![Estimated percent increases [mean (95% CI)] in total, cardiovascular, and respiratory mortality per IQR increase in pollutant concentrations on the current day (lag 0) or the previous 1-3 days (lags 1, 2, and 3), adjusted for temporal trend, day of the week, temperature, relative humidity, and SO 2 and NO 2 concentrations. Percent increase Percent increase Percent increase](https://www.researchgate.net/profile/Junji-Cao/publication/263686286/figure/fig1/AS:669279519260692@1536580080799/Estimated-percent-increases-mean-95-CI-in-total-cardiovascular-and-respiratory_Q320.jpg)
Content uploaded by Junji Cao
Author content
All content in this area was uploaded by Junji Cao on Jul 07, 2014
Content may be subject to copyright.
Environmental Health Perspectives
•
v o l u m e 120 | n u m b e r 3 | March 2012
373
Research
Numerous epidemiological studies during the
past 20 years have confirmed that short- and
long-term exposure to outdoor air pollution
contributes to increased cardiopulmonary mor-
tality and morbidity (Brunekreef and Holgate
2002; Pope and Dockery 2006). Among vari-
ous pollutants in the ambient mixture, fine
particulate matter (PM2.5; particles ≤ 2.5 µm in
aerodynamic diameter) shows the most consis-
tent association with adverse health outcomes
and therefore is of great public health con-
cern (Ito et al. 2011; Ostro et al. 2007; Peng
et al. 2009; urston et al. 2005; Zhou et al.
2011). However, the chemical components
of PM2.5 responsible for these effects are still
unknown. As the U.S. National Academy of
Science pointed out, it is important to under-
stand the contributions of specific components
of ambient particulate matter (PM) to cardio-
pulmonary and other health effects (National
Research Council 1998).
China has one of the highest PM2.5 lev-
els in the world (van Donkelaar et al. 2010).
However, PM2.5 is still not a criteria pollutant
in China, and few studies in the country have
investigated the adverse health effects of PM2.5
because of a lack of monitoring data. Currently,
the Chinese government is reviewing its Air
Quality Standards (AQS) and proposing to
set the annual and daily average PM2.5 stan-
dards as 35 µg/m3 and 75 µg/m3, respectively
(Chinese Ministry of Environmental Protection
2010). To our knowledge, only three published
studies have estimated the effects of PM2.5 on
daily mortality in China (Kan et al. 2007; Ma
et al. 2011; Venners et al. 2003). Kan et al.
(2007) and Ma et al. (2011) found significant
associations between PM2.5 and daily mortal-
ity in Shanghai and Shenyang, China, whereas
Venners et al. (2003) observed negative but
statistically insignificant associations between
PM2.5 and daily mortality in Chongqing.
Obviously, more studies are needed to investi-
gate the health effects of PM2.5 and its chemical
constituents in China.
In the present study, we examined short-
term associations between PM2.5 constituents
and cardiopulmonary mortality in Xi’an, a
heavily polluted Chinese city.
Methods
Data. Xi’an, with an area of 9,983 km2 and a
resident population > 8.1 million in 2005, is
the capital of Shanxi Province, China. Xi’an
is the largest city in northwestern China, and
it experiences some of the worst air pollution
among China’s cities (Cao et al. 2005). Our
study area was limited to the urban area of
Xi’an, an area of 1,166 km2 with a resident
population of > 2.7 million.
Mortality data. We obtained numbers
of deaths among urban residents in Xi’an
for each day for 1 January 2004 through
31 December 2008 from the Shanxi Provincial
Center for Disease Control and Prevention
(SPCDCP). In Xi’an, all deaths, regardless of
whether they occur in a hospital or at home,
must be reported to appropriate authorities
before cremation of the remains. Hospital or
community doctors must indicate the cause
of death on a death certificate card that is
sent to the SPCDCP. SPCDCP staff then
classify the cause of death according to the
International Classification of Diseases, 10th
Revision [ICD-10; World Health Organization
(WHO) 1992] as due to total nonaccidental
causes (ICD-10 codes A00–R99), cardio-
vascular diseases (I00–I99), respiratory diseases
(J00–J98), or injury (S00–T98). e Chinese
government has mandated detailed quality
assurance (QA) and quality control (QC)
programs for the SPCDCP death registry.
Pollutant and meteorological data. For
this study, we measured daily concentrations
of PM2.5, organic carbon (OC), elemental
carbon (EC), and 10 water-soluble ions [i.e.,
sodium ion (Na+), ammonium (NH4+), potas-
sium ion (K+), magnesium ion (Mg2+) calcium
ion (Ca2+), flouride (F–), choride (Cl–), nitrite
(NO2–), sulfate (SO42–) and nitrate (NO3–)]
for 1 January 2004 through 31 December
2008 (1,827 days). We also measured concen-
trations of 15 elements [i.e., sulfur (S), chlorine
(Cl), potassium (K), calcium (Ca), titanium
(Ti), chromium (Cr), manganese (Mn), iron
(Fe), nickel (Ni), zinc (Zn), arsenic (As), boron
(Br), molybdenum (Mo), cadmium (Cd),
Address correspondence to H. Kan, P.O. Box 249,
130 Dong-An Rd., Shanghai 200032, China. E-mail:
haidongkan@gmail.com; or J. Cao, No. 10 Fenghui
South Rd., High-Tech Zone, Xi’an 710075, China.
E-mail: cao@loess.llqg.ac.cn
Supplemental Material is available online (http://
dx.doi.org/10.1289/ehp.1103671).
This study was supported by the National
Basic Research Program (973 program) of China
(2011CB503802), National Natural Science Foundation
of China (40925009), and Gong-Yi Program of China
Ministry of Environmental Protection (200809109,
200909016, and 201209008).
e authors declare they have no actual or potential
competing financial interests.
Received 11 March 2011; accepted 3 January 2012.
