Effect of poverty on the relationship between personal exposures and ambient concentrations of air pollutants in ho chi minh city

Accepted Manuscript
Effect of Poverty on the Relationship between Personal Exposures and Ambient
Concentrations of air pollutants in Ho Chi Minh City
Sumi Mehta, Hind Sbihi, Tuan Nguyen Dinh, Dan Vu Xuan, Loan Le Thi Thanh, Canh
Truong Thanh, Giang Le Truong, Aaron Cohen, Michael Brauer
PII: S1352-2310(14)00532-9
DOI: 10.1016/j.atmosenv.2014.07.011
Reference: AEA 13097
To appear in:
Atmospheric Environment
Received Date: 15 April 2014
Revised Date: 30 June 2014
Accepted Date: 3 July 2014
Please cite this article as: Mehta, S., Sbihi, H., Dinh, T.N., Xuan, D.V., Le Thi Thanh, L., Thanh, C.T., Le
Truong, G., Cohen, A., Brauer, M., Effect of Poverty on the Relationship between Personal Exposures
and Ambient Concentrations of air pollutants in Ho Chi Minh City, Atmospheric Environment (2014), doi:
10.1016/j.atmosenv.2014.07.011.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
1

Effect of Poverty on the Relationship between Personal Exposures
and Ambient Concentrations of air pollutants in Ho Chi Minh City

Sumi Mehta
1
,

Hind Sbihi
2
, Tuan Nguyen Dinh
3
, Dan Vu Xuan
4
,
Loan Le Thi Thanh
5
, Canh
Truong Thanh
6
, Giang Le Truong
7
, Aaron Cohen
1
, Michael Brauer
2

1
Health Effects Institute, Boston.MA, USA.
2
School of Population and Public Health, University of British Columbia, Vancouver.
Canada.

3

Ho Chi Minh City Environmental Protection Agency (HEPA); Institute for Environment
and Resources (IER). The National University Of Ho Chi Minh City, Vietnam.
4
Center for Occupational and Environmental Health, Vietnam.
5
Ho Chi Minh City Bureau of Statistics, Vietnam.
6
Ho Chi Minh City University of Science, Vietnam.
7
Department of Public Health, Vietnam.
Corresponding Author: Hind Sbihi. School of Population and Public Health, 2206 East
Mall. Vancouver, BC. V6T 1Z2. Canada.
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
2

Abstract: Socioeconomic factors often affect the distribution of exposure to air pollution. The 1
relationships between health, air pollution, and poverty potentially have important public health and 2
policy implications, especially in areas of Asia where air pollution levels are high and income disparity is 3
large. The objective of the study was to characterize the levels, determinants of exposure, and 4
relationships between children personal exposures and ambient concentrations of multiple air pollutants 5
amongst different socioeconomic segments of the population of Ho Chi Minh City, Vietnam. Using 6
repeated (N=9) measures personal exposure monitoring and determinants of exposure modeling, we 7
compared daily average PM
2.5
, PM
10
, PM

2.5
absorbance and NO
2
concentrations measured at ambient 8
monitoring sites to measures of personal exposures for (N=64) caregivers of young children from high 9
and low socioeconomic groups in two districts (urban and peri-urban), across two seasons. Personal 10
exposures for both PM sizes were significantly higher among the poor compared to non-poor participants 11
in each district. Absolute levels of personal exposures were under-represented by ambient monitors with 12
median individual longitudinal correlations between personal exposures and ambient concentrations of 13
0.4 for NO
2
, 0.6 for PM
2.5
and PM
10
and 0.7 for absorbance. Exposures of the non-poor were more highly 14
correlated with ambient concentrations for both PM size fractions and absorbance while those for NO
2
15
were not significantly affected by socioeconomic position. Determinants of exposure modeling indicated 16
the importance of ventilation quality, time spent in the kitchen, air conditioner use and season as 17
important determinant of exposure that are not fully captured by the differences in socioeconomic 18
position. Our results underscore the need to evaluate how socioeconomic position affects exposure to air 19
pollution. Here, differential exposure to major sources of pollution, further influenced by characteristics 20
of Ho Chi Minh City’s rapidly urbanizing landscape, resulted in systematically higher PM exposures 21
among the poor. 22
23
MANUS CRIP T

ACCEP TED

ACCEPTED MANUSCRIPT
3

1. Introduction 24
Asia is undergoing significant economic development, population growth, and urbanization with 25
subsequent industrialization and growth in vehicle fleets leading to increased emissions of air pollutants 26
and shifts in environmental risks (HEI International Scientific 2010). As a result, large populations in 27
rapidly developing economies of Asia are exposed to high concentrations of air pollution. These 28
exposures, coupled with ageing populations and increasing burden of chronic diseases, have led to 29
substantial population health impacts from air pollution. The recent Global Burden of Disease estimated 30
over 2.1 million premature deaths and 52 million years of healthy life lost in Asia from ambient fine 31
particle air pollution in 2010, 2/3 of the worldwide burden. In Southeast Asia, the region which includes 32
Vietnam, outdoor air pollution was estimated to contribute to 712,000 deaths in 2010 (Lim et al. 2012; 33
Wang et al. 2012). 34
35
The public health and social policy implications of the relationships between health, air pollution, and 36
socioeconomic position are likely to be important in Asia, where air pollution levels are high and many 37
still live in poverty. Despite what appears to be a similar magnitude of population risk for a given level of 38
exposure to air pollution (Wong et al. 2010, 2008) there is still a lack of evidence about exposure sources 39
and determinants in urban Asia compared to North American and Europe. Economic deprivation has been 40
shown to increase the rates of morbidity and mortality related to air pollution in Europe and North 41
America (Finkelstein et al. 2005; Laurent et al. 2007) , and socioeconomic status dictates the vulnerability 42
of population to environmental risks via factors such as nutritional status and access to medical services. 43
In Asia where large income disparities are more prevalent than in many high-income countries, results of 44
Western studies cannot merely be extrapolated. Variation in socioeconomic status within Asian 45
populations could impact exposures differently than in developed countries, particularly for determinants 46
related to urban planning (residential location, proximity to traffic and small-scale industries), as well as 47
lifestyle and time activity patterns. Exposure to indoor combustion sources in the Asian context (for 48
example from incense use and cooking) also differ from those in the European and North American 49
settings (HEI International Scientific 2010; Le et al. 2012; Smith et al. 2000). 50

