Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 2.
Published in final edited form as: Sci Total Environ. 2018 Feb 20;628-629:1497–1507. doi: 10.1016/j.scitotenv.2018.02.138

Maternal exposure to PM2.5 in south Texas, a pilot study

Misti Levy Zamora a, Jairus C Pulczinski a,b, Natalie Johnson b, Rosa Garcia-Hernandez a, Ana Rule a, Genny Carrillo b, Josias Zietsman c, Brenda Sandragorsian d, Suriya Vallamsundar e, Mohammad H Askariyeh c,f, Kirsten Koehler a,*
PMCID: PMC13041583  NIHMSID: NIHMS2157114  PMID: 30045568

Abstract

In this study, we characterized personal exposure to fine particulate matter (PM2.5), black carbon (BC), and nicotine in pregnant women in Hidalgo County, where the hospitalization rates of childhood asthma are the highest in the state of Texas. The measurements were conducted over three non-consecutive sampling days for 17 participants in their third trimester. Measurements were partitioned into four microenvironments, i.e., Residential, Vehicular, Commercial, and Other, on the basis of GPS coordinates and temperature and humidity measurements. The daily average PM2.5 mass concentration was 24.2 (standard deviation = 22.0) μg/m3, with the highest daily mass concentration reaching 126.0 μg/m3. The daily average BC concentration was 1.44 (SD = 0.82) μg/m3, ranging from 0.5 to 5.4 μg/m3. Hair nicotine concentrations were all near the detection level (i.e., 49.2 pg/mg), indicating that the participants were not routinely exposed to tobacco smoke. The Residential microenvironment contributed dominantly to the mass concentration since the participants chiefly remained at home and cooking activities contributed significantly to the total PM2.5. When compared to an ambient monitoring station, the person-specific PM2.5 was frequently more than double the ambient measurement (10.4 μg/m3 overall), revealing that even in regions where ambient concentrations are below national standards, individuals may be still be exposed to elevated PM2.5 mass concentrations.

Keywords: Microenvironments, Air pollution, PM2.5, Indoor air pollution, Black carbon, Nicotine

Graphical Abstract

graphic file with name nihms-2157114-f0001.jpg

1. Introduction

Particulate air pollutants, such as particulate matter smaller than 2.5 μm (PM2.5) and black carbon (BC), are believed to play a key role in the negative health effects and adverse birth outcomes caused by air pollution due to their ability to penetrate deep into the lungs and even translocate beyond the blood-brain barrier (Nemmar et al., 2001; Elder et al., 2006; Johnson et al., 2016; Pope and Dockery, 2006; Bell et al., 2007; Wilhelm et al., 2012). Pregnant women have been identified as a particularly susceptible population; however, little is known about maternal or prenatal exposures to air pollutants at the individual level during this vulnerable period (van den Hooven et al., 2011). Previous findings have linked prenatal PM2.5 exposure to lung function deficits in childhood, but the mechanism remains uncertain (Leon Hsu et al., 2015). In a study conducted of 736 pregnant woman of a similar demographics as our study population (e.g., mostly Hispanic, 12 or fewer years of education, and nonsmokers), increased PM2.5 exposure concentrations were significantly associated with early childhood asthma development in boys, where maternal PM2.5 exposure was estimated over gestation using a satellite-based, spatiotemporally-resolved model (Leon Hsu et al., 2015). Many studies have performed birth outcome analyses based on ambient monitors to make an estimate of personal exposure during pregnancy, but few maternal exposure studies have been conducted with personal monitors (Johnson et al., 2016; Mortimer et al., 2008; Clark et al., 2015; Bell et al., 2010; Hannam et al., 2013; Nethery et al., 2008; Balakrishnan et al., 2015; Perera et al., 2003). Previous studies have shown that fixed ambient monitoring sites may not be a reliable source for accurately estimating personal exposure to air pollutants (Nerriere et al., 2005; Good et al., 2016; Wallace, 2000), so there is a need for more personal exposure measurements during pregnancy to reduce potential exposure misclassification. In one personal monitoring study, 176 pregnant women were monitored for 2 consecutive days and then the infants were followed for 5 years. They determined that prenatal exposure to PM2.5 mass concentrations higher than 52.6 mg/m3 might have negative impacts due to modifications in the development of the fetal lung (Jedrychowski et al., 2010).

Hidalgo County near the US-Mexico border exhibits the highest hospitalization rates of childhood asthma in the state of Texas, but the reason for these regional health disparities have yet to be elucidated (Services TDoSH, 2013a). The 2011 asthma prevalence among children (0–17 years) in Region 11, which comprises 19 counties in lower south Texas, was 11.4% compared to 8.0% for the entire state of Texas (Services TDoSH, 2013a). Hidalgo County also exhibits a higher prematurity rate (14.8%), a birth outcome that has been observed in regions with high levels of pollution, when compared to the state (12.9%) (Services TDoSH, 2013a; Zuniga et al., 2011; Services TDoSH, 2013b; Statistics THDCfH, 2016; Hyder et al., 2014). A goal of this pilot study was to quantify the maternal exposure to PM to determine if personal exposures may provide insight into these trends. In recent years, the ambient PM2.5 mass concentration measured at the stationary monitoring site in Hidalgo County has remained below the national standard (WHO, 2013) however, few studies have been conducted about the indoor air pollutants or other sources of air pollutants in this region (Services TDoSH, 2013a; Zuniga et al., 2011; Gor et al., 2014; TCEQ, 2015). Additionally, recent studies have confirmed that concentrations below the national standard may still be associated with adverse health effects (WHO, 2013; Beelen et al., 2014; Hansen et al., 2006; Shi et al., 2016; Brook et al., 2010; Miller et al., 2007; Kettunen et al., 2007). PM2.5 mass concentrations as low as 6 μg/m3 has been associated with increased mortality for both short- (2 day) and long-term (1 year) exposures (Shi et al., 2016). An analysis of 22 European cohort studies reported that long-term exposure to fine particulate air pollution was associated with increased mortality, even for concentrations well below the present European annual limit of 25 μg/m3, and the threat significantly increased per 5 μg/m3 rise (Beelen et al., 2014). The primary objective of this pilot study was to assess maternal exposures to particulate air pollution at the personal level in this region where the PM2.5 mass concentration is below attainment levels, but the childhood asthma rates are high. Data collected from this study will serve to explore health effects related to prenatal exposure to air pollution in future work and help develop appropriate intervention strategies that reduce childhood asthma in regions similar to Hidalgo County.

2. Methodology

2.1. Study population

The study consisted of 17 volunteers who resided in Hidalgo County and were in their third trimester of pregnancy. Inclusion criteria were the following: between 21 and 35 years of age, healthy (non-asthmatic and non-diabetic), lived in a non-smoking household, singleton pregnancy, had no history of preterm birth, and received prenatal care from Rio Grande Regional Women’s Clinic. The demographics of the participants are exhibited in Table 1. All participants provided written informed consent, and the Texas A&M University Institutional Review Board approved all study procedures. All participants received compensation for their contribution and access to their identifiable data has been restricted.

Table 1.

Participant demographics.

N %
Ethnicity
 Hispanic 17 100
Education
 <12 years 7 41
 12 years 6 35
 >12 years 1 6
Unknown 3 18
Smoking
 Never 12 71
 Before pregnancy 2 12
 Unknown 3 18
Type of home
 Single family home - detached 7 41
 Single family home - attached 1 6
 Mobile home 5 29
 Unknown 4 24
Heating system
 Central heating 4 24
 Single stoves/heaters 3 18
 Electric 7 41
 Gas 0 0
 Unknown 3 18
Cooling system
 Central air conditioning 5 29
 Window units 6 35
 None 1 6
 Unknown 5 29
Primary cooking method
 Electricity 5 29
 Gas 9 53
 No cooking at home 1 6
 Unknown 2 12
Employed
 Yes 2 12
 No 15 88
Open in a new tab

2.2. Study location

The measurements were conducted in the McAllen-Edinburg-Mission region (MEM), the most populated region in Hidalgo County. Hidalgo County is located on the Texas-Mexico border with a population of 842,300 and a population density of 190/km2 (Census US, 2016). This region is one of the fastest-growing counties in the United States, with a 19% increase in just 5 years (Zuniga et al., 2011), and MEM is projected to experience the state’s second highest population growth (23%) by 2020 (Gor et al., 2014; TCEQ, 2015; Quintana et al., 2015). The population is primarily Hispanic (91.3%), followed by Caucasian (5.4%), Asian (1.1%), African American (0.9%), and Native American (0.5%). The population comprising Hidalgo County is largely urban (93%), and the poverty rate is high (41%), 2.3 times the estimated statewide poverty rate (Zuniga et al., 2011).

