Journal of Exposure Analysis and Environmental Epidemiology (2004) 14, 323–329
r 2004 Nature Publishing Group All rights reserved 1053-4245/04/$30.00
www.nature.com/jea
Personal exposure to particulate matter less than 2.5 lm in Mexico City: a
pilot study
MAITE VALLEJO,a CLAUDIA LERMA,a OSCAR INFANTE,a ANTONIO G. HERMOSILLO,a
HORACIO RIOJAS-RODRIGUEZb AND MANUEL CÁRDENASa
a
Instituto Nacional de Cardiologı´a ‘‘Ignacio Chávez’’, Me´xico City, Mexico
Instituto Nacional de Salud Pública, Cuernavaca Morelos, Mexico
b
This study was aimed to describe the personal exposure of permanent residents in Mexico City’s Metropolitan Area (MCMA) to particulate matter of less
than 2.5 mm diameter (PM2.5) during their daily activities. A total of 40 healthy volunteers (30 women and 10 men) with sedentary activities were
included. All of them carried a PM2.5 personal monitor during 13 h and registered their activities in a written diary that classified them in indoor and
outdoor microenvironments in each 30 min period. All sample collections started at 0900 hours, and even though measurements were obtained during the
rainy season (April–August 2002), the relative humidity was less than 70%. The data were categorized and evaluated under the following criteria:
morning and afternoon exposure, indoor and outdoor activities, and geographical location. The descriptive analysis showed that the overall outdoor
median concentration of PM2.5 (89.50 mg/m3) was higher than the indoor one (67.55 mg/m3). PM2.5 concentrations in the morning to early afternoon were
more elevated than in the late afternoon, suggesting a circadian-like behavior. In the indoor microenvironment, the highest concentration occurred in the
subway (106.2 mg/m3) followed by school (93.27 mg/m3), and the lowest at home (53.1 mg/m3). The outdoor microenvironment with the highest
concentrations was the public transportation (bus) (99.95 mg/m3), while the automobile had the lowest (64.9 mg/m3). The geographical zone with the
highest concentration was the Center city area (87.87 mg/m3), and the one with the lowest concentration was the northeast area of the city (50 mg/m3). All
the differences were statistically significant (Po0.05). Multivariate analysis corroborated that PM2.5 concentrations are mainly determined by
geographical locations and hour of the day, but not by the type of microenvironment. The inclusion of covariables in the multivariable analysis ensures a
more accurate estimation and prediction of the real PM2.5 concentrations. In conclusion, PM2.5 personal exposure of healthy adult permanent residents of
MCMA is usually higher than recommended by the international standards in outdoor and even in indoor microenvironments. Particulate matter
personal exposure varies in relation to hour of the day, daily activities and microenvironments.
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14, 323–329. doi:10.1038/sj.jea.7500328
Keywords: Mexico City, PM2.5, air pollution, personal exposure.
Introduction
The Mexico City Metropolitan Area (MCMA) is one of the
largest urban concentrations in the world with geographical
characteristics that worsen pollution problems. The city is
placed in a hydrological basin at 2240 m above sea level in its
central part. It has a territorial extension of 1200 km2 and is
surrounded by mountains with an average altitude of 1000 m.
These characteristics cause a reduction of 23% in oxygen
concentration in relation to the sea level, which in turn
diminishes the efficiency of combustion, raising the concentration of carbon monoxide and hydrocarbons, and favoring
the transformation of primary pollutants into ozone
and other oxidants. The prevailing winds that flow from
1. Address all correspondence to: Maite Vallejo, Instituto Nacional de
Cardiologı́a Ignacio Chávez, Juan Bandiano No.1, Col. Sección XVI,
14080, Tlalpan D.F. México. Tel.: þ 55-732911 exts 1223 and 1357.
Fax: þ 55-730629.
E-mail: [email protected]
Received 28 May 2003; accepted 28 October 2003
northeast to southwest cause a greater pollutant concentration in the southern area of the city, in which the Ajusco
mountains block its dissipation (Figure 1). In Mexico City,
two climatic periods occur during the year: the winter dry
season from November to April in which thermal inversions
are frequent, and the rainy period from May to October.