Fine Particulate Matter Constituents and Cardiopulmonary Mortality
in a Heavily Polluted Chinese City
Junji Cao,1 Hongmei Xu,1 Qun Xu,2 Bingheng Chen,3 and Haidong Kan3,4
1State Key Laboratory of Loess and Quaternary Geology (SKLLQG), Institute of Earth Environment, Chinese Academy of Sciences, Xi’an,
China; 2Department of Epidemiology and Health Statistics, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences,
Peking Union Medical College, Beijing, China; 3School of Public Health, Key Lab of Public Health Safety of the Ministry of Education,
Fudan University, Shanghai, China; 4G_RIoCE (Research Institute for the Changing Global Environment) and Fudan Tyndall Centre,
Fudan University, Shanghai, China
Background: Although ambient fine particulate matter (PM2.5; particulate matter ≤ 2.5 µm in
aerodynamic diameter) has been linked to adverse human health effects, the chemical constituents
that cause harm are unknown. To our knowledge, the health effects of PM2.5 constituents have not
been reported for a developing country.
oB j e c t i v e s : We examined the short-term association between PM2.5 constituents and daily mortal-
ity in Xi’an, a heavily polluted Chinese city.
Methods: We obtained daily mortality data and daily concentrations of PM2.5, organic carbon
(OC), elemental carbon (EC), and 10 water-soluble ions for 1 January 2004 through 31 December
2008. We also measured concentrations of fifteen elements 1 January 2006 through 31 December
2008. We analyzed the data using over-dispersed generalized linear Poisson models.
results: During the study period, the mean daily average concentration of PM2.5 in Xi’an was
182.2 µg/m3. Major contributors to PM2.5 mass included OC, EC, sulfate, nitrate, and ammonium.
After adjustment for PM2.5 mass, we found significant positive associations of total, cardiovascular,
or respiratory mortality with OC, EC, ammonium, nitrate, chlorine ion, chlorine, and nickel for at
least 1 lag day. Nitrate demonstrated stronger associations with total and cardiovascular mortality
than PM2.5 mass. For a 1-day lag, interquartile range increases in PM2.5 mass and nitrate (114.9
and 15.4 µg/m3, respectively) were associated with 1.8% [95% confidence interval (CI): 0.8%,
2.8%] and 3.8% (95% CI: 1.7%, 5.9%) increases in total mortality.
co n c l u s i o n s : Our findings suggest that PM2.5 constituents from the combustion of fossil fuel may
have an appreciable influence on the health effects attributable to PM2.5 in Xi’an.
ke y words: air pollution, chemical constituents, fine particulate matter, mortality, time-series
studies. Environ Health Perspect 120:373–378 (2012). http://dx.doi.org/10.1289/ehp.1103671
[Online 3 January 2012]
Cao et al.
374
v o l u m e 120 | n u m b e r 3 | March 2012
•
Environmental Health Perspectives
and lead (Pb)] for 1 January 2006 through
31 December 2008 (1,096 days).
e PM2.5 monitoring site was located on
the rooftop of the Chinese Academy of Sciences’
Institute of Earth Environment building in an
urban-scale zone of representation (Chow et al.
2002). The site was surrounded by a residen-
tial area where there were no major industrial
activities nor local fugitive dust sources [see
Supplemental Material, Figure 1 (http://dx.doi.
org/10.1289/ehp.1103671). PM2.5 samples
were obtained 10 m above the ground. Our pre-
vious studies suggest that the measured PM2.5
concentrations at this monitoring station are
representative of the general status of PM2.5 pol-
lution in Xi’an (Cao et al. 2005, 2007, 2009).
Daily PM2.5 samples were collected using
two battery-powered mini-volume samplers
(MiniVol™ TAS; Airmetrics, Eugene, OR,
USA) operating at a flow rate of 5 L/min (Cao
et al. 2003). We used a relatively low flow
rate due to high PM loading in Xi’an. PM2.5
samples were collected on 47-mm Whatman
quartz microfiber filters that were pre-heated
at 900°C for 3 hr before sampling. e quartz-
fiber filters were analyzed gravimetrically for
mass concentrations. We analyzed a 0.5-cm2
punch from each sample for OC and EC using
a Desert Research Institute (DRI) model 2001
thermal/optical carbon analyzer (Atmoslytic
Inc., Calabasas, CA, USA) for eight carbon
fractions following the IMPROVE (Interagency
Monitoring of Protected Visual Environments)
thermal/optical reflectance (TOR) protocol
(Chow et al. 2004). Levels of the five water-
soluble cations (Na+, NH4+, K+, Mg2+ and
Ca2+) and five water-soluble anions (F–, Cl–,
NO2–, SO42– and NO3–) were determined
in aqueous extracts of the sample filters using
an ion chromatograph (Dionex 600; Dionex,
Thermo Fisher Scientific, Inc., Cambridge,
England, UK). Cation concentrations were
determined using a CS12A column (Dionex),
and anions were separated by an AS11-HC
column (Dionex). The elemental concentra-
tions of these samples were then determined by
energy dispersive X-ray fluorescence (ED-XRF)
spectrometry using the PANalytical Epsilon
5 XRF analyzer (PANalytical B.V., Almelo,
the Netherlands). Detailed descriptions of the
sample pretreatment, specific methods, detec-
tion limits, and QA/QC have been discussed
previously (Cao et al. 2003, 2005; Shen et al.
2009a, 2009b).
To adjust for the effect of gaseous pollut-
ants and weather on mortality, we obtained
daily concentrations of sulfur dioxide (SO2)
and nitrogen dioxide (NO2) from the Xi’an
Environmental Monitoring Center, and daily
mean temperature and humidity from the Xi’an
Meteorological Bureau. The SO2 and NO2
concentrations were averaged from the avail-
able monitoring results across seven stations in
our study area. According to the rules of the
Chinese government, we assumed the monitor-
ing data from these stations generally reflected
the background urban air pollution of Xi’an
rather than pollution from local sources.
Statistical methods. Due to different time
periods for measuring PM2.5 constituents,
we constructed two data sets to analyze the
data: e first involved daily measurement of
PM2.5, OC, EC, and ions for 1 January 2004
through 31 December 2008 and the second
included daily concentrations of PM2.5 and
constituent elements for 1 January 2006
through 31 December 2008.
Daily counts of deaths and air pollution
levels were linked by date and analyzed with
time–series analyses (Bell et al. 2004). Because
daily counts of deaths approximate a Poisson
distribution and the relationship between
mortality and explanatory variables is mostly
nonlinear, we used overdispersed generalized
linear Poisson models (quasi-likelihood) with
natural spline (ns) smoothers to analyze mor-
tality, PM2.5 constituents, and covariate data.
