from 76 to 658 L/min individually,
with the mean and the range of PEFR slightly higher in
the evening than in the morning.
Table 2 summarizes the PM monitoring data with
different sizes, temperature and relative humidity. Dur-
ing the study period, 1-h average personal exposures
(mean ± SD) to PM10, PM2.5, and PM1were 51.8 ± 39.5,
34.0 ± 28.9, and 27.8 ± 25.3 μg/m3, respectively. The
microenvironmental meteorological conditions were usu-
ally warm, with an hourly temperature of 27.1 ± 5.2 °C
and a relative humidity of 60.9 ± 10.1%. Additionally,
irrespective of particle size, average ambient PM con-
centrations were lower than the corresponding personal
PM exposures and showed less variability. Neither per-
sonal exposures nor ambient concentrations of PM2.5
(24-h standard of 35 μg/m3for PM2.5) (US EPA, 2006).
To compare the lagged and cumulative effects of PM
exposures on morning and evening PEFR measure-
range (IQR) increase in the PM exposures are presented
in Table 3. Regarding lag modes (lags 0–2), only a slight
variation in the evening PEFR was observed for differ-
ent PM exposure metrics. For instance, regardless of the
particle size, theeveningPEFRchangeswith anincrease
in personal PM exposures measured on the same day
not statistically significant, a negative influence on morn-
ing PEFR changes was reported for different exposures
(Fig. 1A). In addition, stronger lagged effects on morning
PEFR were found for PM2.5–10exposures as compared to
those for the smaller-size particles (PM1–2.5and PM1).
Summarized statistics for 1-h personal exposures and ambient
monitoring data during the study period
Mean±SD IQR⁎Minimum MaximumVariableN
Relativehumidity(%) 3639 60.9±10.1 13.9
3639 51.8±39.5 42.9
3639 17.8±19.6 15.9
3639 34.0±28.9 27.6
3639 27.8±25.3 24.5
Ambient monitoring data
⁎IQR (interquartile range)=Q3–Q1.
3934 48.4±26.5 32.0
3934 17.0±10.6 11.7
3934 31.4±18.8 21.3
3920 72.8±12.6 15.3
Changes in PEFR (95% CI) for interquartile increase in personal PM exposures estimated by linear mixed-effects models⁎
Mode N PM10
2-day mean 36 −59.91 (−133.94, 14.11) −37.15 (−105.01, 30.7) −44.06 (−113.79, 25.67) −32.16 (−102.93, 38.6) −45.67 (−117.09, 25.74)
3-day mean 24 −0.56 (−101.7, 100.57) −35.47 (−127.32, 56.38) −6.01 (−101.48, 89.46) −7.70 (−100.26, 84.87) −5.69 (−105.96, 94.59)
64 −15.99 (−45.07, 13.08) −20.55 (−45.83, 4.73)
37 −44.78 (−117.64, 28.08) −39.05 (−104.16, 26.06) −12.26 (−77.6, 53.09)
25 −41.45 (−98.32, 15.41) −39.56 (−79.56, 0.44)
−6.44 (−30.18, 17.29)
−10.51 (−39.93, 18.91) −6.00 (−29.85, 17.85)
−6.79 (−72.97, 59.4)
−11.73 (−107.13, 83.66) −24.87(−71.49, 21.74)
−12.52 (−77.93, 52.9)
−4.38 (−54.79, 46.03)
2-day mean 57 2.58 (−29.32, 34.49)
3-day mean 26 30.91 (−15.3, 77.12)
87 0.06 (−21.55, 21.68)
58 −1.71 (−23.91, 20.49)
26 11.32 (−17.17, 39.81)
−1.68 (−19.13, 15.78)
1.59 (−14.32, 17.5)
0.86 (−30.84, 32.57)
5.97 (−15.57, 27.5)
29.75 (−1.69, 61.18)
1.17 (−17.79, 20.13)
−4.98 (−27.77, 17.81)
11.30 (−11.55, 34.16)
41.74 (11.36, 72.13)
28.21 (−19.08, 75.5)
4.75 (−15.13, 24.64)
6.60 (−16.05, 29.25)
−6.38 (−44.61, 31.85)
43.90 (20.42, 67.37)
34.92 (−9.71, 79.56)
0.50 (−18.82, 19.82)
16.66 (−7.59, 40.9)
11.60 (−11.1, 34.31)
39.97 (7.1, 72.85)
−3.32 (−66.14, 59.5)
⁎Adjusted for gender, age, BMI, disease history of family, ETS exposure, acute asthmatic exacerbation in past 12 months, temperature, relative
humidity, presence of indoor air pollutants, and presence of outdoor air pollutants.