Finally, anticyclonic systems, generated in the Mexican Gulf
and in the Pacific Sea, prevent air mixture that causes
atmospheric stability (Ponciano-Rodrı́guez and RiveroSerrano, 1996; Onursal and Gautam, 1997).
The MCMA has been divided, for monitoring pollution
purposes, into five areas: center, northeast, northwest,
southeast and southwest. Air pollutants are measured
hourly every day by an environmental monitor network that
has 32 stations, nine in the northeast, nine in the northwest,
four in the southeast, five in the southwest, and five in
the center. These stations have manual and automatic
monitor systems that measure criterion pollutants such as
ozone, carbon monoxide, lead, nitrates, sulfur dioxide,
PM10, and total particulate matters (http://www.sma.df.gob.
mx/-subject: publicaciones, aire, compendios-). Since 1986,
PM2.5 personal exposure
Vallejo et al.
Methods
Population
A total of 40 volunteers, aged 21–40 years, were included in
this study. All participants were nonsmokers and residents
within MCMA as shown in Figure 1. In all, 30 participants
were women and 10 men. Their daily activities were
sedentary: 19 were health professionals (researchers, nurses,
and physicians), 14 were college students, four were housekeepers, and three were office workers.
The Institutional Ethics and Research Committees approved this study and all subjects signed an informed
consent.
Figure 1. Geographical areas and locations of home and work of
participants within MCMA. Black dots indicate the work location of
participants and white ones home location of participants. Dotted line
identifies Distrito Federal political limits. Black line indicates the
geographical division of MCMA.
the Metropolitan Air Quality Index (Indice Metropolitano
de la Calidad del Aire (IMECA)) was established in order to
evaluate air pollution. This index measures air pollution
concentration in parts per million or micrograms per cubic
meter depending on the pollutant being measured on a scale
of 0–500 points, in which a value of 100 corresponds to the
maximum value considered harmless to health and accepted
by air quality standards (Boletı́n Terapéutico, 1992).
At the beginning of the year 2003, the air quality standards
for particulate matter less than 10 mm were reviewed and
modified. PM2.5 standards are in the process of evaluation
and no measurements have begun (Diario Oficial de la
Federación, 2002).
Most health effects studies on exposure to particulate
matter air pollution have been carried out during short
periods of time and using ambient monitors (Dockery
et al., 1993; Pope et al., 1995; Saldivar et al., 1995; Ostro
et al., 1996; Simpson et al., 1997; Brauer et al., 2000;
Levy et al., 2000). Studies carried out in Mexico City were
also performed with ambient monitors, and in some cases
with monitors from the environmental monitors network
(Romieu et al., 1992, 1996; Borja-Aburto et al., 1997; Gold
et al., 1999; Loomis et al., 1999).
There are no published data of PM2.5 personal exposure of
general population exposed in their microenvironments
during their daily activities in Mexico City. The purpose of
this pilot study was to gather data about the PM2.5 personal
exposure of Mexico City healthy residents during their daily
activities, in order to characterize the exposure patterns that
could be of interest for further studies.
324
Particulate Matter Measurement
The participants received and carried a personal PM2.5.
monitor during a single 13-h starting at 0900 hours. No
measurements were made during weekends. Each individual
completed a written diary that classified their daily activities
in indoor and outdoor microenvironments in a preset time
period resolution of 30 min. Indoor situations included
activities at home, at work, at school, or in public places
such as theaters, stores, restaurants, coffee shops, and
subway transportation. If participants were walking, standing, or sitting in an open space, or if they were driving a car,
or using public transportation (bus or taxi), activities were
classified as outdoor. The written diary considered separately
the periods when participants were cooking or near a smoker.
Participants were instructed to classify periods as indoor and
outdoor exposure according to the microenvironment where
they expended most of the time of each 30-min period.
Microenvironments were not preassigned, as each individual
was instructed to perform his/her usual activities. The study
was carried out during the rainy season (April–August 2002).