In the basic model, we incorporated
smoothed spline functions of time, accom-
modating both nonlinear and nonmonotonic
relations between mortality and time and
thus providing a flexible model to control for
long-term and seasonal trends (Hastie and
Tibshirani 1990). Day of the week (DOW)
was included as a dummy variable (a variable
that takes on the values 1 and 0; also called an
indicator variable) in the basic models. Partial
autocorrelation function (PACF) was used to
guide the selection of degrees of freedom (df)
for the time trend until the absolute values of
the sum of PACF of the residuals for lag days
of up to 30 reached a minimal value (Peng
Table 1. Distribution of daily data on mortality and weather conditions in Xi’an, China (2004–2008).
Percentile
Mean ± SD Minimum 25th 50th 75th Maximum
Daily death counts
Total nonaccidental 26.2 ± 9.7 4.0 20.0 25.0 31.0 128.0
Cardiovascular 12.1 ± 5.7 0.0 8.0 11.0 15.0 39.0
Respiratory 7.2 ± 3.8 0.0 4.0 7.0 9.0 29.0
Injury 1.8 ± 1.7 0.0 1.0 1.0 3.0 19.0
Weather conditions
Temperature (°C) 13.4 ± 9.8 –8.0 5.0 14.0 22.0 32.0
Relative humidity (%) 66.5 ± 16.7 15.0 55.0 68.0 79.0 100.0
Table 2. Descriptive statistics for air pollutants in Xi’an, China (2004–2008).
Observation period Pollutant
Observation
(n)
Mean ± SD
(µg/m3) Minimum Maximum IQR (µg/m3)
PM2.5
mass (%)
1 January 2004–31 December 2008
PM2.5 1,756 182.2 ± 110.1 16.4 768.6 114.9 —
SO21,827 48.4 ± 28.9 8.0 260.0 30.0 —
NO21,827 38.2 ± 15.0 6.4 110.0 21.0 —
OC 1,749 28.3 ± 18.3 5.1 142.3 19.3 15.5
EC 1,749 12.0 ± 8.3 0.2 84.2 8.8 6.6
Na+1,649 2.9 ± 1.4 0.0 12.7 1.9 1.6
NH4+1,538 8.8 ± 8.5 0.0 61.1 10.7 4.8
K+1,616 2.2 ± 2.3 0.0 35.3 1.9 1.2
Mg2+ 1,666 0.5 ± 0.3 0.0 3.7 0.3 0.3
Ca2+ 730 2.0 ± 2.4 0.0 22.4 1.9 1.1
F–1,429 0.6 ± 0.3 0.0 3.4 0.5 0.3
Cl–1,670 5.1 ± 3.5 0.3 32.6 3.6 2.8
NO2–563 0.7 ± 0.4 0.0 3.0 0.4 0.4
SO42– 1,666 31.6 ± 24.4 0.8 198.2 27.8 17.4
NO3–1,644 15.2 ± 12.7 0.0 85.5 15.4 8.4
1 January 2006–31 December 2008
S 1,028 5.1 ± 3.5 0.1 24.8 4.3 2.8
Cl 1,027 1.3 ± 1.6 0.0 11.8 1.5 0.7
K 1,007 1.8 ± 1.7 0.0 22.5 1.6 1.0
Ca 904 2.5 ± 3.3 0.0 30.6 2.3 1.4
Ti 1,026 0.14 ± 0.15 0.00 1.63 0.10 0.08
Cr 952 0.01 ± 0.01 0.00 0.10 0.01 0.01
Mn 1,026 0.11 ± 0.08 0.00 0.56 0.09 0.06
Fe 1,013 1.6 ± 1.7 0.0 20.0 1.3 0.87
Ni 836 0.01 ± 0.03 0.00 0.55 0.01 0.01
Zn 1,028 1.4 ± 1.1 0.0 8.6 1.2 0.79
As 676 0.04 ± 0.03 0.00 0.24 0.03 0.02
Br 962 0.04 ± 0.05 0.00 0.56 0.04 0.02
Mo 1,009 0.06 ± 0.05 0.00 0.37 0.03 0.03
Cd 990 0.03 ± 0.02 0.00 0.13 0.03 0.02
Pb 1,025 0.50 ± 0.38 0.00 3.13 0.41 0.27
PM2.5 constituents and cardiopulmonary mortality
Environmental Health Perspectives
•
v o l u m e 120 | n u m b e r 3 | March 2012
375
et al. 2006; Touloumi et al. 2004, 2006). We
used residual plots and PACF plots to examine
residuals of the basic model for discernable
patterns and autocorrelation.
After establishing the basic model, we intro-
duced the PM2.5 constituents and covariates
(including temperature, humidity, and SO2 and
NO2 concentrations) in the model. Based on
previous literature (Dominici et al. 2006), we
used smoothed spline functions with 3 df (for
the whole period of the study) to control for
temperature and relative humidity. To examine
the temporal relationship of PM2.5 constituents
with mortality, we fitted the models with dif-
ferent lag structures from 0 lag days to 3 lag
days because our previous work on PM2.5 and
daily mortality in China showed little evidence
of a significant association with a lag beyond
3 days (Kan et al. 2007; Ma et al. 2011). A lag
of 0 days (lag 0) corresponds to the current-
day PM2.5, and a lag of 1 day (lag 1) refers to
the previous-day PM2.5. We used the smooth-
ing spline, with 3 df for PM2.5, to graphically
describe its relationships with mortality. We
compared the linear and spline models by com-
puting the difference between the deviances of
the fitted two models (Dominici et al. 2002;
Samoli et al. 2005). We estimated associations
of PM2.5 constituents with mortality before
and after adjustment for PM2.5 mass. Finally,
to examine the robustness of our choice on the
optimal values of df for time trend, we per-
formed a sensitivity analysis to test the impact
of df selection on the regression results.
All analyses were conducted in R version
2.10.1 (http://www.R-project.org) using the
MGCV package. The results are presented
as the percent change in daily mortality per
interquartile range (IQR) increase of pollut-
ant concentrations unless specified otherwise.
Statistical significance was defined as p < 0.05.
Results
We identified 47,838 deaths that occured
between 1 January 2004 and 31 December
2008 in our study population. On average,
26.2 nonaccidental deaths occurred per day,
including 12.1 from cardiovascular diseases and
7.2 from respiratory diseases (Table 1). The
mean daily average temperature and humidity
in Xi’an were 13.4°C and 66.5%, respectively.