51
Studies of personal exposure conducted in developed countries indicate that for time series studies of the 52
effect of daily change in air pollution levels, central monitoring sites are adequate surrogates for 53
longitudinal changes in exposures (Janssen et al. 1998, 2005; Sarnat et al. 2000). To date no studies of 54
this type have been conducted in many of the poorer Southeast Asian countries, such as Laos, Cambodia, 55
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
4

and Vietnam. While it is possible to apply existing studies from developed countries to help tailor air 56
quality management strategies, there is a need to assess the extent to which localized sources, time 57
activity patterns, and socioeconomic position may contribute to exposure estimation in the Asian context. 58
59
Under an initiative of the Asian Development Bank, an interdisciplinary collaboration between local and 60
international experts launched assessed the health effect of air pollution and the role of poverty in Ho Chi 61
Minh City (HCMC), Vietnam. An epidemiologic study was conducted to evaluate the impact of air 62
pollution on childhood respiratory infections (children <5 years) between 2003 and 2005 (Le et al. 2012, 63
Mehta et al. 2013). This first study of the health effects of air pollution in HCMC suggested a potential 64
role of air pollution exposure (in particular NO
2
) in the development of Acute Lower Respiratory 65
Infections (ALRI) , but was unable to identify differential effects by socioeconomic position likely due to 66
the small number of patients identified as poor. Given uncertainties about the extent to which differential 67
exposure misclassification by socioeconomic status position (SES) may exist, a companion study, 68
described here, used personal monitoring of young children via their caregivers to 1) evaluate 69
determinants of personal exposure for both poor and non-poor subjects selected from a population-based 70
sample; 2) identify evidence of differential exposure misclassification by SES, and 3) assess the validity 71
of ambient monitoring as a surrogate for personal exposures. 72

2. Methods 73
The study design was a repeated measures survey of subjects selected from a representative sample of 74
households from the extremes of the household income distribution within two of the 19 geographic 75
districts in HCMC (see Figures S1 and S2 in Supplemental Material for study design and sampling 76
scheme). For each participant, personal monitoring of PM
2.5
and PM
10
(mass and filter absorbance) and 77
NO
2
was conducted along with completion of a 24-hr time activity diaries on 9 occasions spanning both 78
the dry and rainy seasons between July 2007 and March 2008. Similar air quality measurements were 79
made at fixed location monitoring sites in each district. Household characteristics were assessed by a 80
questionnaire. 81
2.1 Selection of households and participants
82
Participating households were enrolled from a population-based sample of two districts within Ho Chi 83
Minh City, the largest city in Vietnam and home to 6.1 million inhabitants (2004 Census). In March 84
2007, The Bureau of Statistics conducted a 1,000 household survey (Figure S1 in Supplemental Material) 85
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
5

to identify eligible households in Binh Thanh (BT) and District 2 (Figure 1). These districts are the 86
closest to two key monitoring stations providing air quality data used in the HCMC hospital study (Le et 87
al. 2012, Mehta et al., 2013). To further increase the linkage with the hospital study, we selected 88
households with young children (< 5 years of age).

BT is a densely populated district located in the city 89
center, while District 2, located just across the Saigon river, is much less densely populated, and during 90
the time of the study could be considered somewhat peri-urban. 91
Five wards were selected at random from within each of the two districts (Figure S2). From each ward, 92
local officials provided a list of all households with children less than five years of age and 100 93
households per district were surveyed at random. Information from this home survey (e.g. expenditures, 94
household size, assets) was used to assign households in each district to their corresponding expenditure 95
quintiles and 16 households were selected at random from the lower (lowest 20%) and higher (60-80%) 96
expenditure households in each District (see Figure S3 and Table S1 in Supplemental Material). The 97
primary caregivers for the young child in the household were selected for personal monitoring as they 98
were likely to spend the greatest amount of time in close proximity to the young child, and thus most 99
likely to experience a similar distribution of exposures. 100
101
2.2 Analytical Methods
102
Between July 2007 and March 2008, nine repeated measurements of daily average personal exposures to 103
PM
2.5
, PM
10
, and NO
2
were made for each participant. Participants were asked to wear a small 104
(approximately 1.5 kg) backpack containing all sampling equipment while engaged in normal daily 105
activities. Participants were also trained on the proper removal and placement of backpacks during 106
periods of long inactivity, such as during the night, such that sampling inlets would remain as close as 107
possible to their breathing zones. They completed a daily time activity diary during each measurement 108
period. Detailed information on exposure to potential sources of pollution, including traffic exposure, 109
incense, cottage industries, and tobacco smoke was recorded in half-hourly intervals. Participants also 110

recorded whether or not they were actually wearing the backpack during these intervals. Since 111
participants were being monitored to represent exposures of the young children under their care, they 112
were also asked to document the times when young children were with them during the measurement 113
period. In addition, detailed information on exposure to indoor sources of pollution, including incense and 114
mosquito coil use, tobacco smoke, proximity to traffic, transportation mode and frequency, and 115
ventilation quality in the house was collected at the beginning of the sampling campaign by interview 116
with the primary caregiver. Time-activity pattern (TAP) diaries were completed at each of the home visits 117
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
6

and the initial household questionnaire was filled once with study technicians to obtain information on 118
demographics, self-reported information on indoor pollutant sources and commuting (mode and time). 119
Household locations were measured by GPS (Garmin E-Trex Legend, Garmin International Inc., Olathe, 120
KS) and the distance to the nearest monitoring site and nearest major road were calculated in ArcGIS 121
(v10, ESRI, Redlands, CA). 122
123
Personal PM
10
and PM
2.5
concentrations were measured for 24 hours during each sampling session. 124
Leland Legacy (SKC) pumps were operated with a split valve to simultaneously sample PM
10
and PM
2.5
125
on 37-mm Teflon filters (No. R2PJ037, Pall Life Sciences, Ann Arbor, MI) via PEM impactors (MSP 126