Between 2010 and 2015, the annual PM2.5 concentration ranged between 9.6 and 11.1 μg/m3 according to measurement at the Continuous Ambient Monitoring Station (CAMS) located in Hidalgo County, with only 2 days in the 5 years exceeding the 24-hour PM2.5 standard of 35 μg/m3 (EPA, 2013; TCEQ, 2016). A previous study determined that area sources, such as residential heating and fuel use, industrial processes, use of consumer products, and gasoline stations, contribute 85% of the PM2.5 mass concentration in the region (TCEQ, 2015). Ambient particulate pollution is also comprised of dust from unpaved roads, emissions from light and heavy-duty on-road vehicles, and heavy use of diesel-powered non-road equipment. The annual averages from 2010 to 2015 for PM10 and Ozone were between 19.2–25.5 μg/m3 and 21–25 ppb, respectively, which are also well below national annual standards (TCEQ, 2015, 2016; EPA, 2016).

2.3. Data collection methods

Sampling took place between June 2015 and April 2016. Each participant was asked to complete three, non-consecutive 24-hour measurement periods. Only one participant did not complete all three sampling days, resulting in 50 sampling days. All three measurements were conducted within a 4–6-week period to reduce seasonal effects on personal exposure. The day before a scheduled prenatal care visit, a local community health worker visited the home of a participant and delivered a lightweight backpack that contained the air sampling equipment detailed below, a global positioning system (GPS) device, and instruments to measure temperature and humidity. At the first home visit, participants answered a questionnaire related to their home, commute, and work environments and background information (e.g., health status, education). After carrying the backpack for approximately 24 h, the participants attended their regularly scheduled prenatal care appointment, at which time they supplied an activity log from the measurement period and provided a urine sample. Participants repeated this process two more times. In addition to providing a urine sample, at the last appointment participants also provided a hair sample to evaluate exposure to environmental tobacco smoke.

2.4. Exposure assessment

The participants were asked to carry a backpack with them as they conducted their typical daily activities. They did not have to continuously wear the backpack and could place it at breathing level when seated indoors, driving, sleeping, using the restroom, or showering. This backpack, which weighed <10 lb, contained a personal DataRAM™ pDR-1200 (Thermo Scientific Corp., Waltham, Mass.) along with an external pump (BGI 400, Mesa Labs, Inc.), a GPS receiver (BT1000XT, Qstarz International, Taiwan), and a HOBO Temperature and Humidity Data Logger (Onset Computer Corporation, Pocasset, MA, USA). Data were logged at 10-s resolution or higher for all instruments. Sampling pumps and inlets were calibrated before and after each 24 hour sampling period and averaged after each run.

A Personal Environmental Monitor (PEM, MSP Inc.) was used as a single-stage impactor PM2.5 inlet (flow rate was 4 L min−1) for the pDR. The pDR is a light-scattering nephelometer with a built-in filter downstream to provide a calibration for mass concentration estimation. The inlet was mounted on the backpack’s shoulder strap to sample air at the participant’s breathing level. A standard 37-mm Teflon filter collected all sampled particles for subsequent analysis and gravimetric calibration (Pall Corporation, Ann Arbor, MI). The Teflon filters were individually housed in a clean Petri dish and shipped in a refrigerated container at −20 °C for analysis at Johns Hopkins University. PM2.5 mea surements were corrected for the non-linear instrument response at RH values >60% (Soneja et al., 2014; Benton-Vitz and Volckens, 2008). The average black carbon concentration for each sample day was determined from the Teflon PM2.5 filters collected downstream of the pDR with a Magee OT-21 SootScan Model Transmissometer (Magee Scientific Corporation, Berkeley, CA, USA). The SootScan utilizes a 2 wave-length light source: 880 nm providing the quantitative measurement of black carbon and a 370 nm for qualitative assessment of certain aromatic organic compounds that are found in tobacco smoke and wood and biomass smoke.

To assess the exposure to secondhand smoke (SHS), the concentration of nicotine was quantified using an isotope dilution gas chromatography–mass spectrometry (GC/MS) method developed by S. Kim et al. (2009) and S.R. Kim et al. (2009). Hair samples were stored in a clean envelope and transported to the laboratory for analysis. About 30 to 50 strands of hair were cut near the root from the back of the scalp. Thirty milligrams of hair were then purified using 3 mL of dichloromethane in a polypropylene centrifuge tube (SARSTEDT, no. 62.554.205) and a sonicator (Aquasonic, Model 250HT), shaken using a horizontal shaker (IKA, KS260 basic), and analyzed using a GC/MS (GC-17/MSQP5000, Shimadzu) in selected ion monitoring (S. Kim et al., 2009; S.R. Kim et al., 2009). The detection limit was 49.2 pg/mg. Since hair growth rate has been calculated at 1.1 cm/month, we were able to quantify nicotine exposure from the previous 2–3 month period by analyzing 3 cm of hair (S. Kim et al., 2009; Al-Delaimy, 2002).

The GPS receiver tracked the participant’s location, which allowed for accurate microenvironment identification. Microenvironments were determined by jointly utilizing the GPS monitor and activity log provided by the participants. The participants were assumed to be driving or riding in a vehicle if the speed was above 6.4 km/h (4 mph). For reference, it has been reported that typical pedestrian walking speeds range from 4.51 km/h (2.80 mph) to 5.43 km/h (3.37 mph) (TranSafety, 1997). None of the participants reported using a bicycle. Participants were determined to be at their home or place of work if the latitude and longitude were within 35 m of the geocoded address, unless the speed was >6.4 km/h, such as when driving to or from the location, or the activity log specified otherwise, such as when one participant indicated she was taking a walk in her neighborhood. This process was found to have a high accuracy in identifying the Home, Work, and Vehicular microenvironments, except in cases where the participant was driving and came to a complete stop (i.e., at a stop sign, red light, or in heavy traffic conditions). To correct for this, we assumed that if the participant was driving in the preceding and succeeding time points, the participant was likely continuously within the Vehicular microenvironment. Other residences and commercial locations were determined by overlaying the GPS tracks on Google Maps and identifying the best-fit microenvironment and comparing with the activity log. For this study, the participant’s home and any other residence were treated as one microenvironment, ‘Residential’, since several participants spent multiple hours at another residence (e.g., a friend or family member’s home) during the day. The ‘Commercial’ microenvironment encompassed any indoor businesses (e.g., a grocery store, drug store, or gas station) or public venues (e.g., a school or church) that the participants visited, and the ‘Other’ microenvironment comprised all the locations that did not qualify as a Vehicular, Commercial, or Residential location, such as a park or an RV lot. Participant compliance was determined by evaluating the frequency of backpack movement. Briefly, the analysis was conducted utilizing the 5-minute GPS variance to determine if the backpack was moved at least once each hour between 8 am and 9 pm. The variance in latitude and longitude between 1:00–6:00 am was employed as a baseline variance because participants were instructed to leave the backpack near their bed while they slept. Overall, we concluded that 86% of the participants followed the study guidelines, with 3 or fewer noncompliant hours. The participant compliance is discussed in more detail in the supplemental text and in Fig. S2.

Identification of the cooking emissions was largely based on the activity log. If the participant did not indicate the time of the meal, significant increases during traditional meal times (i.e., 7–9 am, 11 am–1 pm, or 5–8 pm) were assumed to be associated with cooking emissions. The length of the cooking period was determined by the activity log when available, and if the log was unclear or incompatible with the data, the estimated cooking time was from the first spike in PM mass concentration until the mass concentration decreased to <90% of the peak value. If the participant did not provide information about their meals and mass concentration peaks did not correspond to typical meal times, we did not estimate the cooking contribution. Seven of the 50 measurement periods were excluded from the cooking analysis.

The mass to time ratios, also referred to as the “exposure intensity”, was calculated to elucidate the relative contribution of individual micro-environments to the total exposure (Adams et al., 2009; Braniš and Kolomazníková, 2010; Buonanno et al., 2015; Lim et al., 2012).