During 2004–2008, the Xi’an mean
daily average concentration of PM2.5 was
182.2 µg/m3 (Table 2), which was much
higher than the WHO Global Guidelines
(annual average: 10 µg/m3; WHO 2006) and
than the reported levels of PM2.5 for other
Chinese cities such as Beijing (annual aver-
age: 122 µg/m3; Guo et al. 2009), Shanghai
(annual average: 55 µg/m3; Kan et al. 2007),
and Shenyang (annual average: 75 µg/m3; Ma
et al. 2011). Meanwhile, the mean daily aver-
age concentrations of SO2 and NO2 were 48.4
and 38.2 µg/m3.
Over the 5 years (1,827 days) of the study,
we recorded 1,749 observations of OC and EC;
the averaged concentrations were 28.3 µg/m3
for OC and 12.0 µg/m3 for EC, accounting
for 15.5% and 6.6% of the total PM2.5 mass,
respectively (Table 2). Besides OC and EC,
the other largest contributors to PM2.5 were
SO42– (17.4%), NO3– (8.4%), NH4+ (4.8%),
and S (2.8%).
Generally, moderate to high correlations
(r = 0.5–0.8) were observed for PM2.5 with
OC, EC, S, Cl, K, Mg2+, Cl–, K+, SO42–,
NO3–, and NH4+ levels [see Supplemental
Material, Table 1 (http://dx.doi.org/10.1289/
ehp.1103671)]. PM2.5 was modestly corre-
lated with Na+ levels (r = 0.33). Consistent
with previous studies (Ostro et al. 2007),
Ni levels were weakly correlated with PM2.5
(r = 0.13) and other constituents.
Figure 1 summarizes the quantitative
regression results for single-day lags 0–3 of
PM2.5 mass and various constituents (before
adjusting for PM2.5). We found significant
associations of PM2.5 mass with daily mortality;
an IQR increment in the 1-day lagged concen-
trations of PM2.5 (182.2 µg/m3) corresponded
to a1.8% [95% confidence interval (CI):
0.8%, 2.8%], 3.1% (95% CI: 1.6%, 4.6%),
and 4.5% (95% CI: 2.5%, 6.4%) increase of
total, cardiovascular, and respiratory mortality,
respectively. Consistent with previous studies
(Ito et al. 2011; Ostro et al. 2007; Peng et al.
2009), the effect estimates of PM2.5 constitu-
ents varied by lag structures and mortality out-
comes. OC, EC, NH4+, Cl–, NO3–, Cl, and
Ni showed the strongest associations in that
more than half of the associations assessed were
positive and statistically significant. At least one
positive significant association was found for
Na+, K+, Mg2+, SO42–, S, K, and As. We did
not observe positive significant associations for
F–, Ca, Ti, Cr, Mn, Fe, Zn, Br, Mo, Cd, or Pb
[see Supplemental Material, Figure 2 (http://
dx.doi.org/10.1289/ehp.1103671)].
Figure 2 shows the effect estimates of PM2.5
constituents (OC, EC, NH4+, NO3–, Cl–, Cl,
Figure 1. Estimated percent increases [mean (95% CI)] in total, cardiovascular, and respiratory mortality per
IQR increase in pollutant concentrations on the current day (lag 0) or the previous 1–3 days (lags 1, 2, and 3),
adjusted for temporal trend, day of the week, temperature, relative humidity, and SO2 and NO2 concentrations.
Percent increase Percent increase Percent increase
Total Cardiovascular Respiratory
–4.0 1.0 6.0 –4.0 –4.01.0 1.06.0 6.0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
PM2.5
OC
EC
Na+
K+
Cl–
S
Cl
K
Ni
As
SO42–
NO3–
Mg2+
NH4+
Cao et al.
376
v o l u m e 120 | n u m b e r 3 | March 2012
•
Environmental Health Perspectives
and Ni) that were significantly associated with
at least one outcome and lag period after further
adjustment for PM2.5 mass. OC and EC were
positively associated with cardiovascular and
respiratory mortality (for lags 1–3 and lag 3,
respectively), but were not clearly associated with
total mortality. NH4+ and NO3– were signifi-
cantly associated with total and cardio vascular
mortality, but not with respiratory mortality.
Cl–, Cl, and Ni were significantly associated
with all three mortality outcomes for at least one
lagged exposure. It should be noted that NH4+
(lag 3) and Cl– (lag 1) were negatively and statis-
tically significantly associated with cardiovascular
or respiratory mortality. Na+, K+, Mg2+, SO42–,
S, K, and As, after adjustment for PM2.5, were
no longer positively and statistically significantly
associated with any of the outcomes, and some
of the adjusted associations even became nega-
tive and statistically significant [see Supplemental
Material, Figure 3 (http://dx.doi.org/10.1289/
ehp.1103671)]. Interestingly, after adjusting
for PM2.5, associations with an IQR increase
in NO3– were stronger than associations with
an IQR increase in PM2.5 mass for total and
cardiovascular mortality. For instance, for lag 1,
an IQR increase in NO3– (15.2 µg/m3) was
associated with 3.8% (95% CI: 1.7%, 5.9%)
increase in total mortality, compared with 1.8%
(95% CI: 0.8%, 2.8%) for an IQR increase
(182.2 µg/m3) in PM2.5 mass.
Figure 3 shows the exposure– response rela-
tionships for PM2.5 mass (single day lag 1) with
total, cardiovascular, and respiratory mortality
between 2004 and 2008 in Xi’an. For all three
mortality outcomes, we observed almost lin-
ear relationships, with no evidence of obvious
threshold concentrations below which PM2.5
had no effect on mortality outcomes. e dif-
ferences in the deviance between the linear and
spline models did not indicate a significant
improvement in the fit of the spline versus lin-
ear models. In the linear models, a 10-µg/m3
increment in the 1-day lagged PM2.5 was asso-
ciated with 0.2% (95% CI: 0.1%, 0.3%), 0.3%
(95% CI: 0.1%, 0.4%), and 0.4% (95% CI:
0.2%, 0.6%) increases in total, cardiovascular,
and respiratory mortality, respectively.
As expected, deaths due to injury were
not associated with PM2.5 constituents [there
was only 1 significant association out of 92
comparisons when adjusted for PM2.5; see
Supplemental Material, Table 2 (http://dx.doi.
org/10.1289/ehp.1103671)]. Altering the df
per year for time trend within a range of 3–10
df did not substantially change the regression
results (data not shown).