46 C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
The decrement in the morning PEFR for asthmatic
6.00 L/min per increase of IQR for same day (lag 0)
exposures to PM2.5–10, PM1–2.5, and PM1, respectively.
As to the cumulative modes (lag 0, 2-day mean, 3-day
mean), negative impacts (though statistically insignifi-
cant) on morning PEFR were also reported throughout
exposures to different particle sizes (Table 3). In general,
coarse particles (PM2.5–10) had comparatively larger
decreasing effects on morning PEFR than PM1–2.5and
PM1(Fig. 1B). Among the three different cumulative
modes, preceding 2-day mean of PM exposures indicated
of the PM size ranges. For example, the decrement in the
morning PEFR was 39.56 L/min, 11.73 L/min, and
24.87 L/min per increase of IQR for 2-day mean expo-
sures to PM2.5–10, PM1–2.5, and PM1, respectively.
Finally, for different cumulative modes, PM exposures
To investigate the suitability of different types of PM
data for PEFR change assessment, PM2.5and PM10
levels measured at the central monitoring site (Taiwan
Particle Supersite) were also applied in the linear mixed-
Fig. 1. Estimated lagged (A) and cumulative (B) effects of an interquartile increase in personal PM exposures on morning PEFR changes. Error bars
indicate mean±95% CI.
47C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
effect models to replace the corresponding personal PM
measurements for comparison. As shown in Fig. 2,
Akaike's information criterion values in the personal
PM exposure models were consistently lower than those
in the ambient PM models, indicating a better represen-
ment. Moreover, for both of the lagged and cumulative
of particle size, had the lowest AIC values.
In the current study, data of personal exposures to
different sizes of PM were adopted in the analysis. On
the other hand, most existing studies of PM and their
association with lung function outcomes use total
suspended particles (TSP), PM10, or PM2.5 from
stationary ambient monitoring sites as the measurement
for PM exposure, with rather limited data collected for
PM1measurements. Previous studies have suggested
that exposure misclassification from using stationary
PM data may have diminished the accuracy of exposure-
response estimates compared with personal exposures,
thus potentially weakening associations with stationary
PM (Delfino et al., 2004). To measure the participants'
PM exposures during normal daytime activities and to
avoid the change of activity patterns due to sampling
inconvenience, in our study, personal PM exposures
were measured by field staff carrying a DUSTcheck
monitor with the subject throughout different micro-
environments (7 am to 9 pm). At home during the night,
the participants’ parents were asked to take the monitor
48 C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
to different residential microenvironments with the
children and make notes in the time-activity diary. The
information of diary entries were reconfirmed by field
staff the next morning to assure the data quality. Since
the participants were all school children, their daily
activities were actually less variable and were easier to
follow properly as compared to those for adults. As
shown in Fig. 2, the AIC values in the personal PM
models were consistently lower than those in the corre-
sponding ambient PM models. Recently, researchers
found that personal PM was more strongly associated
with forced expiratory volume in 1 s (FEV1) than was
stationary-site PM, where the finding of stronger asso-
ciations with personal PM exposures is also consistent
with those in our study (Delfino et al., 2004).
Public health policies, in terms of establishing guide-
lines or standards for acceptable levels of ambient PM
pollution, have been focusing primarily on indicators of
fine particles (PM2.5) and thoracic particles (PM10)
(WHO, 2005; US EPA, 2006). In the meantime, several
studies suggest relative roles for PM10and for both fine
(PM2.5) and coarse (PM2.5–10) PM mass fractions on
pulmonary function changes, which is consistent with
the findings in our study (Pekkanen et al., 1997; Peters
et al., 1997a,b; Sheppard et al., 1999; Von Klot et al.,
2002; Bourotte et al., 2007). Coarse PM is mainly de-
posited into the trachebronchial region of the respiratory
tract, while the fine PM penetrates deeper into the
alveolar region, where it may contribute to the release
of particulate chemical constituents potentially bio-
available after inhalation and inducing an inflamma-
tory reaction, which may lead to asthma exacerbation
(Lighty et al., 2000; Espinosa et al., 2001). Even so,
there is still limited evidence for a stronger effect of fine
versus coarse fraction particles; also, submicron parti-
cles do not appear to have any notable stronger effect
than other larger-diameter fine particles (Peters et al.,
1997a,b; Yu et al., 2000; Penttinen et al., 2001; Von Klot
et al., 2002; Brunekreef and Forsberg, 2005; Bourotte
et al., 2007). Further studies are needed to identify the
circumstances under which some classes of PM may
cause little or no adverse reduction of lung function, as
well as those under which PM components may cause
The present study showed that the increase in PM
was associated (though statistically insignificant) with
decrements in PEFR, which was similar to those in
previous epidemiological studies. In these studies, for
both PM2.5 and PM10 pollution, the available point
estimates for morning PEFR lagged 0 day (Pekkanen
et al., 1997), lagged 1 day (Romieu et al., 1996, 1997;
Gielen et al., 1997; Peters et al., 1997a,b), and lagged
2 days (Romieu et al., 1996; Timonen and Pekkanen,
1997; Segala et al., 1998) showed decreases, while the
majority of the results were not statistically significant.