Corporation, Shoreview, MN). Pumps were calibrated and flow rates were measured (Defender Model 127
510, BIOS, Butler, NJ) at the start and end of each measurement period to ensure consistent flow rates of 128
4.0 (± 0.17) L/min. Exposure to NO
2
was also monitored for each 24 hour period using Ogawa passive 129
samplers (Ogawa & Company USA, Pompano Beach FL). 130
131
Using the same methods as personal samples, daily fixed location measurements were collected for the 132
duration of the eight month study period at the government fixed ambient monitoring stations closest to 133
the districts, i.e. Zoo (closest to BT district) and District 2, to enable a comparison of personal exposures 134
and ambient concentrations (Figure 1). 135
136
All samples were analyzed at the HCMC Environmental Protection Agency (HEPA) exposure assessment 137
laboratory. The laboratory included a temperature and humidity controlled glovebox (Allen et al. 2001) 138
and a microbalance for gravimetric analysis, a reflectometer to measure particle absorbance, and an ion 139
chromatograph to analyze the Ogawa samples. Staff were fully trained to carry out standardized analytical 140
procedures. 141
142
All filters were equilibrated in a glovebox in the exposure assessment laboratory with controlled 143
temperature (22.5 ± 2.5° C) and relative humidity (40 ±5% RH) for 24 hours prior to weighing with a 144
microbalance (Model SE2, Sartorius) on an anti-static weigh boat. Reflectance was measured, using a 145
smoke-stain reflectometer (UK Diffusion Systems Ltd., London, UK), and absorption coefficients (ABS) 146
were calculated according to ISO 9835 standard (1993). All ABS are reported in m
-1
x 10
-5
. Ogawa 147
passive samplers were assembled in HEPA’s personal exposure assessment laboratory, and kept 148
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
7

refrigerated except during transport to and from the field. Aqueous extracts of filters were analyzed by ion 149
chromatography. The average analyzed nitrite value from the extracts of field blanks was subtracted from 150
each sample extract’s analyzed nitrite value. Subsequently, these blank corrected values were used to 151
calculate the concentrations in ppb. 152
153
HEPA field and lab staff carried out routine quality assurance checks, including the collection of blank 154
and duplicate samples, balance stability testing in the laboratory, and use of blanks and reference samples 155
during laboratory analysis. Specifically, technicians used 1 duplicate for every 15 filters, 1 laboratory 156
blank for every 10 filters, 1 field blank per sampler, per week and one co-located filter blank. In total, 74 157
field blanks and 5 laboratory blanks were used for the household measurements while 49 field and 5 158
laboratory blanks were deployed for the fixed site monitors. Duplicate NO
2
passive samplers were 159
collected with the personal (n= 120 based on 65 pairs of samples) and fixed site measurements (n=161 160
with 85 duplicates and 76 passive samplers) to assess precision. In each case, 4 laboratory blank samples 161
were collected. 162
163
PM concentration and absorbance values were excluded from further analysis when mean flow was 164
beyond +/- 5% of 4 L/min.; All PM concentration and absorbance values were excluded when the filter 165
pre-weight was greater than the post-weight, or when the sample duration was less than 20 hours. 166
167
All study participants signed informed consent forms prior to their participation in the study. In addition, 168
at the end of each monitoring period an honorarium of 100,000 VND (approximately $7.00) was offered 169
to each study participant. This amount, determined by the local members of the collaboration, was 170
intended to compensate participants for their time and efforts without acting as an undue financial 171
incentive that could influence the poorer participants’ participation in the study. The study proposal and 172

protocols were reviewed and approved by the institutional review board of the Biological and Medical 173
Ethical Committee of HCMC Department of Health (Decision no: 2751/SYT-NVY). 174
175
2.3 Statistical analysis
176
Correlations between monitors as well as between personal measurements of NO
2
, PM
10
, PM
2.5
and PM
2.5
177
absorbance were examined in both pooled analysis and after aggregating over the repeated measurements. 178
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
8

All pollutant levels were examined in univariate analysis to determine if any transformation of the data 179
was required. Subsequently, associations were evaluated between all air pollutant measurements and 180
distances to the closest monitor and road, and with each variable from the time activity pattern (TAP) 181
diary and the initial household questionnaire. For TAP variables, associations were analyzed using mixed 182
effects models with the participant as random intercept and with unstructured covariance. For the 183
remaining variables, categorical variables were examined using either a t-test for binary predictors or 184
ANOVA for categorical variables with more than 2 categories. 185
186
Generalized estimating equations were used to account for the correlated responses within each 187

participant and we examined the relationship between personal and ambient concentrations in a sequential 188
process by including as a fixed effect: (Step 1) SES, (Step 2) District, (Step 3) both SES and District, 189
(Step 4) Time Activity Pattern (TAP) initial household visit variables that were significant in the bivariate 190
analyses, (Step 5) SES, district, and all questionnaire variables. In steps 1, and 3 to 5, the analysis was 191
performed for each district where the participants reside, and also regardless of the location, using 192
backward stepwise regression with a cut-off p<0.05.All analyses were conducted using Stata Version 10 193
(StataCorp. 2007. Stata Statistical Software: Release 10. College Station, TX: StataCorp LP). 194
195
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
9

3. Results 196
3.1 Descriptive results
197
3.1.1 Household questionnaire and Time Activity Patterns (TAP)
198
Participants wore the personal air sampler backpack for 16.5 hours on average (max = 29.0; min = 0) and 199
had a child present with them for an average of 17.4 hours. The average time a child was present with the 200
woman who was surveyed was significantly higher (17.8 hours) in the poor households than in the non-201
poor homes (16.7 hours). Overall, participants spent 93% of their time indoors. The majority of time 202
spent in a household microenvironment was spent in the bedroom, followed by the living room. 203
All households reported extensive use of fans, for an average of 15 hours per day. Incense use was also 204
widespread, with 84% of the households reporting burning incense for an average burn time of 40 205
minutes per day. 62/64 participants reported spending time in transit, of which 42% spent one hour or 206
more in traffic each day. While only 7 participants reported current smoking, 60 reported spending time 207
in the presence of smokers and among those the average time of secondhand smoke exposure was 48 208
minutes per day. There were few significant differences by district of residence and/or SES in the time-209