Discussion
Evidence obtained in this time–series analysis
showed that PM2.5 mass and several constitu-
ents were associated with total nonaccidental
and cardiopulmonary disease-related mortality
in Xi’an. e observed levels of PM2.5 and its
constituents in our study were much higher
than earlier health studies of PM2.5 constitu-
ents in developed countries. Several constitu-
ents that were associated with mortality (NH4+,
NO3–, Cl–, OC, EC, Cl) are associated with
the combustion of fossil fuels such as coal and
heavy oil in Xi’an (Cao et al. 2005, 2009). We
found stronger associations for NO3– with total
and cardiovascular mortality than for PM2.5
mass. We did not find evidence of threshold
concentrations below which PM2.5 was not
associated with mortality in Xi’an. To our
knowledge, this is the first study of its kind in
a developing country to investigate the health
effects of PM2.5 constituents.
Figure 2. Estimated percent increases [mean (95% CI)] in total, cardiovascular, and respiratory mortality
per IQR increase in pollutant concentrations on the current day (lag 0) or the previous 1–3 days (lags 1, 2,
and 3), adjusted for PM2.5 mass, temporal trend, day of the week, temperature, relative humidity, and SO2
and NO2 concentrations.
Percent increase Percent increase Percent increase
Total Cardiovascular Respiratory
–8.0 –8.0 –8.08.0 8.0 8.0000
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
3
2
1
0
OC
EC
Cl
Cl–
Ni
NO3–
NH4+
Figure 3. Exposure–response relationships (smoothing plots) of PM2.5 against total (A), cardiovascular (B), and respiratory (C) mortality (df = 3) in Xi’an, adjusted for
temporal trend, day of the week, temperature, relative humidity, and SO2 and NO2 concentrations. The x-axis is the PM2.5 concentrations (single day lag, L1); the
y-axis is the estimated percent change in deaths; the solid blue lines indicate the estimated mean percent change in daily death numbers using the lowest PM2.5
concentration as the reference level; and the dashed lines represent the 95% CI.
0 200 400 600 800 0 200 400 600 800 0 200 400 600 800
PM2.5 (µg/m3)PM2.5 (µg/m3)PM2.5 (µg/m3)
40
30
20
10
0
–10
50
40
30
20
10
0
–10
50
40
30
20
10
0
–10
60
40
20
0
–20
Percent change in
total deaths (%)
Percent change in
cardiovascular deaths (%)
Percent change in
respiratory deaths (%)
PM2.5 constituents and cardiopulmonary mortality
Environmental Health Perspectives
•
v o l u m e 120 | n u m b e r 3 | March 2012
377
e results of our study in Xi’an indicate
considerable risk heterogeneity among the
various PM2.5 constituents. Consistent with
previous epidemiological studies on PM con-
stituents (Ito et al. 2011; Laden et al. 2000;
Ostro et al. 2007, 2008; Peng et al. 2009;
Zhou et al. 2011), we found that PM2.5 con-
stituents resulting from the combustion of fos-
sil fuel (e.g., NH4+, NO3–, Cl–, OC, EC, Cl,
Ni) maintained significant positive associations
with mortality outcomes even after we adjusted
for PM2.5. In contrast, we did not find sig-
nificant associations between mortality and
common crustal elements (e.g., Ca and K) in
Xi’an, which is consistent with a previous study
performed in six U.S. cities that showed PM2.5
crustal particles were not associated with daily
mortality (Laden et al. 2000). It should be
noted that we observed statistically significant
associations for some, but not all, lag struc-
tures of PM2.5 constituents. Further research
is needed to clarify relationships between the
timing of exposures and their potential health
effects.
Our analysis indicates positive associa-
tions of cardiopulmonary mortality with IQR
increases in OC or EC during the previous
1–3 days even after adjusting for PM2.5 mass.
is is consistent with the findings of a meta-
analysis of short-term exposure time–series
studies of EC and daily mortality that reported
positive associations with cardiopulmonary
mortality (Smith et al. 2009). e results of a
recent cohort study in California suggest that
long-term exposure to OC also increase car-
diopulmonary mortality (Ostro et al. 2010).
Additionally, several previous studies support
the biological plausibility of a link between
exposure to OC or EC and exacerbations of
cardiopulmonary diseases (Gold et al. 2005;
Henneberger et al. 2005; Jansen et al. 2005;
Lanki et al. 2006; Lewne et al. 2007; Mar et al.
2005; Shih et al. 2008; von Klot et al. 2009).
For example, one study in Germany examined
weekly electrocardiograms of 56 men with a
history of heart disease and found significant
associations of OC or EC with changes in myo-
cardial repolarization, which could increase the
risk of sudden cardiac death (Henneberger et al.
2005). Gold et al. (2005) found associations of
EC with ST-segment depression among a panel
of 24 elderly Boston residents. Similarly, Lanki
et al. (2006) examined the health effects of five
PM2.5 components (Si, S, Ni, Cl, and EC), and
found only EC had significant association with
ST-segment depression in multipolluant models.
Exposure to OC or EC was also associated with
increased nitric oxide (NO) in exhaled breath,
a marker of airway inflammation (Mar et al.
2005). Thus, exposures to both OC and EC
are associated with a number of indicators that
could contribute to cardio pulmonary mortality.
NO3– was positively associated with
mortality in our study. To date, only a few
epidemiological studies have examined the
relationships of NO3– with mortality, and
their findings were inconclusive. For example,
Klemm et al. (2004) found a positive but insig-
nificant association between NO3– and mortal-
ity in Atlanta (Georgia), whereas Ostro et al.
(2007) found a significant association between
NO3– and mortality in six California counties.
More studies are needed to understand the
health effects of NO3–. In our study, SO42–
(mean level: 31.6 µg/m3) was not associated
with mortality, which is consistent with toxico-
logical studies showing little toxic evidence of
SO42– effects on the cardiopulmonary system
at typical environmental concentrations (Reiss
et al. 2007). As Schlesinger and Cassee (2003)
pointed out, the minimal effective concentra-
tion of SO42– to alter pulmonary mechanical
function in normal humans following acute
exposure is > 1,000 µg/m3.