On the other hand, in our study, stronger lagged effect
(lag 2) and cumulative effect (2-day mean) of PM on
PEFR were observed, which was not necessarily con-
sistent with findings in earlier studies. Results from our
study imply that effects of PM exposures on PEFR
changes could be better predicted for longer exposure
frames. A possible explanation is that the character of
PM varied greatly from place to place and over time,
depending upon factors such as pollution sources and
the prevailing atmospheric conditions. Hence, a distrib-
uted lag model should more correctly capture all asso-
ciations, because selection of any sole lag day would
neglect associations on other lag days (Peters et al.,
2001). Until now, to our knowledge, few data sets con-
taining the everyday data required for a distributed
lag analysis exist. Nevertheless, for those that have
been analyzed for distributed lag, results showed more
excess risk associated with a distributed lag analysis
than with any single day analysis (Mar et al., 2003;
Furthermore, our results show a certain proportion of
subjects (30%) were exposed to ETS during the study
period (Table 1). ETS has been identified as a significant
contributor to measured 24-h personal PM2.5exposures
(Chang et al., 2003), and evidence has shown that
exposure to ETS increases the risk of lower respiratory
tract illness in children (Ferrence and Ashley, 2000). To
control the effects of sidestream smoke in our study,
ETS exposure was included in the mixed-effect model
analysis and no significant impact on decreased PEFR
was found, which could be attributed to the very small
amount of time the subjects were exposed to ETS.
In our results, adverse effect trends of PM were
observed on morning PEFR but not on evening PEFR. It
is not clear why morning PEFR is more responsive to
PM than evening PEFR. A possible reason is that airway
narrowing in people suffering from asthma is frequently
at its worst in the early morning hours. Morning PEFR is
not influenced by daily activities or medication used
later during the day, and it may therefore be a more
sensitive indicator of airway narrowing than evening
PEFR (Pride, 1992; Timonen and Pekkanen, 1997).
limitations should be considered. Although our models
accounted for autocorrelation of the events and for the
effect of temperature and relative humidity, unmeasured
air pollutants could also have been responsible for the
lowering of lung function in the study rather than PM
alone. It should be noted that, while some studies showed
49 C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
of studies where PM was highly correlated with other
gaseous pollutants the PM effect estimates remained ro-
1998; Jalaludin et al., 2000; Tolbert et al., 2000; Yu et al.,
(O3, NO2, SO2, and CO) in the community concerned
were low during the sampling period, confounding of the
study (Taiwan EPA, 2003–2005).
The current study is also limited in the scope of
inferences that can be made as particle exposures were
represented only by total mass concentrations. Further
advancement in assessing personal exposure is needed,
which includes assessment of allergenic sources that
may influence airway inflammation or induce bronch-
oconstriction. These exposure assessment methodolo-
gies should be developed for use in future epidemiologic
research (Delfino et al., 2004).
Finally, our panel of asthmatic children had physician-
diagnosed asthma. Although the mixed-effect models we
applied in the analysis have the advantage of adjustment
for individual differences by random-effect models, it is
possible that a more accurate classification of individuals
based on objective indices of asthma, such as bronchial
hyperactivity, would strengthen the associations. It has
been shown that individuals with greater airway liability,
for example, are more acutely responsive to air pollution,
regardless of asthma diagnosis (Boezen et al., 1998;
McConnell et al., 1999).
The current study presented several trends between
personal PM exposures and asthmatic children's PEFR
measurements, of which some are not necessarily sta-
tistically significant. In brief, notable findings tend to
implicate that not only fine particles (PM2.5) but also
coarse particles (PM2.5–10) are likely to contribute to
exacerbation of asthmatic conditions. Stronger lagged
effect (lag 2) and cumulative effect (2-day mean) of PM
on the decrements in morning PEFR were found in the
study.Finally, personalPMexposures weremoresuitable
for evaluating the association between particle pollution
and the PEFRchanges thanambient PMmonitoring data.