Exposure to PM2.5 was measured using the pDR
nephelometric method (personal DataRAM, pDR1200,
MIE Inc., Bedford, MA, 2000). The equipment was supplied
by the Instituto de Salud Ambiente y Trabajo (ISAT). It
consists of a real-time active system that was connected to a
pump with suction flux set at 4 l/min. The pump introduces
particles into a centrifuge where they are separated according
to their aerodynamic diameter by gravity. The particles are
measured inside a sensor chamber. The equipment was
calibrated previously according to the manufacturer’s
instructions. The sensitivity of this technique allows to detect
concentrations up to 0.001 mg/m3. Each minute concentration was averaged in 30-min periods to relate it to the preset
time periods of the written diary. Since pDR measurements
can be biased if the relative humidity (RH) is more than 70%
(Liu et al., 2002), data with regard to this matter were
obtained from the environmental monitor network for each
specific day in which PM2.5 measurements were performed.
The average daily RH was below 70% on the days in which
measurements were made.
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(4)
PM2.5 personal exposure
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(4)
Descriptive Analysis of PM2.5 Exposure
The time series of PM2.5 median concentrations during the
entire recording at indoor and outdoor microenvironments
are shown in Figure 2. Outdoor levels of PM2.5 were
significantly higher (89.1 mg/m3, 54.9–123.1) than indoor
levels (70.8 mg/m3, 47.4–106.8) (Po0.05), even though the
outdoor PM2.5 median concentration at outdoor microenvironments was lower than the indoor ones at 1100, 1830, and
1900 hours. The time series shows that the PM2.5
concentrations during the day were higher than during the
night (Po0.05). The total number of outdoor measurements
were 218 (21%), while indoor measurements were 821
(79%).
The PM2.5 concentrations measured in different indoor
microenvironments are shown in Figure 3. The highest
120
100
80
60
40
20
Indoor
Outdoor
21:00
20:00
19:00
18:00
17:00
16:00
15:00
14:00
13:00
12:00
11:00
10:00
0
09:00
where Tij is the time variable on k measurement, b0 is the fix
effect coefficient, b1 is the random effect coefficient, b2 is the
coefficient for the amplification factor, b3 is the interaction
term between PM2.5 measurements and time variable, Gi is
the outcome variable in the i subject, Gj are the control
factors measured at the j time, n is the random coefficient
vector, and e is the random error term.
PM2.5 count was log transformed in order to approximate
it to normal distribution. Two different models were
constructed: one for indoor microenvironments and another
for outdoor microenvironments. For indoor microenvironments, the predictors included in the model were geographical location (northeast, northwest, center, southeast,
southwest) and the hours of the day (which had to be
divided into four categories according to PM2.5 average
Results
3
Yij ¼ b0 þ b1 Tij þ b2 Gi þ b3 ðGj Tij Þ þ n0i þ n1i Tij þ eij
concentrations, due to the circadian-like behavior of PM2.5
concentration). For outdoor microenvironments, only the
hour of the day was considered as the indicator variable,
because the geographical location at outdoor microenvironments was difficult to assess, and the number of samples at
outdoor microenvironments was relatively small.
Another model was constructed to evaluate interaction
terms in which microenvironments were taken as a twocategory variable, one for all indoor microenvironments and
the other for all outdoor microenvironments. Geographical
location was also a two-category variable, the north area of
the city in one category, and the center, southeast, and
southwest in the other category. Interaction terms were
constructed between geographical location and hours of the
day and considered for inclusion in the final model if the
significant coefficient was less than 0.2. The confounding bias
was not tested due to the nature of the study design.
Statistical significance was set at alpha level r0.05. Passive
smoking and cooking measurements were 33 samples that
represent 3.2% of the total amount of measurements. These
were excluded from the analysis.
[PM 2.5 ] µg/m
Statistical Analysis
Since the purpose of this pilot study was to describe and
characterize healthy volunteers PM2.5. personal exposure
without pre-established activities pattern, descriptive statistical analysis was employed. The summary and dispersion
estimates are shown as medians, 25 percentile and 75
percentile, because the distribution of data did not fit a
normal pattern (Shapiro Wilk’s test P ¼ 0.000). The
graphical observation of raw data showed a decrease in
PM2.5 concentrations around 1700 hours, therefore, the data
were classified as day measurements (from 0900 to 1700),
and night measurements (from 1700 to 2200). Comparisons
were made using a Mann–Whitney’s U-test. Indoor and
outdoor measurements were categorized according to the
microenvironment previously established in the diary of
activities. The former included the following places: home,
work, school, public places, and the subway. Outdoor
microenvironments were: bus, automobile, standing, or
walking outdoors. Since the study design did not allow
simultaneous measurements of indoor and outdoor microenvironments, it was not possible to perform paired
comparisons among indoor and outdoor microenvironments
PM2.5 concentrations. Therefore, each group of microenvironments was compared using the Kruskal–Wallis’s rank test.