activity patterns (Table 1). Subjects from non-poor households spent more time relaxing and used an air 210
conditioner more frequently, while subjects from poor households spent more time smoking, in a room 211
besides the kitchen, bedroom or living room, and used a fan more frequently. Residents of District 2 spent 212
more time in traffic, sleeping and in the bedroom, while residents of BT spent more time engaged in other 213
activities, in the living room and using a fan. 214
From the initial household questionnaire, only the use of mosquito coils and the ventilation quality 215
differed by district and by SES, while time spent in proximity to traffic when not commuting (e.g. sitting 216
or standing next to majors roads such as road-side stalls or cafés) was significantly different by district 217
and use of Kerosene as cooking fuel differed by SES (Table 1). 218
3.1.2 Quality assurance results
219
For NO
2
, 10% field blanks (n=74 for personal samplers and n=49 for fixed sites monitors) were deployed 220
during the sampling campaign with mean and standard deviation (sd) of 1 ppb (sd= 1.4 ppb) and 0.74 ppb 221
(sd =0.7 ppb) leading to a limit of detection (LOD) of 5.2 and 2.8 ppb for personal and fixed sites, 222
respectively. 88% of the personal monitoring samples were above the LOD while only 75% of the fixed 223
sites samples were above their corresponding LOD. Duplicate samples were 13% and 21% of the total 224
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
10

sample size for personal and fixed site samples, respectively. There were no significant differences 225
between paired samples and a high Pearson correlation between paired samples was found for the field 226
study pairs (r= 0.8 for personal and r= 0.6 for fixed sites). 227
228
Of the 566 expected personal PM samples, 32 PM
2.5

and 36 PM
10
values from the household 229
measurement campaign were excluded from the database prior to analysis due to either a large drift (i.e. 230
± 5% of 4L/min) in the pump flow (n=17 for PM
2.5
and n=20 for PM
10
), missing pump flow rates (n=11 231
for PM
2.5
and n=11 for PM
10
), invalid filter weights including missing values and pre/post-weight blank 232
filters that were too high to enable the calculation of PM concentration (n=11 for PM
2.5
and n=11 for 233
PM
10
). Similarly, of the 86 and 103 fixed-site samples collected at each of the District 2 and BT 234
monitoring sites for each PM fraction size, there were 9 and 14 that were discarded in District 2 and BT, 235
respectively for PM
10
, and 10 and 13 for PM
2.5
in District 2 and BT, respectively, for similar reasons 236
(drift in pump flow, missing flow rates, unusable pre-post weights) as well as non-plausible PM 237
concentration values where PM
2.5
/PM

10
>1 (n=2 for each of PM size fraction). For absorbance, 4 samples 238
for each size fraction in the fixed site measurements were excluded due to negative values while 1 239
absorbance measurement for each of PM
2.5
and PM
10
was eliminated in the personal samples. 240
3.2 Pollutant levels
241
After blank correction was applied, the mean ambient concentration of NO
2
in the BT District was 242
statistically higher than that measured in District 2 (Table 2). The two monitors’ measurements were 243
moderately correlated (r=0.48 p<0.001). The mean personal concentrations of NO
2
were higher among 244
the participants classified as non-poor (21.5 µg/m
3
, sd = 9.9) compared with those classified as poor (18.9 245
µg/m
3
, sd = 10.6) ( p=0.06). However, when examining these differences in personal concentrations by 246
district, we found that in the BT district this difference was in the expected (poor > non-poor) direction 247
unlike in District 2. 248
249


250
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
11

Following a stratified (by district) analysis, there were no statistically significant differences in personal 251
PM concentrations by SES in BT district, but the difference was still significant in District 2 for both 252
PM
10
and PM
2.5
(Table 3). As hypothesized, personal concentrations across districts for both PM sizes 253
were significantly higher among those classified as poor compared to non-poor. 254
Ambient concentrations of both PM
2.5
and PM
10
were significantly higher in the BT district compared 255
with district 2. However, since the correlation between the 2 monitors was high and statistically 256
significant PM
10
: mean r= 0.8 p<0.001 and PM
2.5
: mean r=0.9 p<0.001, ambient PM levels were 257
averaged across monitors for the longitudinal comparisons with personal PM concentrations. Table 4 258
below displays the levels of PM concentration and absorbance after averaging across the two monitoring 259
sites. 260
When examining the difference in personal and ambient concentrations by season (Figure 2), we found 261
statistically significant differences in PM
2.5

, PM
10
and absorbance for personal and fixed site levels with 262
higher levels in the dry compared with the rainy season. For NO
2
, personal levels were slightly (p=0.1) 263
higher in the rainy season (18.8 ppb) compared with the dry season (17.2 ppb). Only the BT district fixed 264
site measurements were significantly different by season, with higher levels in the rainy (23.1 ppb) vs. dry 265
(17.9 ppb) dry season. 266
267
3.3 Correlations between outdoor and personal pollutants
268
Overall, personal exposures were more highly correlated with concentrations measured at the fixed sites 269
for particulate matter (median Spearman’s r=0.7 for both size fractions with BT monitor) compared with 270
NO
2
(r= 0.42 for BT). 271
Regardless of the residential location and for all pollutants examined, the correlations between personal 272
measurements and measurements at the D2 fixed site were consistently lower than those at the BT fixed 273
site. The difference was more pronounced for NO
2
(D2: r=0.38) compared with particulate matter (D2: r 274
=0.5 and 0.65 respectively for PM
10
and PM
2.5
). 275
Since the concentrations from the fixed sites were highly correlated we examined the correlation between 276
the average of the two monitoring stations with the repeated measurements of study participants. Figure 3 277
shows the results of the analysis of the correlation by SES for absorbance, PM, and NO

2
with the latter 278
being the correlation with the closest monitor to which a participant’s home was located. 279
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
12