The mass to time ratio is determined by the following equation,

IntensityME=ConcentrationMETotalTimeMETotal

where Concentration is the cumulative PM2.5 mass concentration measured in that microenvironment divided by the total PM2.5 mass concentration and Time is the amount of time spent in a given microenvironment compared to the full measurement period. If the ratio is greater than unity, then that microenvironment contributed more to the total PM2.5 exposure than would be expected based on the proportion of time spent in that microenvironment. For example, if a participant spent 80% of her time in the residential microenvironment and her exposure was constant, it would be predicted that 80% of the measured PM2.5 mass concentration would have been observed at that location; however, this is frequently not the case, as certain activities will have disproportionate contributions to the cumulative PM2.5 mass concentration.

In addition to personal monitoring, concurrent hourly averaged PM2.5 mass concentrations were obtained from an ambient monitoring site operated and maintained by the Texas Commission on Environmental Quality (TCEQ) located in Hidalgo County (site number 48-215-0043, latitude: 26.226210°, longitude: −98.291069°).

2.5. Statistical methods

Linear mixed models were developed to examine the associations between personal exposures and explanatory covariates using a random intercept for each participant to account for possible within-subject correlation:

Yi,j=μ+bi+βiCi,j+ei,j (Model 1)

In Model 1, Yi,j is the mean personal PM2.5 or cumulative BC mass concentration, for participant i and replicate j, μ is the overall mean exposure, bi is the participant-specific mean, ei,j is the unexplained error, and an unstructured covariance structure was specified. β represents a fixed-effects coefficient for covariate, C. Several covariates were considered. The potential correlation within participant was accounted for with random intercepts by participant ID, since multiple measurements per participant were used in the analysis. No repeated effect was specified. A solution for the fixed-effects parameters was produced. The confidence intervals for all of the parameters were constructed around the estimate by using a radius of the standard error times a percentage point from the t distribution. Model 1 was used to establish the relationship between the reported traffic conditions, measurements from the TCEQ ambient site, or presence of kitchen vents and the mean personal PM2.5 mass concentration and between cooking activities and the overall cumulative PM2.5 mass concentration. All covariates were considered in separate, univariate analyses.

Model 2 was used to establish the relationship between the known covariates (e.g., cooking emissions, the amount of time spent driving, amount of time at a commercial location, and time at home) and the cumulative BC concentration, where all covariates were considered in separate, univariate analyses:

Yi,j=μ+bi+βi+aiCi,j+ei,j (Model 2)

In Model 2, Yi,j is the cumulative BC concentration, for participant i and replicate j, μ is the overall mean exposure, bi is the participant-specific mean, ei,j is the unexplained error, and an unstructured covariance structure was specified. β represents a fixed-effects coefficient for covariate, C, and ai represents a random slope for covariate, C. For the random effects, we utilized intercepts for participants for the effect of all the exposure determinants on BC, as well as, by-participants random slopes for the effect of cooking emissions on BC. For all associations considered with Model 1, we also evaluated the influence of accounting for random slopes with Model 2, but found nearly equivalent results for all analyses except for the effect of cooking emissions on the BC concentrations.

Pearson correlation coefficients were also calculated on within-person replicates for the BC mass concentrations with potential sources (i.e., BC concentration vs time spent in each microenvironment, the amount of time cooking, and time spent at home not cooking). Participant 10 was excluded from the correlation coefficient analysis since she only completed two sampling days.

For all models, visual inspection of residual plots did not reveal any obvious deviations from homoscedasticity or normality. p-Values were obtained by likelihood ratio tests of the full model with the effect in question against the model without the effect in question. The measured pollutants were not normally distributed and were log-transformed for all analyses. All analyses were completed with R (Team RC, 2016), and the lme4 (Bates et al., 2015) package was utilized to perform a linear mixed effects analysis of the relationship between the mass concentrations and the exposure determinants.

3. Results

The PM2.5 mass concentration (μg/m3) plotted as a function of time and as a function of GPS coordinates from one sampling day is shown in Fig. 1, and the PM2.5 mass concentrations (μg/m3) as a function of GPS coordinates for all 17 participants are shown in Fig. 2. The daily average mass concentration from the 50 measurement days ranged between 1.9 and 126.0 μg/m3, with an average PM2.5 mass concentration of 24.2 μg/m3 (Table 2). The Residential microenvironment was the most prevalent; participants spent 86% of their time in this microenvironment, on average. The Commercial microenvironment accounted for, on average, 9% of the sampling period, followed by the Vehicular and Other microenvironments at 5 and 0.1%, respectively (Fig. 3). The time-weighted average PM2.5 mass concentration was greatest in the Residential microenvironment for 8 out of 17 of the participants and for the study average over all participants (25.6 μg/m3, Table 2). Typically, the Residential mass concentrations were lowest overnight, and the greatest mass concentrations were observed during meal times. The overall average PM2.5 mass concentrations for the Vehicular, Commercial, and Other categories were 19.5, 15.6, and 2.0 μg/m3, respectively. The time-weighted average PM2.5 mass concentration was greatest in the Vehicular microenvironment for 6 out of 17 participants. The remaining three participants were exposed to the highest time-weighted average mass concentrations in the Commercial microenvironment.

Fig. 1.

Fig. 1.

Open in a new tab

A) The temporal variations of the PM2.5 mass concentration (μg/m3) from one sampling day. The colors correspond with the points in panel B. All times are shown in the local time zone (CST). B) The mass concentration plotted as a function of GPS coordinates. The size of the point corresponds to the mass concentration, and the shape indicates microenvironment, i.e., Residential (blue diamond), Commercial (yellow star), Vehicular (circles; colored by trip), and Other (pink triangle).

Fig. 2.

Fig. 2.

Open in a new tab

The PM2.5 mass concentration (μg/m3) as a function of GPS coordinates for all 17 participants. Line color corresponds to the measured mass concentrations, with warmer colors corresponding to higher values and cooler colors indicating lower mass concentrations. The red dot indicates the location of the TCEQ ambient monitoring site.

Table 2.

The average, median, and 1-min peak PM2.5 mass concentrations from each participant, and the mass concentrations and mass-time ratio of exposures in the four microenvironments for each participant during the three sampling periods. Data is presented as mean ± SD.

Participant Number of samples Daily average mass conc. (μg/m3) Daily median mass conc. (μg/m3) Peak 1-min mass conc. (μg/m3) Residential (μg/m3)/mass-time ratio Vehicular (μg/m3)/mass-time ratio Commercial (μg/m3)/mass-time ratio Other (μg/m3)/mass-time ratio
1 3 5.7 ± 3.4 5.0 ± 2.3 132.4 5.8 ± 3.5/1.03 4.4 ± 3.1/0.99 5.2 ± 2.8/0.8 0.8 ± 0.2/0.62
2 3 29.4 ± 2.3 21.3 ± 9.4 1136.4 28.8 ± 46.8/0.97 66.7 ± 134.7/0.8 26.6 ± 13.1/2.08 –
3 3 52.4 ± 16.2 29.2 ± 17.6 384 54.1 ± 64.2/1.02 24.6 ± 21.7/0.59 21.4 ± 18.8/0.72 33.1 ± 57.0/1.66
4 3 7.7 ± 1.7 6.0 ± 2.5 168.3 7.7 ± 8.5/1.00 9.6 ± 4.9/1.09 6.6 ± 4.9/0.91 –
5 3 22.8 ± 23.3 9.9 ± 9.9 1760.7 23.9 ± 94.9/1.05 7.0 ± 7.3/0.35 7.3 ± 3.7/0.35 –
6 3 24.8 ± 7.5 14.4 ± 6.4 408 23.9 ± 33.6/1.01 32.9 ± 14.1/0.87 20.1 ± 7.6/0.97 –
7 3 17.8 ± 11.2 7.6 ± 6.5 440.2 19.9 ± 45.8/1.09 10.2 ± 12.6/0.48 9.1 ± 1.6/0.6 –
8 3 41.9 ± 86.6 17.0 ± 19.6 1510.7 41.9 ± 106.1/0.95 37.9 ± 71.1/2.82 50.1 ± 78.1/1.35 –
9 3 91.2 ± 60.2 10.3 ± 8.6 2701.3 108.5 ± 292.7/1.13 13.8 ± 31.5/0.08 8.6 ± 17/0.2 –
10 2 13.9 ± 2.1 6.4 ± 2.8 138.3 15.0 ± 23.4/1.09 6.6 ± 7.9/0.65 6.2 ± 5.8/0.58 –
11 3 33.6 ± 31.4 6.9 ± 3.1 2922.8 36.5 ± 115.7/1.04 16.8 ± 19/0.76 22.8 ± 22.9/0.61 –
12 3 8.0 ± 6.8 3.0 ± 2.3 324.7 6.5 ± 16.4/0.86 6.0 ± 8.9/1.36 10.4 ± 14.5/0.76 –
13 3 4.7 ± 2.1 4.1 ± 2.0 70.5 4.6 ± 4.9/0.97 3.7 ± 2.3/1.58 6.6 ± 2.9/0.98 –
14 3 21.7 ± 2.5 11.2 ± 8.8 3477.9 22.2 ± 101.9/1.03 23.1 ± 73.6/0.63 12.2 ± 6.9/1.23 –
15 3 8.5 ± 4.0 5.4 ± 3.4 215.7 8.3 ± 13.3/0.98 18.3 ± 36.4/0.81 6.9 ± 3.4/1.95 –
16 3 5.8 ± 4.6 3.3 ± 1.9 165.3 6.2 ± 13.3/1.02 3.8 ± 1.9/0.93 4.4 ± 2.5/0.87 –
17 3 20.8 ± 8.7 10.1 ± 3.6 298.5 20.4 ± 37.4/0.95 45.8 ± 35.3/1.24 32.6 ± 21.7/1.86 –
Overall 2.9 24.2 ± 22.0 10.1 ± 6.5 956.2 ± 1115.6 25.6 ± 25.6/1.01 ± 0.06 19.5 ± 17.56/0.94 ± 0.61 15.57 ± 12.2/0.99 ± 0.54 2.0 ± 8.0/0.91 ± 0.66
Open in a new tab