The authors thank all of the subjects who participated
Taiwan EPA. This study was funded by the Environmen-
tal Protection Bureau of Taipei County.
Akaike H. A new look at the statistical model identification. IEEE
Trans Automat Contr 1974;19:716–23.
Barnes PJ, Jonsson B, Klim JB. The costs of asthma. Eur Respir J
Boezen M, Schouten J, Rijcken B, Vonk J, Gerritsen J, Zee S van der,
et al. Peak expiratory flow variability, bronchial responsiveness
and susceptibility to ambient air pollution in adults. Am J Respir
Crit Care Med 1998;158:1848–54.
Bourotte C, Curi-Amarante AP, Forti MC, Pereira LAA, Braga AL,
aerosol soluble fraction and peak expiratory flow of asthmatic
Brunekreef B, Holgate ST. Air pollution and health. Lancet 2002;360
Brunekreef B, Forsberg B. Epidemiological evidence of effects of
coarse airborne particles on health. Eur Respir J 2005;26:309–18.
Chan CC, Wu TH. Effects of ambient ozone exposure on mail carriers'
peak expiratory flow rates. Environ Health Perspect 2005;113
Chang LT, Koutrakis P, Catalano PJ, Suh HH. Assessing the
importance of different exposure metrics and time-activity data
to predict 24-h personal PM2.5 exposures. J Toxicol Environ
Health-Part A 2003;66(16-19):1825–46.
Chang LT, Lin YJ, Tang CS. The prevalence of asthma for children of
elementary school in eight towns of Taipei County. Fu Jen J Med
ChuangKJ, Chan CC, Chen NT, Su TC, Lin LY. Effects of particle size
fractions on reducing heart rate variability in cardiac and
hypertensive patients. Environ Health Perspect 2005;113:1693–7.
for respiratory illnesses among the elderly in Montreal: association
with low level ozone exposure. Environ Res 1998;76:67–77.
Delfino RJ, Quintana PJE, Floro J, Gastañaga VM, Samimi BS,
Kleinman MT, et al. Association of FEV1 in asthmatic children
with personal and microenvironmental exposure to airborne
particulate matter. Environ Health Perspect 2004;112(8):932–41.
Diggle PJ, Heagerty P, Liang KY, Zeger SL. Analysis of longitudinal
data. New York: Oxford University Press; 2002.
An association between air pollution and mortality in six US cities.
N Engl J Med 1993;329:1753–9.
Espinosa AJF, Rodrıguez MT, De la Rosa FJB, Sanchez JCJ. Size
distribution of metals in urban aerosols in Seville (Spain). Atmos
Ferrence R, Ashley MJ. Protecting children from passive smoking. Br
Med J 2000;321:310–31.
Acute effects of summer air pollution on respiratory health of
asthmatic children. Am J Respir Crit Care Med 1997;155: 2105–8.
Howard-Reed C, Rea AW, Zufall MJ, Burke JM, Williams RW, Suggs
JC, et al. Use of a continuous nephelometer to measure personal
exposure to particles during the U.S. Environmental Protection
Agency Baltimore and Fresno panel studies. J Air Waste Manage
Hsieh KH, Shen JJ. Prevalence of childhood asthma in Taipei, Taiwan,
and other Asian Pacific countries. J Asthma 1988;25:73–82.
Jalaludin BB, Chey T, O'Toole BI, Smith WT, Capon AG, Leeder SR.
rate in a cohort of Australian children. Int J Epidemiol 2000;29:
50 C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51
Jan IS, Chou WH, Wang JD, Kuo SH. Prevalence and major risk Download full-text
factors for adult bronchial asthma in Taipei City. J Formos Med
Kuo LC, Shau WY, Yang PC, Kuo SH. Trends in asthma mortality in
Taiwan, 1981–2000. J Formos Med Assoc 2003;102:534–8.
Lighty JS, Veranth JM, Sarofim AF. Combustion aerosols: factors
governing their size and composition and implications to human
health. J Air Waste Manage Assoc 2000;50(9):1565–618.
Lipfert FW, Wyzga RE. Statistical considerations in determining the
health significance of constituents of airborne particulate matter.
J Air Waste Manag Assoc 1999;49(9):182–91.
Mar TF, Norris GA, Larson TV, Wilson WE, Koenig JQ. Air pollution
and cardiovascular mortality in Phoenix, 1995–1997. Revised
analyses of time-series studies of air pollution and health. special
report. Boston, MA: Health Effects Institute; 2003. p. 177–82.