In our analysis, an IQR increase of
0.01 µg/m3 in 1-day lagged Ni was associated
with 0.4% (95% CI: 0.0%, 0.8%), 0.6%
(95% CI: –0.1%, 1.2%) and 0.9% (95% CI:
0.2%, 1.7%) increases in total, cardiovascular,
and respiratory mortality. As a transition metal,
Ni may affect health by producing reactive
oxygen species and increasing oxidative stress
(Lippmann et al. 2006; Schlesinger et al. 2006).
In fact, existing epidemiological studies provide
evidence of adverse effects for several transition
metals (Dominici et al. 2007; Huang et al.
2003; Lippmann et al. 2006; Ostro et al. 2007,
2008). For example, Huang et al. (2003) found
that exposure to a factor including vanadium
(V), Zn, and copper (Cu) from concentrated
ambient particles was associated with increased
blood fibrinogen levels. Using the National
Morbidity, Mortality, and Air Pollution Study
(NMMAPS) database, Lippmann et al. (2006)
found that daily mortality rates in the 60 U.S.
cities with speciation data were significantly
associated with average levels of Ni and V, but
not other measured species. In Xi’an, the major
source of Ni in PM2.5 is fossil fuel combustion,
especially heavy oil (Shen et al. 2009b). e
role of Ni in PM2.5 health hazards should be
investigated further.
In our analysis, a 10-µg/m3 increment in
the 1-day lagged concentrations of PM2.5 was
associated with 0.2% (95% CI: 0.1%, 0.3%),
0.3% (95% CI: 0.1%, 0.4%), and 0.4% (95%
CI: 0.2%, 0.6%) increases in total, cardiovas-
cular, and respiratory mortality, respectively.
Compared with studies of PM2.5 and daily
mortality in developed countries (Franklin
et al. 2007; Ostro et al. 2006; Ueda et al. 2009;
Zanobetti and Schwartz 2009), our estimations
of the associations of PM2.5 with mortality were
somewhat lower in magnitude per amount of
PM2.5 mass. For example, a multicity analy-
sis in 112 U.S. cities found that a 10-µg/m3
increase in PM2.5 was associated with a 1.0%
increase in total mortality, a 0.9% increase in
cardiovascular mortality, and a 1.7% increase in
respiratory mortality (Zanobetti and Schwartz
2009), whereas our findings are in agreement
with earlier evidence (Aunan and Pan 2004)
suggesting weaker associations between health
outcomes and unit increases in air pollution
exposures in China than in developed coun-
tries. is may be explained by differences in
the composition and toxicity of PM, as well
as differences in local PM concentrations and
population sensitivity to PM in addition to dif-
ferences in age structure and other population
characteristics. Lower risks of death per unit
increases in pollutants when concentrations
are high may reflect the selective attrition of
vulnerable members of the population who die
before concentrations reach the maximum level
(Wong et al. 2008). Also, associations between
mortality and PM exposures ranging from low
(e.g., exposure levels associated with ambient
air pollution) to high (e.g., exposure levels asso-
ciated with cigarette smoking) concentrations
suggest that the exposure–response curve of
PM often tends to become flat at higher con-
centrations (Pope et al. 2009).
Accurate information on the shape of
exposure–response relationships is crucial for
public health assessment, and the demand for
providing the relevant curves has been grow-
ing (Dominici et al. 2002). Dose–response
relationships may vary by location depend-
ing on factors such as the air pollution mix-
ture, climate, and overall health of the studied
population (Samoli et al. 2005). In our study
population, we did not observe evidence for
a threshold concentration below which PM2.5
was not associated with mortality, suggest-
ing that linear models without a threshold are
appropriate for assessing the effect of PM2.5 on
daily mortality for the high-exposure settings
typical of developing countries.
Our study has limitations. First, we evalu-
ated the associations of multiple constituents
and lags with three different mortality out-
comes; some significant associations, therefore,
may have occurred by chance. Second, because
of moderate-to-high collinearity among PM2.5
constituents, we could not adjust for multiple
exposures, and some associations may reflect
the effects of other correlated components. We
did not measure several elements such as sele-
nium (Se), V, and silicon (Si), although previ-
ous studies reported significant associations
between these elements and adverse health out-
comes (Laden et al. 2000; Ostro et al. 2007),
and we could not evaluate ozone (O3) due to
a lack of monitoring data in Xi’an. As in many
previous time–series studies, we used PM2.5
monitoring results from a fixed station as a
proxy measure for population exposures to air
pollution. As a result, a number of issues may
arise given that ambient monitoring results
differ from personal exposure level to air pol-
lutants (Sarnat et al. 2001, 2005). In addition,
Cao et al.
378
v o l u m e 120 | n u m b e r 3 | March 2012
•
Environmental Health Perspectives
variation in the extent of exposure misclassi-
fi cation among individual constituents may
influence associations. Finally, we did not con-
duct formal source apportionment of PM2.5
constituents, and therefore cannot identify the
source components that contributed most to
the associations between PM2.5 and mortality.
Conclusions
Our findings suggest that PM2.5 constitu-
ents from fossil fuel combustion may have
an appreciable influence on the health effects
attributable to PM2.5. Associations of PM2.5
with mortality in Xi’an are somewhat lower in
magnitude per amount of PM2.5 mass com-
pared with associations reported for popula-
tions in developed countries. Our findings
add support to previously reported evidence of
PM2.5-related health effects in China and sug-
gest that combustion-associated pollutants are
particularly important.
Re f e R e n c e s
Aunan K, Pan XC. 2004. Exposure–response functions for health
effects of ambient air pollution applicable for China—a
meta-analysis. Sci Total Environ 329(1–3):3–16.
Bell ML, Samet JM, Dominici F. 2004. Time-series studies of
particulate matter. Annu Rev Public Health 25:247–280.
Brunekreef B, Holgate ST. 2002. Air pollution and health. Lancet
360(9341):1233–1242.
Cao JJ, Chow J, Lee S, Li Y, Chen S, An Z, et al. 2005.
Characterization and source apportionment of atmospheric
organic and elemental carbon during fall and winter of 2003
in Xi’an, China. Atmos Chem Phys 5:3127–3137.
Cao JJ, Lee SC, Chow JC, Watson JG, Ho KF, Zhang RJ, et al.
2007. Spatial and seasonal distributions of carbona-
ceous aerosols over China. J Geophys Res 112(D22S11);
doi:10.1029/2006JD008205 [Online 10 November 2007].