Measurements were also classified into geographical location
(north, center, southwest, and southeast) and comparisons
among them were also carried out with Kruskal–Wallis’s
rank test.
Multivariate analysis was performed in order to describe
the variability of PM2.5 concentrations at indoor as well as at
outdoor microenvironments controlling for other factors
such as geographical location of the participants during the
measurement period and hour of the day. Since particulate
matter sequential measurements are known to be correlated,
a statistical linear mixed model that accounts for this
situation was used with an exchangeable correlation structure
(Diggle et al., 1995); each measurement Yij was modeled as
Vallejo et al.
Hours
Figure 2. Medians of PM2.5 concentrations by indoor and outdoor
microenvironments.
325
Vallejo et al.
Figure 3. Indoor microenvironment PM2.5 concentrations.
concentrations occurred inside the subway (106.20 mg/m3,
76.4–125.4), followed by schools, in which the concentrations were 93.3 mg/m3 (59.7–133.5). The lowest concentrations occurred at home (54.5 mg/m3, 39.7–88.9) (Po0.05),
where measurements of this pollutant were highly disperse
with a range of 15–525.8 mg/m3. On the other hand, the
subway was the indoor microenvironment in which measurements were less disperse (range 37.7–174.3 mg/m3).
Figure 4 shows the PM2.5 concentrations measured in
outdoor microenvironments. Bus transportation showed the
highest concentrations (101.7 mg/m3, 61.5–136.7) with less
dispersion of measurements (range 27.3–218 mg/m3), while
the automobile (private automobile or taxi) had the lowest
concentration in this group (64.2 mg/m3, 44.6–109.1) with the
highest dispersion with measurements ranging from 23.8 to
329.8 mg/m3. The median concentrations of PM2.5 in the bus
and standing on the street were very similar with a difference
of only 12.4 mg/m3. The medians in this group showed
statistically significant differences (Po0.05).
Figure 5 shows PM2.5 concentrations according to the
geographical areas in which MCMA is officially divided.
The center locations showed the highest concentrations
(89.4 mg/m3, 67.9–113.2) and the northeast area the lowest
(54.4 mg/m3, 42.6–89.4). In the southwest, the median
concentrations were 69 mg/m3 (48.4–100.3), but it was the
area that showed the highest dispersion of measurements
ranging from 15 to 525.8 mg/m3, while the lowest dispersion
was observed in the northeast area (Po0.05).
Multivariate Analysis
The average PM2.5 concentrations predicted by the model
among the different indoor microenvironments, when controlling for the geographical location and the hour of the
day, are shown in Table 1. The public places microenvironments showed the highest concentrations. This result
is different from the one obtained with the descriptive
analysis of raw data, in which the school microenvironment
showed the highest concentration (Figure 3). With regard to
326
PM2.5 personal exposure
Figure 4. Outdoor microenvironment PM2.5 concentrations.
Figure 5. PM2.5 concentrations by geographical area of MCMA.
geographical location, the multivariate analysis confirms
the observations from raw data descriptive analysis: the
microenvironments at the northeast and northwest had
lower concentrations than the center, southeast, and southwest (Figure 5); likewise PM2.5 concentrations were higher
during day measurements than during night measurements
(Figure 2).
As for outdoor microenvironments, the predicted particulate matter concentrations (Table 2) were similar in the three
microenvironments. During day measurements, PM2.5 concentrations were higher than during night measurements,
which indicates that any outdoor microenvironment represents the same risk of exposure to this pollutant.
Coefficients predicted by multivariable model in which an
interaction term was introduced are shown in Table 3.
According to the model estimations, PM2.5 concentrations
are mainly determined by the geographical location and the
hour of the day, and they are not influenced by the type of
microenvironment (indoor or outdoor). Moreover, the
interaction term shows that PM2.5 concentrations are
different among geographical locations only in the interval
between 1400 and 1500 hours.
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(4)
PM2.5 personal exposure
Vallejo et al.