Summary estimates (mean and median) of correlations showed much stronger differences by SES, with 280
the correlations among the non-poor much better than those among the poor for all pollutants except NO
2
281
(Figure 3). Results for PM
2.5
were amplified (PM
10
: r=0.57 for non-poor vs. r=0.43 for poor participants; 282
PM
2.5
: r=0.62 for non-poor vs. r=0.37 for poor participants). Differences by season were also found only 283
for PM
2.5
with higher correlation in the rainy season compared with the dry season (Table 5). 284
Since we collected repeated measurements for each participant, we also fit a mixed effects model with 285
subject as random intercept to account for the correlation between visits (Table 6). The modeling results 286
showed the effect of place in the association between personal, ambient and SES. 287
SES was an important modifier of the association between ambient and personal PM
2.5
and PM

10
288
concentrations regardless of the ambient monitor used for comparison. When examining our models 289
separately in each district, we found that in District 2, being classified as poor (vs. non-poor) explains 290
significant additional variability in personal concentrations above what was explained by the ambient PM 291
concentrations alone. In BT however, SES did not influence the association between ambient and 292
personal PM
2.5
or PM
10
concentration. 293
Unlike PM, the associations between personal and ambient NO
2
concentrations were not affected by the 294
participant’s SES, confirming the results shown with the summary estimates of correlations. 295
Finally, being exposed to air pollution in the rainy season or the dry season did not affect the association 296
between ambient and personal concentrations in both districts. 297
298
3.4 Exposure factors: determinants of personal concentrations
299
Given the high correlation between the two fixed-site PM measurements, ambient PM was averaged 300
between the 2 sites for all subsequent modeling. In examining whether the association between ambient 301
and personal concentrations was modified by SES or other activities and/or time spent in different micro-302
environments and/or activities, we built determinants of personal PM concentrations models (Table 7). 303
The determinants of exposure modeling indicated that SES and the time in which air conditioning (AC) 304
was used both predicted the personal exposure for PM
2.5
and PM
10
in the expected direction (i.e. stronger 305

association for non-poor compared with poor participants and lower personal concentration with 306
increased time of AC usage). For a standard deviation increase in ambient concentration of PM
2.5
(21 307
µg/m
3
) and PM
10
(38 µg/m
3
), the personal concentration increased by 18.5 and 57 µg/m
3
respectively. For 308
a 120 minute (1 standard deviation) increase in AC use, the personal PM
2.5
and PM
10
concentration 309
decreased by 1.4 and 5.5 µg/m
3
respectively. In addition, smoking was a significant predictor of PM
2.5
310
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
13

exposures, while distance to the nearest road (as provided by the initial household questionnaire) was 311

positively associated with the personal concentration of PM
10
, but not PM
2.5
nor absorbance. Season, a 312
categorical variable relatively balanced among the two strata of SES (31 poor subjects provided samples 313
in each of the rainy and dry season vs. 35 and 33 for the non-poor study participants), had a different 314
effect on the personal level of PM
2.5
absorbance compared with the personal levels of NO
2
. For the latter, 315
being in the rainy season increased the personal concentration of NO
2
by 1.8 ppb, implying a stronger 316
association between ambient and personal concentration. In contrast, the personal PM
2.5
absorbance 317
decreased by 0.62 m
-1
x 10
-5
, for the rainy vs. dry season leading to a weaker outdoor to personal 318
association in the rainy season compared with the dry season. 319
For NO
2
, both in District 2 and BT, questionnaire variables explained more variability in personal 320
concentration than the socioeconomic position of the study participants. The quality of the ventilation in 321
the kitchen was an important factor in the personal concentration as every unit drop in ventilation quality 322
(e.g. from moderate to bad) was associated with 2.5 and 2.3 ppb decrease in the personal concentration in 323

D2 and BT respectively which corresponds approximately to a five percentile downshift. 324
Regarding model fit, the determinants of personal PM concentration for both PM size fractions explained 325
less between-subject variability compared with absorbance and NO
2
. It is important to note however, that 326
direct comparison of goodness of fit for these models is not feasible since the main predictors differed as 327
a function of the pollutant that was considered. 328
329
4. Discussion 330
Using monitoring and modeling based approaches, we evaluated whether poorer children in Ho Chi Minh 331
City systematically experienced higher exposures to air pollution per level of ambient air pollution on any 332
given day compared to non-poor children, regardless of district of residence. By comparing more precise 333
estimates of individual personal exposure with estimates based on the ambient monitoring stations, we 334
were able to explore systematic daily differences in exposure – major sources and levels - across 335
socioeconomic position. 336
We found that measured personal exposure was not well represented by ambient concentration 337
measurements in most circumstances. This is because exposure while partly reflected by ambient 338
concentration measurements is also influenced by neighborhood “hot spots” as well as micro-339
environmental levels experienced by individuals according to their personal behaviors. We compared 340
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
14

measurements of individual personal exposure with estimates based on concentrations measured at 341
ambient monitoring stations and found that there were systematic differences in these relationships across 342
socioeconomic position and seasons for both PM
2.5
and PM

10
. Measured personal exposures of the poor 343
were less correlated to those estimated from ambient monitors. 344
In addition, ambient monitoring substantially underestimated personal exposures for all measured 345
pollutants in Ho Chi Minh City, with a significantly higher underestimation among the poor for fine PM. 346
Daily mean concentrations for PM measured at the fixed sites during the same time period were lower 347
than the personal measurements, with BT district showing higher levels compared to those measured in 348
District 2 (95.2 vs. 77.8 µg/m
3
for PM
10
and 50.1 vs.39.2 µg/m
3
for PM
2.5
). Similar results were apparent 349
for NO
2
with higher personal measurements compared with those from fixed sites, with significantly 350
higher concentrations in BT district compared with District 2, and significant differences between poor 351
and non-poor participants only in District 2. 352
Thus, localized sources appeared to contribute to exposure error arising from the use of ambient 353
monitoring site data for health effects assessments, Further, the relative contribution of different sources 354
of exposure differed by socioeconomic position. 355
A wide distribution of daily personal exposures to PM
10
and PM
2.5
were measured, with average 356
exposures of 103.4 and 64.6 µg/m