Fig. 3.

Fig. 3.

Open in a new tab

The cumulative mass concentration (μg/m3) from the 17 participants apportioned into four microenvironments: Residential (blue), which includes the participant’s home and other residences, Commercial (yellow), Vehicular (green), and Other (red).

The mass to time ratios revealed that the Residential microenvironment contributed slightly more to the total cumulative mass concentration than would be predicted based on the division of time (i.e., mass to time ratio > 1) (Table 2); whereas, the other three microenvironments contributed less to total exposure (<1). There are a few notable exceptions when analyzing the person-specific results. Participants 8, 12, 13, and 17 demonstrated much higher mass-time contributions from the Vehicular microenvironment, and the measurements from Participant 3 demonstrated a higher contribution from the Other microenvironment when she visited an RV park. The Commercial microenvironment exhibited a broad range of mass-time ratios (0.20 to 2.08) reflecting that this category includes a large spectrum of businesses, such as markets, grocery stores, fast food venues, and places of worship.

Participants who reported ‘heavy traffic’ near their residences (n = 4) exhibited the greatest average mass concentrations (29.2 μg/m3) in this microenvironment, whereas, participants who indicated “light traffic” near their residences (n = 6) exhibited an average mass concentrations of 16.0 μg/m3 (t-test, p-value = 0.1). To evaluate the accuracy of the participant’s perception of the traffic volume near their homes, a comparison of the traffic volumes measured in 2014 and 2015 by the Texas Department of Transportation (TxDOT) with the reported traffic intensity by the participants was conducted (Fig. S1). Visually, the reported volumes appear to agree with the TxDOT observations, as the women living near the major roads generally reported higher traffic volumes, and the overall Pearson correlation coefficient was 0.50.

Cooking emissions were found to be a significant PM2.5 source for many of the participants (Fig. 4). The average PM2.5 mass concentration during the identified cooking periods was 111.3 μg/m3, with 1 min peak exposures exceeding 700 μg/m3. On average, 27% of the cumulative PM2.5 (Braniš et al., 2010) mass concentration could be attributed to emissions during meal times; however, the contribution was over 50% for several of the sampling days. In addition, differences in the cumulative mass (Fig. 4) were strongly correlated with variations in cooking emissions (p-value < 0.0001) indicating that the cooking activities can have a substantial impact on an individual’s exposure. An analysis of the method of cooking (e.g., frying, sautéing, grilling, or type of stove) was not found to be significantly associated with cooking mass concentrations. The daily average mass concentration in kitchens with a working fan was 11.9 μg/m3 (n = 3), compared to 20.9 μg/m3 in kitchens that did not have a working fan (n = 10). However, we did not ask the participants to indicate if the fan was used, so although the results are suggestive, we are unable to determine how influential this factor was on personal exposures. The remaining participants did not provide information about the status of their kitchen fans (n = 4).

Fig. 4.

Fig. 4.

Open in a new tab

Contribution (%) of the four microenvironments and cooking activities to the cumulative PM2.5 mass concentrations.

The daily average BC mass concentrations from the 50 sampling days ranged between 0.5 and 5.4 μg/m3, with an average mass concentration of 1.4 μg/m3 (Table 3). On average, BC accounted for 12% of the PM2.5 mass concentration. Participant 1 exhibited the highest BC percent contribution to PM2.5 at 35% (Fig. 5). She was the only participant who indicated that she utilized the city bus systems, so it is possible that she was exposed to higher concentrations of BC from the bus’s diesel exhaust. There does not appear to be a common, dominant BC source for the participants. Two of the participants (Participant 5 and Participant 7) exhibited strong correlations between the measured BC mass concentrations and estimated cooking emissions (i.e., r2 = 0.66, 0.98, respectively), whereas, five participants (Participants 3, 8, 11, 14, and 15) exhibited positive correlations (r2 = 0.51, 0.98, 0.89, 0.92, and 0.60, respectively) with the amount of time in their vehicle. In addition, three participants (Participants 9, 13, and 17) exhibited the strongest correlations between BC and the time spent at their residence excluding cooking activities (r2 = 0.82, 0.61, 0.81, respectively), suggesting that another source in or around the home may be the main BC source for these participants.

Table 3.

The time-weighted average black carbon (BC) mass concentration (μg/m3) per participant, and the ratio of the BC mass concentration to the total PM2.5 mass concentration (%). Data is presented as mean ± SD.

Participant Number of samples Black carbon concentration (μg/m3) BC (%)
1 3 2.5 ± 1.0 35 ± 16
2 3 2.0 ± 0.5 6 ± 1
3 3 1.4 ± 0.5 2 ± 1
4 3 1.2 ± 0.5 15 ± 7
5 3 1.3 ± 0.5 6 ± 0
6 3 2.9 ± 2.3 11 ± 8
7 3 1.0 ± 0.2 6 ± 3
8 3 1.1 ± 0.3 11 ± 14
9 3 1.4 ± 0.7 2 ± 1
10 2 0.9 ± 0.1 6 ± 0
11 3 1.8 ± 0.4 8 ± 5
12 3 1.0 ± 0.2 20 ± 14
13 3 1.1 ± 0.1 29 ± 18
14 3 1.5 ± 0.7 7 ± 2
15 3 1.1 ± 0.2 13 ± 3
16 3 0.8 ± 0.3 17 ± 7
17 3 1.2 ± 0.1 6 ± 2
Overall 2.9 1.4 ± 0.5 12 ± 6
Open in a new tab

Fig. 5.

Fig. 5.

Open in a new tab

The BC mass concentration (black bar, left axis) and the percent contribution of black carbon to the PM2.5 mass concentration (blue bars, right axis).

The average of the hair nicotine concentration was 50.2 (SD = 4) pg/mg, with 7 of the participants exhibiting concentrations below the detection level (49.2 pg/mg). The nicotine concentrations in the other 10 participants were <62 pg/mg.

We compared the daily average PM2.5 mass concentrations from the personal exposure measurements with the ambient stationary monitoring site that is located about 8 miles from the Rio Grande Regional Women’s Clinic (Fig. 6). The site is within 20 miles from all women’s home addresses. The average PM2.5 mass concentration measured at the stationary site from the corresponding dates of our personal measurements was 10.4 μg/m3. 74% of the ambient measurement days were considered ‘Good’, which is the best rating on the air quality index scale (PM2.5 < 12.1 μg/m3), and 26% were at the ‘Moderate’ levels (12.0 μg/m3 < PM2.5 < 35.4 μg/m3) (TCEQ, 2016; EPA, 2016). Overall, the stationary site observed lower PM2.5 mass concentrations compared to personal exposures, with only 13 out of the 50 sampling days exhibiting higher mass concentrations than found for personal exposures. Interestingly, 25 out of the 50 personal exposure measurement days exhibited average mass concentrations that were more than double the mass concentration measured at the ambient site. An analysis between the personal exposure and the fixed ambient site revealed no significant association between the personal exposures and ambient concentrations.