Table 1. Average PM2.5 concentrations (mg/m3) at indoor microenvironments predicted by the linear mixed-effect model.
Table 3. Linear mixed-effect model PM2.5 regression coefficients.
P
Coefficient* 95% CI
Hour of the day
Indoor microenvironments
NE
NW
Center
SE
SW
900–1330
1400–1500
1530–1800
1830–2200
Home
86.3
79.0
55.9
49.0
75.3
67.3
48.8
42.8
102.2
91.3
66.2
58.0
100.5
89.7
65.1
53.6
900–1330
1400–1500
1530–1800
1830–2200
School
88.5
79.0
57.3
50.2
77.2
68.9
50.0
43.8
104.8
93.5
67.8
59.5
103.0
92.0
66.7
58.5
96.7
86.4
62.6
54.9
900–1330
1400–1500
1530–1800
1830–2200
Work
82.9
74.0
53.7
47.0
72.3
64.6
46.8
47.0
98.1
87.6
63.5
55.7
96.5
86.1
62.5
54.8
90.6
80.9
58.7
51.4
900–1330
1400–1500
1530–1800
1830–2200
Public places
111.3
99.4
72.1
63.2
97.2
86.7
62.9
55.2
131.8
117.7
85.4
74.8
129.6
115.7
83.9
73.6
121.7
108.7
78.8
69.1
94.4
84.3
61.1
53.6
Microenvironment
Geographical location
900–1330
1400–1500
1530–1800
Geographical location+1400–1500
Constant
0.1299
0.1595
0.5191
0.2227
0.1058
0.2464
3.8529
0.1033;
0.0467;
0.4549;
0.0609;
0.0325;
0.0617;
3.6903;
0.3633
0.2724
0.5832
0.3845
0.1792
0.4312
4.0155
NS
0.006
0.000
0.007
0.005
0.009
0.000
*Logarithmic units. NS ¼ no significant.
Table 2. Average PM2.5 concentrations (mg/m3) at outdoor microenvironments predicted by the linear mixed-effect model.
Hour of the day
900–1330
1400–1500
1530–1800
1830–2200
Outdoor microenvironments
Bus
Automobile
On-street
92.8
85.3
66.8
68.7
92.6
85.1
66.7
68.6
94.5
86.8
68.0
70.0
Discussion
This work presents novel data related to PM2.5 personal
exposure measurements in MCMA, in a group of healthy
residents during their daily activities. Exposure patterns
were identified within different microenvironments as well
as the PM2.5 temporal behavior. The microenvironments
and geographical locations that explain PM2.5 variations
along the study period were identified by multivariate
analysis.
In Mexico City, most research about air pollution has been
carried out using ambient monitors from the environment
monitor network that gives a good estimate but could be
altered by weather conditions such as wind, rain, or relative
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(4)
humidity (Romieu et al., 1992, 1996; Borja-Aburto et al.,
1997; Loomis et al., 1999). It is known that outdoor
concentrations of PM2.5 are well correlated with ambient
measurements. However, both indoor concentrations from
fix monitors and personal measurements are only weakly
correlated with outdoor concentrations (Ozkayanak et al.,
1996). Therefore, personal exposure measurements allow a
better estimation of the real exposure of an individual during
an ordinary working day.
Microenvironment and Geographical Location Exposure
Concentrations
In the present study, it was found that PM2.5. concentration
levels had a circadian-like behavior probably related to an
increase in the population daily activities during the morning
hours, which decrease in the evening, especially at indoor
microenvironments. It has been previously reported that
people tend to stay for longer periods indoors than outdoors
(GB Parliament House of Commons Environment Committee, 1991; Wigzell et al., 2000). As expected, outdoor
microenvironments had higher concentrations than indoor,
even though the former represented only the 22% of the
overall sample measurements. Persons who stay for longer
periods at outdoor places (i.e., public transport drivers,
police officers, sales men, etc) could be both overly and
chronically exposed to high concentrations of PM2.5, as it
was shown in a study with Bangkok traffic police (Jinsart
et al., 2002).
Subway exposure was the indoor microenvironment with
the highest concentrations of PM2.5, even though the
Mexican subway works with electrical energy. The system
has several long lines on the surface, and because of the city
soil characteristics, the tunnels are very superficial and have
ample and direct connection with the surface for ventilation.