3
respectively, along with mean NO
2
personal exposure of 16.2 ppb. 357
This is consistent with the distribution of ambient air pollution levels in HCMC, which are generally 358
higher than those reported in developed countries, but lower than levels observed in other Asian mega-359
cities. Personal concentrations for both PM sizes were significantly higher among those classified as poor 360
compared to participants who were classified as non-poor. Zhou and colleagues also demonstrated an 361
SES gradient in PM levels in Accra, Ghana (lowest PM in the high-SES neighborhood, and highest in two 362
of the low SES slums with geometric means reaching 71 and 131 µg/m3 for fine and coarse PM) (Zhou 363
et al. 2011). 364
365
Median longitudinal correlations between personal and ambient monitors were 0.4 for NO
2
, 0.6 for PM
2.5
366
and PM
10
and 0.7 for absorbance. These correlations were somewhat lower than those observed in similar 367
studies (Brunekreef et al. 2005; Janssen et al. 1998, 2005; Noullett et al. 2006; Wallace 2000) conducted 368
in developed countries (median longitudinal correlation (# days) = 0.74 (4-8), 0.73 (10), 0.49 (2days for 369
23 weeks), for PM
2.5
, PM
10
, Absorbance, and NO
2
, respectively). 370
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
15

Along with the socioeconomic gradient found in exposure to PM in HCMC, the exposures of the non-371
poor were more highly correlated with ambient measurements for both PM size fractions while those 372
found for NO
2
were not significantly affected by SES. This suggests that different PM sources may be 373
influencing the exposures of the poor and non-poor. Our analysis of the household characteristics and 374
time activity patterns collected along with the personal sampling campaign shed some light on these 375
sources as well as factors that would alter the relationship between fixed site and personal measurements. 376
For instance, the quality of the ventilation in the kitchen was significantly different between the two SES 377
strata, with the poor having worse ventilation quality than non-poor study participants. This modifier was 378
among the main predictors of the model for personal exposures. From the TAP diaries, differences in 379
personal factors between the poor and non-poor were more predominant than time spent in different 380
micro-environments as we observed statistically significant differences between poor and non-poor 381
HCMC residents participating in the study: the poor smoked and used fans more, while the non-poor 382
were more frequent users of AC. In order, to disentangle the roles of all the factors captured in the 383
questionnaires and the daily diaries from the role played by SES, we examined the association between 384
personal and ambient in two steps: first without including SES and offering all significant predictors in 385
the bivariate analysis; second forcing SES in the same models. Should the TAP and questionnaire 386
variables be explained by the socioeconomic position, the multicollinearity would lead to only the 387
stronger predictors remaining in the final model of the determinants of personal exposures. 388
Overall, the models for the determinants of personal exposure to NO
2
, PM
10
, PM

2.5
and absorbance 389
indicated ventilation quality and time spent in the kitchen, AC use and season as important factors that 390
were not fully captured by SES differences. These results indicate that epidemiologic analysis examining 391
the effects of air pollution on health may be biased if surrogates of SES are not included. Furthermore, 392
more detailed information capturing the specificities of developing countries (e.g. ventilation quality and 393
AC use) would reduce the potential for different degrees of exposure misclassification that may be related 394
to SES. Other influential indoor air quality determinants, such as type of cooking devices used may have 395
provided further insight in the SES gradient found in the examined pollutants;

for although nearly all 396
households (92%) used LPG as their cooking fuel, kerosene use was elevated in the poor (12.5%) 397
compared to the non-poor (3%) households. 398
Results of this study also aid in the interpretation of the companion hospital study, where analyses were 399
not able to identify differential effects by socioeconomic position (Mehta et al.2013). In the hospital 400
study, a single daily measurement of pollution was assigned to all children for a particular day. As such, 401
daily differences in individual exposures across districts or socioeconomic groups could not be adequately 402
assessed. This study lends further support to the hypothesis that poorer children in Ho Chi Minh City 403
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
16

systematically experience higher exposures to air pollution per unit of reported ambient air quality on any 404
given day compared to non-poor children, regardless of district of residence. If the exposures of the poor 405
are less well correlated with measurements made at the fixed sites used in epidemiologic analyses, there 406
will be more exposure misclassification among the poor. This would be expected to result in a decreased 407
ability to assess the true association between short-term air pollution exposure and adverse health 408
outcomes among the poor, and will limit the ability to assess differences in risk by socioeconomic 409

position. Our investigation is based on the premise that the siting of the two ambient monitors is 410
representative of average ambient concentrations within the surrounding area where participants resided. 411
We examined and confirmed that (1) residents were living at similar distances to the nearest major road 412
(245m in BT vs. 267m in District 2 based on study technicians report), and (2) that road density was not 413
significantly different around households and the corresponding monitor in each district. However, we 414
have no data to examine the distribution of industries across the two districts, although most industries are 415
small-scales operations and located mainly within residential areas. 416
Differential exposure to major sources of pollution, further influenced by characteristics of Ho Chi 417
Minh City’s rapidly urbanizing landscape, resulted in systematically higher exposures among the poor. 418
Our experience documents potential for differential misclassification of air pollution exposure by SES 419
when using ambient pollution monitors located in areas that differ in the relative contribution of different 420
sources of pollution and other aspects of the urban environment correlated with SES. These results 421
underscore the need to carefully evaluate how socioeconomic position may influence exposure to air 422
pollution. 423
Acknowledgments 424
The authors would like to acknowledge the contributions of HEPA field and lab staff, the 425
International Scientific Oversight Committee, and Timothy McAuley as well as the Bureau of Statistics 426
field staff 427
This project is supported with funds from the Health Effects Institute and the Poverty Reduction 428
Cooperation Fund of the Asian Development Bank, as well as in-kind support from the Government of 429
Vietnam. 430
431
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
17

References 432
Allen R, Box M, Liu L-JS, Larson TV. 2001. A Cost-Effective Weighing Chamber for

433
Particulate Matter Filters. Journal of the Air & Waste Management Association 51:1650–
434
1653; doi:10.1080/10473289.2001.10464392.
435
Brunekreef B, Janssen NAH, de Hartog JJ, Oldenwening M, Meliefste K, Hoek G, et al. 2005.
436
Personal, indoor, and outdoor exposures to PM2.5 and its components for groups of
437
cardiovascular patients in Amsterdam and Helsinki. Res Rep Health Eff Inst 1–70;
438
discussion 71–79.
439
Finkelstein MM, Jerrett M, Sears MR. 2005. Environmental inequality and circulatory disease
440
mortality gradients. J Epidemiol Community Health 59:481–487;
441
doi:10.1136/jech.2004.026203.
442
HEI International Scientific. 2010. Outdoor air pollution and health in the developing countries
443
of Asia: a comprehensive review. Special report 18.
444
Janssen NA, Hoek G, Brunekreef B, Harssema H, Mensink I, Zuidhof A. 1998. Personal
445
sampling of particles in adults: relation among personal, indoor, and outdoor air
446
concentrations. Am. J. Epidemiol. 147: 537–547.
447
Janssen NAH, Lanki T, Hoek G, Vallius M, de Hartog JJ, Van Grieken R, et al. 2005.