Cao JJ, Lee S, Ho K, Zhang X, Zou S, Fung K, et al. 2003.
Characteristics of carbonaceous aerosol in Pearl River Delta
Region, China during 2001 winter period. Atmos Environ
37(11):1451–1460.
Cao JJ, Zhu CS, Chow JC, Watson JG, Han YM, Wang GH,
et al. 2009. Black carbon relationships with emissions and
meteorology in Xi’an, China. Atmos Res 94(2):194–202.
Chinese Ministry of Environmental Protection. 2010. Proposed
ambient air quality standards Beijing [in Chinese].
Available: http://www.zhb.gov.cn/gkml/hbb/bgth/201011/
W020101130374443014849.pdf [accessed 24 December 2011].
Chow JC, Engelbrecht JP, Watson JG, Wilson WE, Frank NH,
Zhu T. 2002. Designing monitoring networks to represent
outdoor human exposure. Chemosphere 49(9):961–978.
Chow JC, Watson JG, Chen LW, Arnott WP, Moosmuller H,
Fung K. 2004. Equivalence of elemental carbon by thermal/
optical reflectance and transmittance with different tem-
perature protocols. Environ Sci Technol 38(16):4414–4422.
Dominici F, Daniels M, Zeger SL, Samet JM. 2002. Air pollution
and mortality: estimating regional and national dose–
response relationships. J Am Stat Assoc 97(457):100–111.
Dominici F, Peng RD, Bell ML, Pham L, McDermott A, Zeger SL,
et al. 2006. Fine particulate air pollution and hospital
admission for cardiovascular and respiratory diseases.
JAMA 295(10):1127–1134.
Dominici F, Peng RD, Ebisu K, Zeger SL, Samet JM, Bell ML.
2007. Does the effect of PM10 on mortality depend on
PM nickel and vanadium content? A reanalysis of the
NMMAPS data. Environ Health Perspect 115:1701–1703.
Franklin M, Zeka A, Schwartz J. 2007. Association between
PM2.5 and all-cause and specific-cause mortality in 27 US
communities. J Expo Sci Environ Epidemiol 17(3):279–287.
Gold DR, Litonjua AA, Zanobetti A, Coull BA, Schwartz J,
MacCallum G, et al. 2005. Air pollution and ST-segment
depression in elderly subjects. Environ Health Perspect
113:883–887.
Guo Y, Jia Y, Pan X, Liu L, Wichmann HE. 2009. The asso-
ciation between fine particulate air pollution and hospital
emer gency room visit s for card iovascu lar d isease s in
Beijing, China. Sci Total Environ 407(17):4826–4830.
Hastie TJ, Tibshirani RJ. 1990. Generalized Additive Models.
London:Chapman & Hall.
Henneberger A, Zareba W, Ibald-Mulli A, Ruckerl R, Cyrys J,
Couderc JP, et al. 2005. Repolarization changes induced
by air pollution in ischemic heart disease patients. Environ
Health Perspect 113:440–446.
Huang YC, Ghio AJ, Stonehuerner J, McGee J, Carter JD,
Grambow SC, et al. 2003. The role of soluble components
in ambient fine particles-induced changes in human lungs
and blood. Inhal Toxicol 15(4):327–342.
Ito K, Mathes R, Ross Z, Nadas A, Thurston G, Matte T. 2011.
Fine particulate matter constituents associated with car-
diovascular hospitalizations and mortality in New York
City. Environ Health Perspect 119:467–473.
Jansen K, Larson T, Koenig J, Mar T, Fields C, Stewart J, et al.
2005. Associations between health effects and particulate
matter and black carbon in subjects with respiratory dis-
ease. Environ Health Perspect 113:1741–1746.
Kan H, London SJ, Chen G, Zhang Y, Song G, Zhao N, et al.
2007. Differentiating the effects of fine and coarse par-
ticles on daily mortality in Shanghai, China. Environ Int
33(3):376–384.
Klemm RJ, Lipfert FW, Wyzga RE, Gust C. 2004. Daily mortality
and air pollution in Atlanta: two years of data from ARIES.
Inhal Toxicol 16(suppl 1):131–141.
Laden F, Neas LM, Dockery DW, Schwartz J. 2000. Association of
fine particulate matter from different sources with daily mor-
tality in six U.S. cities. Environ Health Perspect 108:941–947.
Lanki T, de Hartog JJ, Heinrich J, Hoek G, Janssen NA,
Peters A, et al. 2006. Can we identify sources of fine par-
ticles responsible for exercise-induced ischemia on days
with elevated air pollution? The ULTRA study. Environ
Health Perspect 114:655–660.
Lewne M, Plato N, Gustavsson P. 2007. Exposure to particles,
elemental carbon and nitrogen dioxide in workers exposed
to motor exhaust. Ann Occup Hyg 51(8):693–701.
Lippmann M, Ito K, Hwang JS, Maciejczyk P, Chen LC. 2006.
Cardiovascular effects of nickel in ambient air. Environ
Health Perspect 114:1662–1669.
Ma Y, Chen R, Pan G, Xu X, Song W, Chen B, et al. 2011. Fine
particulate air pollution and daily mortality in Shenyang,
China. Sci Total Environ, 409:2473–2477.
Mar TF, Jansen K, Shepherd K, Lumley T, Larson TV, Koenig JQ.
2005. Exhaled nitric oxide in children with asthma and short-
term PM2.5 exposure in Seattle. Environ Health Perspect
113:1791–1794.
National Research Council. 1998. Research Priorities for
Airborne Particulate Matter. Washington DC:National
Academy Press.
Ostro B, Broadwin R, Green S, Feng WY, Lipsett M. 2006. Fine par-
ticulate air pollution and mortality in nine California counties:
results from CALFINE. Environ Health Perspect 114:29–33.
Ostro B, Feng WY, Broadwin R, Green S, Lipsett M. 2007. The
effects of components of fine particulate air pollution
on mortality in California: results from CALFINE. Environ
Health Perspect 115:13–19.
Ostro B, Feng WY, Broadwin R, Malig BJ, Green RS, Lipsett MJ.
2008. The impact of components of fine particulate matter
on cardiovascular mortality in susceptible subpopulations.
Occup Environ Med 65(11):750–756.
Ostro B, Lipsett M, Reynolds P, Goldberg D, Hertz A, Garcia C,
et al. 2010. Long-term exposure to constituents of fine par-
ticulate air pollution and mortality: results from the California
teachers study. Environ Health Perspect 118:363–369.