It is important to note that regardless of the microenvironment, PM2.5 concentrations in this study were very high and
unsafe for the susceptible population, according to international standards (WHO-EURO Update and Revision of the
Air Quality Guidelines for Europe, 1995; US Environmental
Protection Agency, Air Quality Strategies and Standards
Division, 1998).
327
Vallejo et al.
The highest concentrations of PM2.5 were observed in the
central region (downtown) and in the southwest areas, while
the northeast and northwest regions showed the lowest, even
though most industries are located in these areas of the
MCMA. This may be explained by the fact that the
prevailing winds blow the particles to the southwest, where
they are blocked by the Ajusco mountains and cannot
disperse (Figure 1). Other authors have reported higher
concentrations of pollutants such as ozone in the southwest
area (Borja-Aburto et al., 1997), even though measurements
were made with ambient monitors. These results indicate that
particulate matter behaves in a similar fashion as to other
pollutants, such as ozone in this area.
Microenvironment Relationship to PM2.5 Concentrations
Many factors influence personal exposure to air pollution.
PM2.5 concentrations change due to emission sources,
weather conditions (temperature, relative humidity, wind
velocity, etc.), geographical location, time of the day, and
type of microenvironment among others. In this study, two
covariants were related to PM2.5 concentration in the
different microenvironments, the geographical location and
the hour of the day. The average concentrations of PM2.5
predicted by the model were similar to those observed in the
descriptive analysis with raw data. However, while the
descriptive analysis pointed out the highest concentrations at
the school indoor microenvironment, the model predicted
that the higher concentration were in public places. This
disagreement between the descriptive analysis and the model
can be explained by the effect of controlling covariants. The
same occurred at outdoor microenvironments, where exposure in the bus was the highest (in descriptive analysis), but
the multivariate model predicted similar PM2.5 concentrations for the three outdoor microenvironments. Furthermore,
PM2.5 concentrations had a different distribution pattern
between 1400 and 1500 hours within the geographical
locations, while the rest of the time PM2.5 had a similar
distribution pattern at the different geographical locations.
These findings could be related to an increase in vehicles
movement due to working and school day off, or may be due
to changes in weather conditions, but further studies are
needed in which these variables are considered.
Study Considerations and Limitations
The population evaluated in this work included subjects who
were willing to participate voluntarily, whose daily activities
were sedentary, and mainly in indoor microenvironments. In
order to generalize the present results, a study with a large
sample that includes a more heterogeneous population with
different occupational activities should be carried out,
since there is an influence of occupational activities on
exposure patterns (Brauer et al., 2000). Street workers, such
as police officers, salesmen, and bus drivers, as well as high
school students and office workers, among others, should be
328
PM2.5 personal exposure
included. Nevertheless, in this study the participants’
location, either at home or at work covered most of the
total area of MCMA, as shown in Figure 1.
In this study, the participants performed their daily
activities and wrote down the microenvironment in which
they stayed for the most part of each preset 30-min period.
This allowed us to identify the microenvironment without
recall bias, although microenvironment classification errors
could be introduced by the participants. Other studies
assigned activities to the participants into different microenvironments (Levy et al., 2000), or interviewed the study
subjects in order to identify the microenvironments in which
they stayed during the whole recording period (Brauer et al.,
2000). Preassigned activities avoid microenvironment misclassification, but do not allow the assessment of the real
exposure during activities on an ordinary day.
In order to compare the PM2.5 exposure between indoor
and outdoor microenvironments, some studies have measured this pollutant simultaneously in both microenvironments (Brauer et al., 2000; Levy et al., 2000; Rojas-Bracho
et al., 2000). Since in this work PM2.5 real-time personal
measurements were obtained, measurements at indoor and
outdoor microenvironments were not paired by time, and
indoor and outdoor measurements were grouped independently of the day time.
The identification of emission sources and particle
composition are of special concern if the purpose is to study
the health effects associated to particulate matter exposure.
This matter is beyond the scope of the present work.