448
Associations between ambient, personal, and indoor exposure to fine particulate matter
449
constituents in Dutch and Finnish panels of cardiovascular patients. Occup Environ Med
450
62:868–877; doi:10.1136/oem.2004.016618.
451
Laurent O, Bard D, Filleul L, Segala C. 2007. Effect of socioeconomic status on the relationship
452
between atmospheric pollution and mortality. J Epidemiol Community Health 61:665–
453
675; doi:10.1136/jech.2006.053611.
454
Le TG, Ngo L, Mehta S, Do VD, Thach TQ, Vu XD, et al. 2012. Effects of short-term exposure
455
to air pollution on hospital admissions of young children for acute lower respiratory
456
infections in Ho Chi Minh City, Vietnam. Res Rep Health Eff Inst 5–72; discussion 73–
457
83.
458
Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, et al. 2012. A comparative
459
risk assessment of burden of disease and injury attributable to 67 risk factors and risk
460
factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of
461
Disease Study 2010. The Lancet 380:2224–2260; doi:10.1016/S0140-6736(12)61766-8.
462
Mehta S, Ngo LH, Dzung DV, Cohen A, Thach TQ, Dan VX, et al. 2013. Air pollution and

463
admissions for acute lower respiratory infections in young children of Ho Chi Minh City.
464
Air Qual Atmos Health 2013;6(1):167–79; doi:10.1007/s11869-011-0158-z.
465
Noullett M, Jackson PL, Brauer M. 2006. Winter measurements of children’s personal exposure
466
and ambient fine particle mass, sulphate and light absorbing components in a northern
467
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
18

community. Atmospheric Environment 40:1971–1990;
468
doi:10.1016/j.atmosenv.2005.11.038.
469
Sarnat JA, Koutrakis P, Suh HH. 2000. Assessing the relationship between personal particulate
470
and gaseous exposures of senior citizens living in Baltimore, MD. J Air Waste Manag
471
Assoc 50: 1184–1198.
472
Smith KR, Samet JM, Romieu I, Bruce N. 2000. Indoor air pollution in developing countries and
473
acute lower respiratory infections in children. Thorax 55: 518–532.
474
Wallace L. 2000. Correlations of Personal Exposure to Particles with Outdoor Air

475
Measurements: A Review of Recent Studies. Aerosol Science and Technology 32:15–25;
476
doi:10.1080/027868200303894.
477
Wang H, Dwyer-Lindgren L, Lofgren KT, Rajaratnam JK, Marcus JR, Levin-Rector A, et al.
478
2012. Age-specific and sex-specific mortality in 187 countries, 1970–2010: a systematic
479
analysis for the Global Burden of Disease Study 2010. The Lancet 380:2071–2094;
480
doi:10.1016/S0140-6736(12)61719-X.
481
Wong C-M, Vichit-Vadakan N, Kan H, Qian Z. 2008. Public Health and Air Pollution in Asia
482
(PAPA): A Multicity Study of Short-Term Effects of Air Pollution on Mortality. Environ
483
Health Perspect 116:1195–1202; doi:10.1289/ehp.11257.
484
Wong CM, Vichit-Vadakan N, Vajanapoom N, Ostro B, Thach TQ, Chau PYK, et al. 2010. Part
485
5. Public health and air pollution in Asia (PAPA): a combined analysis of four studies of
486
air pollution and mortality. Res Rep Health Eff Inst 377–418.
487
Zhou Z, Dionisio KL, Arku RE, Quaye A, Hughes AF, Vallarino J, et al. 2011. Household and
488
community poverty, biomass use, and air pollution in Accra, Ghana. PNAS 108:11028–
489
11033; doi:10.1073/pnas.1019183108.

490
ISO 9835:1993 Ambient Air - determination of a black smoke index.
491
492
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
Tables:
Table 1- Average Time (Standard Deviation in hour/day) spent in Microenvironments and on
Activities (TAP diaries), and Initial household questionnaire descriptive statistics by District and SE
Table 2- NO
2
summary statistics by District and by SES
Table 3 - Personal and Ambient PM concentration and absorbance levels by District and by SES.
Table 4 – Mean Fixed Sites Ambient PM Concentration and Absorbance.
Table 5- Summary estimates of individual longitudinal correlation between Ambient (mean levels
between two fixed site monitors for PM, and nearest monitor for NO2) and Personal Pollutants levels by
SES and by Season.
Table 6- Effect of District and SES in personal/ambient concentrations repeated measures models
Table 7 - Final explanatory models showing significant variables affecting association between
personal and ambient NO
2
, PM
2.5
and PM
10
concentrations and absorbance.

MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
Table 1- Average Time (Standard Deviation in hours per day) spent in Microenvironments and on Activities (TAP
diaries), and Initial household questionnaire descriptive statistics by District and SES
SES District
Overall Non-poor Poor D2 BT
Wearing backpack
16.5 (6.2) 16.7 (6.3) 16.3 (6.1) 16.4 (5.9) 16.7 (6.5)
Child present

§

17.2 (5.6) 16.7 (5.8) 17.8 (5.4) 17.5 (5.8) 16.9 (5.4)
Household activities
Sleeping
*
8.6 (2.9) 8.4 (2.8) 8.8 (3.0) 9.2 (2.6) 8 (3.1)
Cooking 1.5(1.6) 1.4 (1.4) 1.5 (1.8) 1.5 (1.8) 1.5 (1.5)
Housework 4.1 (4.2) 4 (3.9) 4.2 (4.4) 4.3 (4.3) 3.9 (4.0)
Working outside near home 1.2 (2.7) 1.2 (2.7) 1.1 (2.7) 1.1 (2.6) 1.2(2.8)
Relaxing
§
4.8 (3.8) 5.2 (3.6) 4.5 (3.8) 4.8 (3.7) 4.8 (3.8)
Other activity
*

2.4 (2.7) 2.2 (2.7) 2.6 (3.5) 1.9* (2.8) 2.8* (3.4)
Microenvironment (home)
Kitchen 2.5 (3.3) 2.3 (2.3) 2.7(4.1) 2.7 (3.9) 2.3 (2.6)
Living room
*
§
7.5 (5.0) 7.9 (4.5) 7.1(5.5) 6.8* (4.5) 8.2* (5.4)
Bedroom 8.2 (4.1) 8.5 (3.7) 7.9(4.5) 8.6* (4.3) 7.8* (3.9)
Other room
*
§
1.1 (2.7) 0.7(1.3) 1.5 (3.5) 1.4(3.1) 0.8(2.1)
Microenvironment (outside)
Working away from home 0.7 (2.3) 0.8 (2.3) 0.6 (2.2) 0.6 (2.1) 0.9 (2.5)
Transit (foot, car, bike, etc.) 0.6 (1.4) 0.6(1.1) 0.5(1.7) 0.5 (1.5) 0.6 (1.4)
Other 0.2 (0.9) 0.2 (0.9) 0.2 (0.9) 0.2 (1.0) 0.2(0.9)
Other personal factors

Smoking (self)

§

0.2 (0.8) 0.1 (0.4) 0.3 (1.1) 0.2 (1.0) 0.1 (0.6)
Smoking (other)
0.8 (1.7) 0.8(1.9) 0.8(1.5) 0.7 (1.4) 0.9 (2.0)
Burning incense
0.6 (1.6) 0.6 (2.0) 0.5 (1.1) 0.4(1.0) 0.7 (2.0)
Traffic
*

0.6 (1.6) 0.5 (1.5) 0.7 (1.6) 0.8 (1.8) 0.4 (1.3)
Air conditioner
§

0.4 (2.0) 0.7 (2.7) 0.04(0.4) 0.4 (1.9) 0.3 (2.1)
Fan
*
§

12.8 (8.2) 11.4 (8.1) 14.3 (8.1) 11.9 (8.5) 13.7 (7.9)
Other
§

0.08 (0.7) 0 (0) 0.2 (1.0) 0.1 (0.6) 0.10 (0.8)
Household Questionnaire


Do you burn incense?


Yes

30 26 27 29
No

5 6 6 5
Do you use mosquito coils?

*

§



Yes


6 11 10 7
No


30 21 23 28
Do you cook for sale outside?


Yes

2 2 2 2
No

32 30 30 32
MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
Do you spend time close to
traffic when not commuting?
*

Yes

8 8 11 5
No

27 24 22 29
Ventilation Quality in Kitchen
*
§




Poor


2 9 4 7
Moderate
23 16 19 20
Good
10 7 10 7
Cooking fuel
§


LPG
34 28 31 31
Kerosene
1 4 2 3

§
Significant difference by SES (p<0.05).
* Significant difference by District (p<0.05)


MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
Table 2- NO
2
summary statistics by District and by SES


District 2

BT

Personal
concentrations

Overall
Poor
(n=104)

Non
-
Poor
(n=112)
Overall

Poor
(n=116)

Non-Poor
(n=120)
Mean
(µg/m
3
)
17.95 16.1 19.6 17.86 18.7 17
SD
10.1

9.6

11.5

10.7

10.4

9.8

Max
56.8

49.4

56.8

54

54

40.3

Min
.15

.3

.15

.4

2

.4

Ambient
concentrations
N
376

293

Mean
16.1

19.5

SD
8.6

10.2

Median
15.9

19.3

Max
46.9

46.9

Min
0.6

0.2



MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT

Table 3 – Personal and Ambient PM concentration and absorbance levels by District and by SES.

District 2 BT

Personal

PM
10
PM
2.5

ABS
PM
2.5

PM
10
PM
2.5

ABS
PM
2.5


Poor

Non-
Poor
Poor
Non-
Poor

Poor

Non-
Poor
Poor

Non-
Poor
Poor

Non-
Poor
Poor

Non-
Poor

N
113 125 108 127 108 125 114 114 117 114 117

117
Mean
113 92 73.6 54.7 5.7 5.2 113.7 96.9 67.7

64.3 5.5 5.6
Standard
deviation
55.5 44.8 42.1 24.2 2.1 2.1 151.8 37 26.3

32 1.9 2.3

Max
424.7 230.8 375.9 128 11.2

12.7

1675.5 204.5

147.5

175.7

10.5

18.1

Min
9.8 9.5 4.6 5 0.16 0.3

22.5 19 61.3 57.7 0.5 0.1

Ambient


District 2

BT






Ratio
PM
2.5/
PM
10

PM
10
PM
2.5
ABS PM
2.5

Ratio
PM
2.5:
PM
10

PM
10
PM
2.5

ABS PM
2.5

N
313 256 262 262 256 340 330 317
Mean
0.54 77.8 39.2 4.9 0.54 95.2 50.1 5.4
Standard
deviation
0.09 41 18.5 1.9 0.16 34.8 21.2 1.7
Median
0.52 66 35.4 4.9 0.54 89.9 44.3 5.19
Max
0.77 167 99.1 8.7 0.97 180.1 103.4 9.3
Min
0.36 27 10.1 0.9 0.1 34.2 14.6 0.01


MANUS CRIP T

ACCEP TED
ACCEPTED MANUSCRIPT
Table 4 – Mean Fixed Sites Ambient PM Concentration and Absorbance.


PM concentrations
(ug/m3)
PM ABS
(m
-1

x 10
-5
)

PM
10

PM
2.5

PM
2.5

N

442

420

414

Mean

90.8

46.2

5.1

Standard Deviation

38.3

21

1.8

Median

88.7

42.4

5.2

Min

27.1

10.1

.9

Max

173.7

103.4

5.1