Fig. 6.

Fig. 6.

Open in a new tab

Comparison of average PM2.5 mass concentrations measured by our personal monitors (pDR, red) and at the ambient monitoring station located in Hidalgo County (dark grey).

4. Discussion

In this pilot study, we analyzed the person-specific exposures to PM2.5, BC, and nicotine during three non-consecutive 24-hour periods for 17 pregnant women in their third trimester. Our study-average PM2.5 mass concentration was comparable with previously reported personal monitoring of non-smoking households. The average PM2.5 mass concentrations from population-based studies in Basel, Switzerland, Detroit, MI, and Baltimore, MD were 17.5 μg/m3, 17.6, and 25.87 μg/m3, respectively, compared to 24.2 μg/m3 found in this study (Lim et al., 2012; Oglesby et al., 2000; Rodes et al., 2010; Breysse et al., 2005). Because the activities a person chooses to participate impact the timing, location, and degree of pollutant exposure, they play a crucial role in explaining exposure variations. We examined the temporal and microenvironmental variations of the PM2.5 mass concentrations and determined that the Residential microenvironment contributed dominantly to the cumulative mass concentrations. This was due to both the amount of time spent in the home compared with the other microenvironments and the high mass concentrations observed in this environment. Common residential sources include cooking, cleaning, the burning of candles or incense, cigarette smoke (though not observed in this study), and the infiltration of outdoor pollutants (Braniš and Kolomazníková, 2010; Buonanno et al., 2015; Breysse et al., 2005; Buonanno et al., 2009; Cao et al., 2012, 2005; Dennekamp et al., 2002; Wan et al., 2011). The mass concentrations observed during meal times were generally the largest mass concentrations observed in a given sampling day and frequently exceeded 100 μg/m3. Buonanno et al. also report that the greatest mass concentrations for their female cohort were also observed during periods of cooking (Buonanno et al., 2009). Their cohort came from a variety of lifestyles, such as students, workers, and homemakers, and had an average age of 42. The correlation between the cumulative PM2.5 and cooking mass concentrations in our study suggests that the variations in the total mass concentration between the three 24-hour periods for a given participant were partially governed on the amount of cooking activities performed by the participant during that sampling day. Previous work focusing on cooking emissions have found that the method of cooking (e.g., frying, boiling, or grilling), the fuel type (e.g., electric stove, biomass, or kerosene), and the food cooked (e.g., fatty foods vs vegetables) considerably affect the mass concentration in that microenvironment (Wan et al., 2011; Pokhrel et al., 2015). The participants in our study used electric or gas stoves, which have been found to increase the mass concentration by up to 20–40 fold in the kitchen and 10-fold in adjacent rooms compared to background concentrations (Wan et al., 2011). Vehicle emissions contributed considerably to the exposures of several participants; however, our participants comparatively spent much less time in this microenvironment, so the overall contribution to the cumulative PM2.5 mass concentration was generally small compared to the Residential microenvironment.

We did not request any additional information about activities conducted in the microenvironments so we are unable to comment on peaks that were observed during non-meal times; however, the National Human Activity Pattern Survey (NHAPS) performed a telephone-based survey of 9386 Americans that characterized exposure related activities over broad geographical and age scales in the United States (Klepeis et al., 2001a). NHAPS participants reported spending an average of 87% of their time in enclosed buildings and about 6% of their time in vehicles during the 24-hour period prior to when the participant completed the interview. The percentage of time spent indoors, outdoors, and in vehicles was fairly consistent across people in different regions of the U.S. The NHAPS exposure activities were grouped into eight categories: 1) cooking/food preparation; 2) laundry/dishes/cleaning kitchen; 3) housekeeping; 4) bathing/showering/washing/using bathroom; 5) yardwork/gardening/car- or house-maintenance; 6) sports/exercise; 7) eating/drinking, and 8) other activities (Klepeis et al., 1996). It was determined that the exposure related activities conducted during that 24-hour period were most frequently comprised of consuming meals at 35.82%, followed by exercise at 13.26%, housekeeping at 12.42%, food preparation at 11.81%, maintenance at 11.28%, bathing at 9.38%, and kitchen cleaning at 6.02%. The “other” category was excluded from the analysis since the exposure sources were unknown or too diverse. The authors also did not include vehicle-related activities in this comparison.

Typically, the highest average mass-time ratios over all participants have been reported to occur during transport and at restaurants/bars (Buonanno et al., 2015; Breysse et al., 2005; Bekö et al., 2015). The high ratios in restaurants and bars were stated to be due to the frequent cooking and the presence of smoking (Buonanno et al., 2015; Breysse et al., 2005). Mass-time ratios while driving have been frequently reported to exceed 3. The mass-ratios in the home have varied significantly depending on the study participants. If the participants did not cook frequently in their residence, the mass-time ratios were generally less than unity (Braniš and Kolomazníková, 2010), but if the subjects cooked at home, the mass-time ratios were typically greater than unity (Buonanno et al., 2015; Braniš et al., 2010; Buonanno et al., 2009). Mass-time ratios during periods of shopping were found to be generally less than unity (Buonanno et al., 2015). The Residential micro-environment was the only location in our study where the study-average mass-time ratio was greater than unity, but the highest individual mass-time ratios were observed in the commercial and vehicular microenvironments. It is possible that the mass-time ratios in the Vehicular microenvironment for our study were generally lower than previous studies since the majority of our participants drove outside of peak traffic times.

The mass concentration was nearly twice as high in residences in which the participants reported heavy traffic near the home compared to participants who reported light traffic, suggesting that the influx of outdoor pollutants may contribute to residential mass concentrations. It is interesting to note that one participant in our study reported ‘light traffic’ near her home, when the back of her residence is <0.35 km (~0.2 miles) from a major highway. A closer inspection revealed that she lives on a road that likely does not receive much direct traffic, which highlights how an individual’s perception may not be a reliable metric when evaluating personal exposure to ambient pollution sources.

The “Relationships of Indoor, Outdoor, and Personal Air” study was conducted in approximately 300 nonsmoking households in Houston, Texas; near New York City, New York; or Los Angeles County, California (Weisel et al., 2005). This study measured the indoor, outdoor, and personal PM2.5 mass concentrations and the indoor/out air exchange rates in homes with different structures. Between the three cities, differences in indoor and outdoor PM2.5 levels were minimal, but differences in personal PM2.5 exposures were more pronounced. The authors estimated that outdoor air contributed about 60% of the indoor PM2.5 mass concentration. Cao et al. coupled indoor and outdoor PM2.5 monitors at homes in Hong Kong located at roadside, urban, and rural environments (Cao et al., 2005). They found that the indoor PM2.5 mass concentration was greatest in the homes located near the roadside, followed by homes in the urban environment, and the rural homes exhibited the lowest average indoor PM2.5 mass concentrations. The ratio of the indoor to outdoor PM2.5 was also greatest at the rural locations, suggesting that outdoor pollutant sources may be less influential in this environment.

Nicotine concentrations were all below or near the detection limit signifying that the participants were not routinely exposed to SHS in the last three months and likely lived in non-smoking residences with minimal exposure to environmental tobacco smoke (Table S1). Previous studies have reported that the average hair nicotine concentration for nonsmoking women who were exposed to SHS at home is 440 pg/mg or greater (S. Kim et al., 2009). We expected the participants’ exposure to nicotine to be low since they all reported living in nonsmoking homes and smoking rates among Hispanic populations tend to be low (Goldman, 2016). In the US, 9.9% of Hispanic adults are current smokers, compared to 17.4% of Caucasians and 16.8% of African Americans (Clarke et al., 2015). The maternal smoking rate in Hidalgo County (0.35%) is also markedly lower than the statewide average (5.18%) (Statistics THDCfH, 2016).