Air pollution and bronchitic symptoms in Southern California
Children with asthma. Environ Health Perspect 1999;107:757–60.
Neas LM, Dockery DW, Koutrakis P, Tollerud DJ, Speizer FE. The
association of ambient air pollution with twice daily peak
expiratory flow rate measurements in children. Am J Epidemiol
Pekkanen J, Timonen KL, Ruuskanen J, Reponen A, Mirme A. Effects
of ultrafine and fine particles in urban air on peak expiratory flow
among children with asthmatic symptoms. Environ Res 1997;74:
Penttinen P, Timonen KL, Tiittanen P, Mirme A, Ruuskanen J,
Pekkanen J. Number concentration and size of particles in urban
air: effects on spirometric lung function in adult asthmatic subjects.
Environ Health Perspect 2001;109:319–23.
Peters A, Dockery DW, Heinrich J, Wichmann HE. Short-term effects
of particulate air pollution on respiratory morbidity in asthmatic
children. Eur Respir J 1997a;10:872–9.
Peters A, Wichmann HE, Tuch T, Heinrich J, Heyder J. Respiratory
effects are associated with the number of ultrafine particles. Am J
Respir Crit Care Med 1997b;155:1376–83.
Peters A, Dockery DW, Muller JE, Mittleman MA. Increased
particulate air pollution andthe triggeringof myocardial infarction.
Pope III CA, Thun MJ, Namboodiri MM, Dockery DW, Evans JS,
SpeizerFE,et al.Particulateair pollutionas apredictorof mortality
in a prospective study of US adults. Am J Respir Crit Care Med
Pride NB. Physiology. In: Clark THJ, Godlfrey S, Lee HT, editors.
Romieu I, Meneses F, Ruiz S, Sienra JJ, Huerta J, White MC, et al.
Effects of air pollution on the respiratory health of asthmatic
children living in Mexico City. Am J Respir Crit Care Med
Romieu I, Meneses F, Ruiz S, Huerta J, Sienra JJ, White M, et al.
Effects of intermittent ozone exposure on peak expiratory flow and
respiratory symptoms among asthmatic children in Mexico City.
Arch Environ Health 1997;52:368–76.
analyses of time-series studies of air pollution and health. Special
report. Boston, MA: Health Effects Institute; 2003. p. 211–8.
Schwartz J, Neas LM. Fine particles are more strongly associated than
coarse particles with acute respiratory health effects in school-
children. Epidemiology 2000;11:6–10.
Segala C, Fauroux B, Just J, Pascual L, Grimfeld A, NeukirchF. Short-
termeffect of winter air pollutionon respiratory health of asthmatic
children in Paris. Eur Respir J 1998;11:677–85.
Sheppard L, Levy D, Norris G, Larson TV, Koenig JQ. Effects of
ambient air pollution on nonelderly asthma hospital admissions in
Seattle, Washington, 1987–1994. Epidemiology 1999;10:23–30.
Timonen KL, Pekkanen J. Air pollution and respiratory health among
childrenwith asthmatic or cough symptoms. Am J Respir Crit Care
Tolbert PE, Mulholland JA, Maclntosh DL, Xu F, Daniels D, Devine
OJ, et al. Air quality and pediatric emergency room visits for
asthma in Atlanta, Georgia. Am J Epidemiol 2000;151:798–810.
Trenga CA, Sullivan JH, Schildcrout JS, Shepherd KP, Shapiro GG, Liu
LJS, et al. Effect of particulate air pollution on lung function in adult
and pediatric subjects in a Seattle panel study. Chest 2006;129:
US EPA. National Ambient Air Quality Standards for particulate
matter. Final rule. Fed Reg 2006;71:61144–233.
Vedal S, Petkau J, White R, Blair J. Acute effects of ambient inhalable
particles in asthmatic and nonasthmatic children. Am J Respir Crit
Care Med 1998;157:1034–43.
ultrafine particles. Eur Respir J 2002;20:691–702.
WHO. Air quality guidelines global update 2005. Copenhagen: World
Health Organization, Regional Office for Europe; 2005.
Yearbook of Environmental Protection Statistics, Republic of China,
2003–2005. Taipei: Environmental Protection Administration,
Executive Yuan; 2003–2005.
Yu O, Sheppard L, Lumley T, Koenig JQ, Shapiro GG. Effects of
ambient air pollution on symptoms of asthma in Seattle-area
children enrolled in the CAMP study. Environ Health Perspect
51 C.-S. Tang et al. / Science of the Total Environment 382 (2007) 43–51