Peng RD, Bell ML, Geyh AS, McDermott A, Zeger SL, Samet JM,
et al. 2009. Emergency admissions for cardiovascular and
respiratory diseases and the chemical composition of fine
particle air pollution. Environ Health Perspect 117:957–963.
Peng RD, Dominici F, Louis TA. 2006. Model choice in time
series studies of air pollution and mortality. Journal of the
Royal Statistical Society, Series A 169(2):179–203.
Pope CA III, Burnett RT, Krewski D, Jer rett M, Shi Y, Calle EE,
et al. 2009. Cardiovascular mortality and exposure to airborne
fine particulate matter and cigarette smoke: shape of the
exposure–response relationship. Circulation 120(11):941–948.
Pope CA III, Dockery DW. 2006. Health effects of fine particu-
late air pollution: lines that connect. J Air Waste Manage
Assoc 56(6):709–742.
Reiss R, Anderson EL, Cross CE, Hidy G, Hoel D, McClellan R, et al.
2007. Evidence of health impacts of sulfate- and nitrate-con-
taining particles in ambient air. Inhal Toxicol 19(5):419–449.
Samoli E, Analitis A, Touloumi G, Schwartz J, Anderson HR,
Sunyer J, et al. 2005. Estimating the exposure–response rela-
tionships between particulate matter and mortality within the
APHEA multicity project. Environ Health Perspect 113:88–95.
Sarnat JA, Brown KW, Schwartz J, Coull BA, Koutrakis P. 2005.
Ambient gas concentrations and personal particulate
matter exposures—implications for studying the health
effects of particles. Epidemiology 16(3):385–395.
Sarnat JA, Schwartz J, Catalano PJ, Suh HH. 2001. Gaseous
pollutants in particulate matter epidemiology: confounders
or surrogates? Environ Health Perspect 109:1053–1061.
Schlesinger RB, Cassee F. 2003. Atmospheric secondary inorganic
particulate matter: the toxicological perspective as a basis for
health effects risk assessment. Inhal Toxicol 15(3):197–235.
Schlesinger RB, Kunzli N, Hidy GM, Gotschi T, Jerrett M. 2006.
The health relevance of ambient particulate matter char-
acteristics: coherence of toxicological and epidemiologi-
cal inferences. Inhal Toxicol 18(2):95–125.
Shen Z, Cao J, Arimoto R, Han Z, Zhang R, Han Y, et al. 2009a.
Ionic composition of TSP and PM2.5 during dust storms
and air pollution episodes at Xi’an, China. Atmos Environ
43(18):2911–2918.
Shen Z, Cao J, Tong Z, Liu S, Reddy L, Han Y, et al. 2009b.
Chemical characteristics of submicron particles in winter
in Xi’an. Aerosol Air Qual Res 9(1):80–93.
Shih TS, Lai CH, Hung HF, Ku SY, Tsai PJ, Yang T, et al. 2008.
Elemental and organic carbon exposure in highway toll-
booths: a study of Taiwanese toll station workers. Sci Total
Environ 402(2-3):163–170.
Smith KR, Jerrett M, Anderson HR, Burnett RT, Stone V, Derwent R,
et al. 2009. Public health benefits of strategies to reduce
greenhouse-gas emissions: health implications of short-lived
greenhouse pollutants. Lancet 374(9707):2091–2103.
Thurston GD, Ito K, Mar T, Christensen WF, Eatough DJ,
Henry RC, et al. 2005. Workgroup report: workshop on
source apportionment of particulate matter health effects—
intercomparison of results and implications. Environ Health
Perspect 113:1768–1774.
Touloumi G, Atkinson R, Tertre AL, Samoli E, Schwartz J,
Schindler C, et al. 2004. Analysis of health outcome time
series data in epidemiological studies. Environmetrics
15(2):101–117.
Touloumi G, Samoli E, Pipikou M, Le Tertre A, Atkinson R,
Katsouyanni K. 2006. Seasonal confounding in air pollution
and health time-series studies: effect on air pollution effect
estimates. Stat Med 25(24):4164–4178.
Ueda K, Nitta H, Ono M, Takeuchi A. 2009. Estimating mortality
effects of fine particulate matter in Japan: a comparison
of time-series and case-crossover analyses. J Air Waste
Manag Assoc 59(10):1212–1218.
van Donkelaar A, Martin RV, Brauer M, Kahn R, Levy R,
Verduzco C, et al. 2010. Global estimates of ambient fine
particulate matter concentrations from satellite-based
aerosol optical depth: development and application. Environ
Health Perspect 118:847–855.
Venners SA, Wang B, Peng Z, Xu Y, Wang L, Xu X. 2003.
Particulate matter, sulfur dioxide, and daily mortality in
Chongqing, China. Environ Health Perspect 111:562–567.
von Klot S, Gryparis A, Tonne C, Yanosky J, Coull BA, Goldberg RJ,
et al. 2009. Elemental carbon exposure at residence and
survival after acute myocardial infarction. Epidemiology
20(4):547–554.
WHO (World Health Organization). 1992. International
Statistical Classification of Diseases and Related Health
Problems 10th Revision (ICD-10). Geneva. World Health
Organization. Available: http://apps.who.int/classifications/
icd10/browse/2010/en [accessed 21 January 2012].
WHO (World Health Organization). 2006. WHO Air Quality
Guidelines for Particulate Matter, Ozone, Nitrogen
Dioxide and Sulfur Dioxide. Global Update 2005. Geneva,
Switzerland:World Health Organization. Available: whqlib-
doc.who.int/hq/2006/WHO_SDE_PHE_OEH_06.02_eng.pdf
[accessed 21 January 2012].
Wong CM, Vichit-Vadakan N, Kan H, Qian Z. 2008. Public health
and air pollution in Asia (PAPA): a multicity study of short-
term effects of air pollution on mortality. Environ Health
Perspect 116:1195–1202.
Zanobetti A, Schwartz J. 2009. The effect of fine and coarse
particulate air pollution on mortality: a national analysis.
Environ Health Perspect 117:898–903.
Zhou J, Ito K, Lall R, Lippmann M, Thurston G. 2011. Time-series
analysis of mortality effects of fine particulate matter com-
ponents in Detroit and Seattle. Environ Health Perspect
119:461–466.