The daily average BC concentration was 1.44 ± 0.82 μg/m3, and BC accounted for 12% of the PM2.5 mass concentration on average. For comparison, a study conducted in Guangzhou, China found that about 9% of indoor PM2.5 was comprised of BC (Cao et al., 2012), and ambient measurements at the US-Mexico border near Tijuana, where cars are frequently idling for extended periods of time, reported that BC accounted for 19% of the total PM2.5 (Takahama et al., 2014; Levy et al., 2014). LaRosa et al. compared the BC mass concentration inside and outside a suburban Virginia house occupied by two nonsmokers. The main outdoor source of BC was due to regional background sources, contributing about 84% of the total mass concentration during the two-year study. Rush-hour traffic and seasonal wood burning were determined to be responsible for a majority of the remaining 16%. The main indoor sources of BC were cooking (16%) and candle burning (31%), and increased BC concentrations could be detected indoors during peak traffic hours and periods of outdoor wood burning, highlighting the influence of outdoor BC sources on indoor concentrations (LaRosa et al., 2002). However, the paired indoor/outdoor measurements at roadside, urban, and rural homes conducted by Cao et al. (Cao et al., 2005) determined that there were no significant indoor BC sources and that the majority of indoor BC mass concentrations could be attributed to outdoor sources. The indoor BC mass concentrations were significantly higher in roadside homes (4.0 μg/m3) than in the urban (2.1 μg/m3) and rural (1.8 μg/m3) homes (Cao et al., 2005). Since vehicles are a dominant contributor to BC, it is often used as a surrogate for exposure to diesel exhaust and other combustion processes (LaRosa et al., 2002). Our results suggest that cooking is an important contribution to personal BC exposures.

An important factor was that the majority of our participants were not employed, which contributed to the large amount of time spent in their residences. In addition, the women who were employed had two scheduled appointments (i.e., receiving and returning the personal monitors), so it is possible that the distribution of exposures in this study was not representative of pregnant American women more generally. It would be expected that if the participants worked full-time outside of the home that the residential microenvironment would likely contribute less. NHAPS found that the overall US population spent 72.4% of their time at their residence (Klepeis et al., 2001b). This was higher for females (75.4%) than males (69.0%) and for preschool aged children (89.9%) and the retired (85.3%) than those employed (66.6%) (Klepeis et al., 1996). Dons et al. measured the personal exposure to BC in 8 couples of which one person was a full-time worker and the other was a homemaker. The exposure differed between partners by up to 30%, despite the fact that they lived at the same location. It was determined that the greatest difference was due to the amount of time spent in a vehicle, likely from commuting to work (Dons et al., 2011).

Many exposure assessments use fixed regional monitors to estimate the exposure to pollutants for a given population. This is particularly true for epidemiological studies that endeavor to quantify the long-term health effects of prenatal pollutant exposure. It has been reported that fixed monitoring sites are more representative for homes with no significant cooking activities (Braniš and Kolomazníková, 2010) and for non-smoking homes with central air-conditioning systems, which leads to higher outdoor-to-indoor particle penetration ratios (Buonanno et al., 2015; Adgate et al., 2002). When compared to the ambient monitoring station located in Hidalgo County, the personal PM2.5 mass concentrations were higher for 76% of the sampling days. Moreover, 25 out of the 50 personal exposure measurements exhibited more than double the mass concentration measured at the ambient site, and nine sampling days exhibited more than a 5-fold difference in the daily averaged PM2.5 mass concentrations. No significant association between personal exposures and ambient concentrations was found in this study, which may suggest that even in regions where ambient PM2.5 concentrations are below national standards, exposure to PM2.5 may be substantially higher than previous estimates and that the fixed ambient measurements would not be a reliable proxy for estimating personal exposures for this population.

4.1. Study limitations

The major limitation of this study was the low participant interest, which resulted in only 17 participants. Although, our analyses suggest that the participants were generally compliant, and only one participant did not complete all three rounds. The use of self-report activity time sheets could have led to some biases. We also did not request any additional information about activities conducted in the microenvironments, which limits our ability to identify potentially influential peaks that were observed during non-meal times. Lastly, the findings may not be generalizable to the community since these women were mostly unemployed and they had two appointments during each 24-hour window (i.e., receiving the backpack and their prenatal appointment), which may not be representative of their typical day.

5. Conclusions

This pilot study has demonstrated the utility of personal exposure monitoring in a previously understudied region in Hidalgo County, Texas. The primary objective of this pilot study was to assess maternal exposure to particulate air pollution in this region where the ambient PM2.5 mass concentration meets EPA attainment levels, but the asthma rates are high. The daily average mass concentration from the 50 measurement days ranged between 1.9 and 126.0 μg/m3, with an average PM2.5 mass concentration of 24.2 μg/m3. The time-weighted average PM2.5 mass concentration was greatest in the Residential microenvironment for 8 out of 17 of the participants. Vehicle emissions contributed considerably to the exposures of several participants; however, participants comparatively spent much less time in this microenvironment, so the overall contribution to the cumulative PM2.5 mass concentration was generally minor compared to the Residential microenvironment. Cooking emissions were found to be a significant PM2.5 source for many of the participants. Overall, the nearby stationary site observed lower PM2.5 mass concentrations compared to personal exposures, and 25 out of the 50 personal exposure measurement days exhibited average mass concentrations that were more than double the mass concentration measured at the ambient site. An analysis between the personal exposure and the fixed ambient site revealed no significant association between the personal exposures and ambient concentrations. Given the high rates of childhood asthma and pre-term birth in this region, future work should further investigate the relationship between measured personal exposures and those estimated from ambient monitoring networks and the impact of variable toxicity between these types of pollution.

Supplementary Material

Sup mat

HIGHLIGHTS.

  • Our results demonstrate that the women were exposed to the greatest mass concentrations when in the Residential microenvironment.

  • Compared to an ambient monitoring station, the person-specific PM2.5 was frequently more than double the observed mass concentration.

  • This study has demonstrated that individuals may be still be exposed to elevated PM2.5 mass concentrations even in regions where ambient concentrations are below national standards.

Acknowledgements

This research was supported by a seed grant from the Texas A&M Transportation Institute and the Texas A&M Health Science Center. J. Pulczinski was also supported by a Texas A&M One Health Initiative Summer Research Program. R. Garcia-Hernandez was supported by the Johns Hopkins University Pulmonary and Critical Care Medicine Summer Internship Program and the SIP grant (NIH HL084762). The authors would like to thank Catherine Hess for her contribution to the exposure monitoring. N. Johnson would like to acknowledge the Community Health Workers at Texas A&M University Colonias Program for their contribution to the project. M. Levy Zamora would like to acknowledge the Johns Hopkins Institute for Clinical and Translational Research (Grant Number UL1TR001079) for the contribution to the statistical analysis and Andrew Patton for his contribution to the TxDOT analysis.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.scitotenv.2018.02.138.

Footnotes

Conflict of interest

Authors declare no competing financial interests in relation to the work described.