A final consideration must be given to the exposure
measurement method used in this study. Until recently, most
epidemiological studies have been conducted using fixed
monitors either in an indoor and/or in outdoor microenvironment (Borja-Aburto et al., 1997; Chen et al., 1999; Gold
et al., 1999). These devices yield only an average particulate
matter concentration and do not reflect the fluctuations in
particulate levels over the sampling period, and therefore, a
less accurate measurement of PM personal exposure is
obtained in comparison to light-scattering nephelometers that
provide a more accurate and precise resolution of particulate
matter measurements (Quintana et al., 2000; Wigzell et al.,
2000). A published work that evaluated the performance of
Personal DataRAM (2000) (pDRs) and Radiance nephelometer against measurements from both Harvard impactors
(HI2.5) and Harvard personal environmental monitors
(HPEM2.5) for PM2.5 indoor, outdoor and personal settings
showed that pDRs measurement of PM2.5 could systematically overestimate the HPEM2.5 measurements by approximately 27% (Liu et al., 2002). However, since this error
occurs in every measurement, the real values of PM2.5 can be
estimated by adjusting the averaged measurements.
In conclusion, PM2.5 personal exposure in a specific
population of Mexican adult permanent residents of
MCMA, whose activities were sedentary, was high in almost
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(4)
PM2.5 personal exposure
all microenvironments, even higher than the international
standard recommendations. It was found that concentrations
of PM2.5 are influenced by the hour of the day as well as
geographical location, and at certain microenvironments is
not relevant to this matter.
Acknowledgments
We express our deepest gratitude to the individual subjects
who participated voluntarily in the study and acknowledge
Ms Betty Lou Chin R.N. for her support in the preparation
of the manuscript and extend out thanks to Dr Victor H.
Borja for his comments and pertinent suggestions. This
project was supported partially by the Instituto Nacional de
Cardiologia ‘‘Ignacio Chávez’’.
References
Boletı́n Terapéutico. El ı́ndice metropolitano de la calidad del aire. Gac
Med Mex 1992: 128: 51–54.
Borja-Aburto V.H., Loomis D.P., Bangdiwala S.I., Shy C.N., and
Rascon-Pacheco R.A. Ozone, suspended particulate, and daily
mortality in Mexico City. Am J Epidemiol 1997: 145: 258–268.
Brauer M., Hruba F., Mihalikova E., Fabianova E., Miskovic P.,
Plzikova A., Lendacka M., Vandenberg J., and Cullen A. Personal
exposure to particles in Banska Bystrica, Slovakia. J Expos Anal
Environ Epidemiol 2000: 10: 478–487.
Chen M.L., Mao I.F., and Lin I.K. The PM2.5 and PM10 particle in
urban areas of Taiwan. Sci Total Environ 1999: 226: 227–235.
Diario Oficial de la Federación. Proyecto de modificación a la NOM025-SSA1-1993, Salud ambiental. Criterio para evaluar la calidad del
aire ambiente, con respecto a las partı́culas menores de 10 micras
(PM10). Diario Oficial de la Federación, Mexico, 2002.
Diggle P.J., Liang K.Y., and Zeger S.L. Analysis of Longitudinal Data.
Clarendon Press, Oxford, UK, 1995.
Dockery D.W., Pope III C.A., Xu X., Spengler J.D., Ware J.H., Fay
M.E., Ferris B.G. Jr., and Speizer F.E. An association between
pollution and mortality in six U.S. cities. N Engl J Med 1993: 329:
1753–1759.
GB Parliament House of Commons Environment Committee. Indoor
Pollution. Vol. 1 F Report together with the Proceeding of the
Committee relating to the report, session 1990–1991. Sixth report,
HMSO, London, 1991.
Gold D.R., Damokosh A.I., Pope III C.A., Dockery D.W., McDonnell
W.F., Serrano P., Retama A., and Castillejos M. Particulate and
ozone pollutant effects on the respiratory function of children in
southwest Mexico City. Epidemiology 1999: 10: 8–16.
Jinsart W., Tamara K., Loetkamonwit S., Thepanondh S., Karita K.,
and Yano E. Roadside particulate air pollution in Bangkok. J Air
Waste Manage Assoc 2002: 52: 1102–1110.
Levy J.I., Houseman E.A., Ryan L., Richardson D., and Spengler J.D.
Particle concentrations in urban microenvironments. Environ Health
Perpect 2000: 108: 1051–1057.