References

  1. Adams C, Riggs P, Volckens J, 2009. Development of a method for personal, spatiotemporal exposure assessment. J. Environ. Monit 11, 1331–1339. [DOI] [PubMed] [Google Scholar]
  2. Adgate JL, Ramachandran G, Pratt G, Waller L, Sexton K, 2002. Spatial and temporal variability in outdoor, indoor, and personal PM2.5 exposure. Atmos. Environ 36 (20), 3255–3265. [Google Scholar]
  3. Al-Delaimy W, 2002. Hair as a biomarker for exposure to tobacco smoke. Tob. Control 11 (3), 176–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Balakrishnan K, Sambandam S, Ghosh S, Mukhopadhyay K, Vaswani M, Arora NK, et al. , 2015. Household air pollution exposures of pregnant women receiving advanced combustion cookstoves in India: implications for intervention. Ann. Glob. Health 81 (3), 375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bates D, Mächler M, Bolker B, Walker S, 2015. Fitting linear mixed-effects models using lme4. J. Stat. Softw 67 (1), 1–48. [Google Scholar]
  6. Beelen R, Raaschou-Nielsen O, Stafoggia M, Andersen ZJ, Weinmayr G, Hoffmann B, et al. , 2014. Effects of long-term exposure to air pollution on natural-cause mortality: an analysis of 22 European cohorts within the multicentre ESCAPE project. Lancet 383 (9919), 785–795. [DOI] [PubMed] [Google Scholar]
  7. Bekö G, Kjeldsen BU, Olsen Y, Schipperijn J, Wierzbicka A, Karottki DG, et al. , 2015. Contribution of various microenvironments to the daily personal exposure to ultrafine particles: personal monitoring coupled with GPS tracking. Atmos. Environ 110, 122–129. [Google Scholar]
  8. Bell ML, Ebisu K, Belanger K, 2007. Ambient air pollution and low birth weight in Connecticut and Massachusetts. Environ. Health Perspect 1118–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bell ML, Belanger K, Ebisu K, Gent JF, Lee HJ, Koutrakis P, et al. , 2010. Prenatal exposure to fine particulate matter and birth weight: variations by particulate constituents and sources. Epidemiology 21 (6), 884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Benton-Vitz K, Volckens J, 2008. Evaluation of the pDR-1200 Real-time Aerosol Monitor. pp. 353–359. [DOI] [PubMed] [Google Scholar]
  11. Braniš M, Kolomazníková J, 2010. Monitoring of long-term personal exposure to fine particulate matter (PM2.5). Air Qual. Atmos. Health 3 (4), 235–243. [Google Scholar]
  12. Braniš M, Vyškovská J, Malý M, Hovorka J, 2010. Association of size-resolved number concentrations of particulate matter with cardiovascular and respiratory hospital admissions and mortality in Prague, Czech Republic. Inhal. Toxicol 22 (sup2), 21–28. [DOI] [PubMed] [Google Scholar]
  13. Breysse PN, Buckley TJ, Williams DA, Beck CM, Jo S-J, Merriman B, et al. , 2005. Indoor exposures to air pollutants and allergens in the homes of asthmatic children in inner-city Baltimore. Environ. Res 98 (2), 167–176. [DOI] [PubMed] [Google Scholar]
  14. Brook RD, Rajagopalan S, Pope CA, Brook JR, Bhatnagar A, Diez-Roux AV, et al. , 2010. Particulate matter air pollution and cardiovascular disease an update to the scientific statement from the American Heart Association. Circulation 121 (21), 2331–2378. [DOI] [PubMed] [Google Scholar]
  15. Buonanno G, Morawska L, Stabile L, 2009. Particle emission factors during cooking activities. Atmos. Environ 43 (20), 3235–3242. [Google Scholar]
  16. Buonanno G, Fuoco FC, Russi A, Stabile L, 2015. Individual exposure of women to fine and coarse PM. Environ. Eng. Manag. J 14 (4), 827–836. [Google Scholar]
  17. Cao J, Lee S, Chow J, Cheng Y, Ho K, Fung K, et al. , 2005. Indoor/outdoor relationships for PM2.5 and associated carbonaceous pollutants at residential homes in Hong Kong–case study. Indoor Air 15 (3), 197–204. [DOI] [PubMed] [Google Scholar]
  18. Cao J, Huang H, Lee S, Chow JC, Zou C, Ho K, et al. , 2012. Indoor/outdoor relationships for organic and elemental carbon in PM2.5 at residential homes in Guangzhou, China. Aerosol Air Qual. Res 12 (5), 902–910. [Google Scholar]
  19. Census US, 2016. QuickFacts - Hidalgo County, Texas. Available from:. http://www.census.gov/quickfacts/table/RHI305210/48215 (cited 12 July 2016).
  20. Clark NA, Demers PA, Karr CJ, Koehoorn M, Lencar C, Tamburic L, et al. , 2015. Effect of Early Life Exposure to Air Pollution on Development of Childhood Asthma. University of British Columbia. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Clarke TC, Ward BW, Freeman G, Schiller JS, September 2015. Early Release of Selected Estimates Based on Data From the January–March 2015 National Health Interview Surveyx. 201(5). National Center for Health Statistics. [Google Scholar]
  22. Dennekamp M, Mehenni O, Cherrie J, Seaton A, 2002. Exposure to ultrafine particles and PM2.5 in different micro-environments. Ann. Occup. Hyg 46 (Suppl. 1), 412–414. [Google Scholar]
  23. Dons E, Panis LI, Van Poppel M, Theunis J, Willems H, Torfs R, et al. , 2011. Impact of time–activity patterns on personal exposure to black carbon. Atmos. Environ 45 (21), 3594–3602. [Google Scholar]
  24. Elder A, Gelein R, Silva V, Feikert T, Opanashuk L, Carter J, et al. , 2006. Translocation of inhaled ultrafine manganese oxide particles to the central nervous system. Environ. Health Perspect 1172–1178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. EPA, 2013. Criteria Air Pollutants National Ambient Air Quality Standards (NAAQS) Table. Available from:. https://www.epa.gov/criteria-air-pollutants/naaqs-table (updated July 5, 2016; cited 2016 July 25, 2016).
  26. EPA, 2016. Outdoor Air Quality Data. Available from:. https://www.epa.gov/outdoor-air-quality-data/air-data-concentration-plot (updated September 14, 2016).
  27. Goldman N, 2016. Will the Latino mortality advantage endure? Res. Aging 38 (3), 263–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Good N, Mölter A, Ackerson C, Bachand A, Carpenter T, Clark ML, et al. , 2016. The Fort Collins Commuter Study: impact of route type and transport mode on personal exposure to multiple air pollutants. J. Expo. Sci. Environ. Epidemiol 26 (4), 397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Gor P, Martinez A, Reyna-Caamaño R, Mejia-Velazquez GM, Guerrero-Medrano L, Santos-Guzman J, 2014. Environmental health indicators for the border region between South Texas and Northeastern Mexico. Environmental Sustainability Issues in the South Texas–Mexico Border Region. Springer, pp. 25–42. [Google Scholar]
  30. Hannam K, McNamee R, De Vocht F, Baker P, Sibley C, Agius R, 2013. A comparison of population air pollution exposure estimation techniques with personal exposure estimates in a pregnant cohort. Environ. Sci.: Processes Impacts 15 (8), 1562–1572. [DOI] [PubMed] [Google Scholar]
  31. Hansen C, Neller A, Williams G, Simpson R, 2006. Maternal exposure to low levels of ambient air pollution and preterm birth in Brisbane, Australia. BJOG Int. J. Obstet. Gynaecol 113 (8), 935–941. [DOI] [PubMed] [Google Scholar]
  32. van den Hooven EH, de Kluizenaar Y, Pierik FH, Hofman A, van Ratingen SW, Zandveld PY, et al. , 2011. Air pollution, blood pressure, and the risk of hypertensive complications during pregnancy the generation R study. Hypertension 57 (3), 406–412. [DOI] [PubMed] [Google Scholar]
  33. Hyder A, Lee HJ, Ebisu K, Koutrakis P, Belanger K, Bell ML, 2014. PM2.5 exposure and birth outcomes: use of satellite-and monitor-based data. Epidemiology 25 (1), 58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Jedrychowski WA, Perera FP, Maugeri U, Mroz E, Klimaszewska-Rembiasz M, Flak E, et al. , 2010. Effect of prenatal exposure to fine particulate matter on ventilatory lung function of preschool children of non-smoking mothers. Paediatr. Perinat. Epidemiol 24 (5), 492–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Johnson NM, Carrillo G, Mehta RK, 2016. Academic and psychosocial impact of air pollution on children. Pediatric Neurotoxicology. Springer, pp. 167–180. [Google Scholar]
  36. Kettunen J, Lanki T, Tiittanen P, Aalto PP, Koskentalo T, Kulmala M, et al. , 2007. Associations of fine and ultrafine particulate air pollution with stroke mortality in an area of low air pollution levels. Stroke 38 (3), 918–922. [DOI] [PubMed] [Google Scholar]
  37. Kim S, Wipfli H, Navas-Acien A, Dominici F, Avila-Tang E, Onicescu G, et al. , 2009. Determinants of hair nicotine concentrations in nonsmoking women and children: a multicountry study of secondhand smoke exposure in homes. Cancer Epidemiol. Biomark. Prev 18 (12), 3407–3414. [DOI] [PubMed] [Google Scholar]
  38. Kim SR, Wipfli H, Avila-Tang E, Samet JM, Breysse PN, 2009. Method validation for measurement of hair nicotine level in nonsmokers. Biomed. Chromatogr 23 (3), 273–279. [DOI] [PubMed] [Google Scholar]
  39. Klepeis NE, Tsang AM, Behar JV, 1996. Analysis of the National Human Activity Pattern Survey (NHAPS) Respondents From a Standpoint of Exposure Assessment. US EPA, Las Vegas, NV. [Google Scholar]
  40. Klepeis NE, Nelson WC, Ott WR, Robinson JP, Tsang AM, Switzer P, et al. , 2001a. The national human activity pattern survey (NHAPS): a resource for assessing exposure to environmental pollutants. J. Expo. Anal. Environ. Epidemiol 11, 252–321. [DOI] [PubMed] [Google Scholar]
  41. Klepeis NE, Nelson WC, Ott WR, Robinson JP, Tsang AM, Switzer P, et al. , 2001b. The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants. J. Expo. Anal. Environ. Epidemiol 11 (3), 231–252. [DOI] [PubMed] [Google Scholar]
  42. LaRosa LE, Buckley TJ, Wallace LA, 2002. Real-time indoor and outdoor measurements of black carbon in an occupied house: an examination of sources. J. Air Waste Manage. Assoc 52 (1), 41–49. [DOI] [PubMed] [Google Scholar]
  43. Leon Hsu H-H, Mathilda Chiu Y-H, Coull BA, Kloog I, Schwartz J, Lee A, et al. , 2015. Prenatal particulate air pollution and asthma onset in urban children. Identifying sensitive windows and sex differences. Am. J. Respir. Crit. Care Med 192 (9), 1052–1059. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Levy ME, Zhang R, Zheng J, Tan H, Wang Y, Molina LT, et al. , 2014. Measurements of submicron aerosols at the California–Mexico border during the Cal–Mex 2010 field campaign. Atmos. Environ 88, 308–319. [Google Scholar]
  45. Lim S, Kim J, Kim T, Lee K, Yang W, Jun S, et al. , 2012. Personal exposures to PM2.5 and their relationships with microenvironmental concentrations. Atmos. Environ 47, 407–412. [Google Scholar]
  46. Miller KA, Siscovick DS, Sheppard L, Shepherd K, Sullivan JH, Anderson GL, et al. , 2007. Long-term exposure to air pollution and incidence of cardiovascular events in women. N. Engl. J. Med 356 (5), 447–458. [DOI] [PubMed] [Google Scholar]
  47. Mortimer K, Neugebauer R, Lurmann F, Alcorn S, Balmes J, Tager I, 2008. Air pollution and pulmonary function in asthmatic children: effects of prenatal and lifetime exposures. Epidemiology 19 (4), 550–557. [DOI] [PubMed] [Google Scholar]
  48. Nemmar A, Vanbilloen H, Hoylaerts M, Hoet P, Verbruggen A, Nemery B, 2001. Passage of intratracheally instilled ultrafine particles from the lung into the systemic circulation in hamster. Am. J. Respir. Crit. Care Med 164 (9), 1665–1668. [DOI] [PubMed] [Google Scholar]
  49. Nerriere É, Zmirou-Navier D, Blanchard O, Momas I, Ladner J, Le Moullec Y, et al. , 2005. Can we use fixed ambient air monitors to estimate population long-term exposure to air pollutants? The case of spatial variability in the Genotox ER study. Environ. Res 97 (1), 32–42. [DOI] [PubMed] [Google Scholar]
  50. Nethery E, Leckie SE, Teschke K, Brauer M, 2008. From measures to models: an evaluation of air pollution exposure assessment for epidemiological studies of pregnant women. Occup. Environ. Med 65 (9), 579–586. [DOI] [PubMed] [Google Scholar]
  51. Oglesby L, Künzli N, Röösli M, Braun-Fahrländer C, Mathys P, Stern W, et al. , 2000. Validity of ambient levels of fine particles as surrogate for personal exposure to outdoor air pollution—results of the European EXPOLIS-EAS Study (Swiss Center Basel). J. Air Waste Manage. Assoc 50 (7), 1251–1261. [DOI] [PubMed] [Google Scholar]
  52. Perera FP, Rauh V, Tsai W-Y, Kinney P, Camann D, Barr D, et al. , 2003. Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population. Environ. Health Perspect 111 (2), 201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Pokhrel AK, Bates MN, Acharya J, Valentiner-Branth P, Chandyo RK, Shrestha PS, et al. , 2015. PM2.5 in household kitchens of Bhaktapur, Nepal, using four different cooking fuels. Atmos. Environ 113, 159–168. [Google Scholar]
  54. Pope CI, Dockery D, 2006. 2006 critical review-health effects of fine particulate air pollution: lines that connect. J. Air Waste Manage 56, 709–742. [DOI] [PubMed] [Google Scholar]
  55. Quintana PJ, Ganster P, Stigler Granados PE, Muñoz-Meléndez G, Quintero-Núñez M, Rodríguez-Ventura JG, 2015. Risky borders: traffic pollution and health effects at US–Mexican ports of entry. J. Borderl. Stud 30 (3), 287–307. [Google Scholar]
  56. Rodes CE, Lawless PA, Thornburg JW, Williams RW, Croghan CW, 2010. DEARS particulate matter relationships for personal, indoor, outdoor, and central site settings for a general population. Atmos. Environ 44 (11), 1386–1399. [Google Scholar]
  57. Services TDoSH, 2013a. Asthma Burden Among Children in Hidalgo County, Public Health Region 11, and Texas, 2005–2011 [Web Report]. https://www.dshs.texas.gov/asthma/data.shtm (July 25, 2016).
  58. Services TDoSH, 2013b. Asthma Burden Among Children in Hidalgo County, Public Health Region 11, and Texas. 2005–2011 Available from:. http://www.dshs.texas.gov/asthma/data/shtm (12 July 2016).
  59. Shi L, Zanobetti A, Kloog I, Coull BA, Koutrakis P, Melly SJ, et al. , 2016. Low-concentration PM2.5 and mortality: estimating acute and chronic effects in a population-based study. Environ. Health Perspect 124 (1), 46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Soneja S, Chen C, Tielsch JM, Katz J, Zeger SL, Checkley W, et al. , 2014. Humidity and gravimetric equivalency adjustments for nephelometer-based particulate matter measurements of emissions from solid biomass fuel use in cookstoves. Int. J. Environ. Res. Public Health 11 (6), 6400–6416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Statistics THDCfH, 2016. Texas Health Data Center for Health Statistics of Live Birth Outcome 2005–2013. http://healthdata.dshs.texas.gov/VitalStatistics/Birth: Center For Health Statistics Department of State Health Services. [Google Scholar]
  62. Takahama S, Russell L, Shores C, Marr L, Zheng J, Levy M, et al. , 2014. Diesel vehicle and urban burning contributions to black carbon concentrations and size distributions in Tijuana, Mexico, during the Cal-Mex 2010 campaign. Atmos. Environ 88, 341–352. [Google Scholar]
  63. TCEQ, 2015. Texas Five-year Ambient Monitoring Network Assessment. https://www3.epa.gov/ttn/amtic/files/networkplans/TXAssess2015.pdf.
  64. TCEQ, 2016. Continuous Ambient Monitoring Station Mission C43/AP143 [Internet]. TCEQ Harlingen Regional Office, UT Brownsville: Available from:. http://www.tceq.state.tx.us/cgi-bin/compliance/monops/site_photo.pl?cams=43 (cited June 6, 2016). [Google Scholar]
  65. Team RC, 2016. R: A Language and Environment for Statistical Computing. Available from: https://www.R-project.org.
  66. TranSafety I, 1997. Study compares older and younger pedestrian walking speeds. Road Manage. Eng. J Available from: http://www.usroads.com/journals/p/rej/9710/re971001.htm. [Google Scholar]
  67. Wallace L, 2000. Correlations of personal exposure to particles with outdoor air measurements: a review of recent studies. Aerosol Sci. Technol 32 (1), 15–25. [Google Scholar]
  68. Wan M-P, Wu C-L, To G-NS, Chan T-C, Chao CY, 2011. Ultrafine particles, and PM2.5 generated from cooking in homes. Atmos. Environ 45 (34), 6141–6148. [Google Scholar]
  69. Weisel CP, Zhang J, Turpin BJ, Morandi MT, Colome S, Stock TH, et al. , 2005. Relationships of Indoor, Outdoor, and Personal Air (RIOPA). Part I. Collection methods and descriptive analyses. Res. Rep. Health Eff. Inst (130 Pt 1), 1–107 (discussion 9–27). [PubMed] [Google Scholar]
  70. Wilhelm M, Ghosh JK, Su J, Cockburn M, Jerrett M, Ritz B, 2012. Traffic-related air toxics and term low birth weight in Los Angeles County, California. Environ. Health Perspect 120 (1), 132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. World Health Organization, 2013. Review of evidence on health aspects of air pollution–REVIHAAP Project. World Health Organization, Copenhagen, Denmark. [Google Scholar]
  72. Zuniga GC, Hernandez T, Kirk S, Nadeau N, Chong-Menard B, Lucio RL, et al. , 2011. A multi-institutional collaboration to develop asthma education for school settings in South Texas. Public Health Rep 126 (1), 139–144. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Sup mat
NIHMS2157114-supplement-Sup_mat.doc (234KB, doc)

RESOURCES