Liu L.J., Slaughter J.C., and Larson T.V. Comparison of light scattering
devices and impactors for particulate measurements in indoor,
Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(4)
Vallejo et al.
outdoor and personal environments. Environ Sci Technol 2002: 36:
2977–2986.
Loomis D., Castillejos M., Gold D., McDonnell W., and Borja-Aburto
V. Air pollution and infant mortality in Mexico City. Epidemiol
1999: 10: 118–123.
Onursal B., and Gautam S.P. Contaminación atmosférica por vehı́culos
automotores. Experiencias recogidas en siete centros urbanos de
América Latina. Documento técnico del Banco Mundial No. 373S.
Banco Mundial, Washington, 1997, pp. 135–289.
Ostro B., Sánchez J.M., Aranda C., and Eskeland G.S. Air pollution
and mortality results from study of Santiago, Chile. J Expos Anal
Environ Epidemiol 1996: 6: 97–114.
Ozkayanak H., Xue J., Spengler J., Wallace J., Pellizzari E., and Jerkins
P. Personal exposure to airborne particle and metals: results from the
particle TEAM study in Riverside, California. J Expos Anal Environ
Epidemiol 1996: 60: 57–78.
Personal DataRAM. Models pDR-1000AN and pDR-1200, MIE, Inc.
Bedford, MA, 2000.
Ponciano-Rodrı́guez G., and Rivero-Serrano O. Factores ambientales y
salud en el ambiente urbano. El caso de la Zona Metropolitana de la
Ciudad de México. In: Rivero S.O., Ponciano R.G. (Eds.). Riesgos
ambientales para la salud en la ciudad de México. Programa
Universitario de Medio Ambiente, Mexico, 1996, pp 443–456.
Pope III C.A., Thun M.J., Namboodiri M.M., Dockery D.W.,
Evans J.S., Speizer F.E. and Heath C.W. Jr. Particulate air
pollution as predictor of mortality in a prospective study of U.S.
adults. Am J Respir Crit Care Med 1995: 151: 669–674.
Quintana P.J., Samimi B.S., Kleinman M.T., Liu L.J., Soto K., Warner
G.Y., Bufalino C., Valencia J., Francis D., Hovell M.H., and
Delfino R.J. Evaluation of real-time passive personal particle
monitors in fixed site residential indoor and ambient measurements.
J Expos Anal Environ Epidemiol 2000: 10: 437–445.
Rojas-Bracho L., Suh H.H., and Koutrakis P. Relationships among
personal, indoor and outdoor fine and coarse particle concentrations
for individuals with COPD. J Expos Anal Environ Epidemiol 2000:
10: 294–306.
Romieu I., Cortés-Lugo M., Ruiz-Velasco S., Sánchez S., Meneses F.,
and Hernandez M. Air pollution and school absenteeism among
children in Mexico City. Am J Epidemiol 1992: 136: 1524–1531.
Romieu I., Meneses F., Ruiz S., Sienra J., Huerta J., and White M.
Effects of air pollution on the respiratory health of asthmatic
children living in México City. Am J Respir Crit Care 1996: 154:
300–307.
Saldiva P.H., Pope III C.A., Schwartz J., Dockery D.W., Lichtenfels
A.J., Salge J.M., Barone I., and Bohm G.M. Air pollution and
mortality in elderly people: a time series study in Sao Paulo, Brazil.
Arch Environ Health 1995: 50: 159–163.
Simpson R.W., Williams J., Morgan G., and Rutherford S. Associations between particulate air pollution and daily mortality in
Brisbane, Australia. Arch Environ Health 1997: 52: 442–454.
US Environmental Protection Agency, Air Quality Strategies and
Standards Division. Revised Ozone and Particulate Matter (PM)
National Ambient Air Quality Standards. US Environmental Protection Agency, Washington, DC, 1998.
WHO-EURO Update and Revision of the Air Quality Guidelines for
Europe. EUR/ICP/EHAZ 94 05/PB01, World Health OrganizationEuropean Region, Copenhagen, Denmark, 1995.
Wigzell E., Kendall M., and Nieuwenhuijsen M.J. The spatial and
temporal variation of particulate matter within the home. J Expos
Anal Environ Epidemiol 2000: 10: 307–314.
329