Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Dec 15.
Published in final edited form as: Environ Pollut. 2022 Oct 10;315:120401. doi: 10.1016/j.envpol.2022.120401

Fossil-fuel and combustion-related air pollution and hypertension in the Sister Study

Jing Xu a,b, Nicole M Niehoff a, Alexandra J White a, Emily J Werder a, Dale P Sandler a,*
PMCID: PMC9746069  NIHMSID: NIHMS1843757  PMID: 36228848

Abstract

Hypertension is a leading risk factor for disease burden, with more than 200 million disability-adjusted life-years attributed to high blood pressure in 2015. While outdoor air pollution is associated with cardiovascular disease, the joint effect of exposure to air pollution from combustion products on hypertension has rarely been studied. We conducted a cross-sectional analysis to explore the association between combustion-related air pollution and hypertension. Census-tract levels of ambient concentrations of nine fossil-fuel and combustion-related air toxics (biphenyl, naphthalene, polycyclic organic matter, diesel emissions, 1,3-butadiene, acetaldehyde, benzene, acrolein, and formaldehyde) from the 2005 National Air Toxics Assessment database and NO2 from 2005 monitoring data were linked to baseline residential addresses of 47,467 women in the Sister Study cohort. Hypertension at enrollment (2003-2009) was defined as high systolic (≥140 mm Hg) or diastolic (≥90 mm Hg) blood pressure or taking antihypertensive medication. We used log-binomial regression and quantile-based g-computation to estimate the individual and joint effects of fossil-fuel and combustion-related air pollution on hypertension. Comparing the highest to lowest quartiles, diesel emissions (prevalence ratio (PR) = 1.05, 95% confidence interval (CI)=1.01,1.08), 1,3-butadiene (PR=1.04, 95%CI=1.00,1.07), acetaldehyde (PR=1.08, 95%CI=1.04,1.12), benzene (PR=1.05, 95%CI=1.02,1.08), formaldehyde (PR=1.08, 95%CI=1.04,1.11), and NO2 (PR=1.08, 95%CI=1.05,1.12) were individually associated with higher prevalence of hypertension. The PR for the joint effect of increasing all ambient air toxics and NO2 by one quartile was 1.02 (95%CI=1.01,1.04). Associations varied by race/ethnicity, with stronger associations observed among women reporting races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) other than non-Hispanic White. In conclusion, we found that air pollution from fossil fuel and combustion may be a risk factor for hypertension.

Keywords: Fossil fuel, combustion, air toxics, hypertension

Graphical Abstract

graphic file with name nihms-1843757-f0001.jpg

Introduction

Hypertension is an increasingly important global health challenge. In 2000, there were 972 million adults with hypertension worldwide (Kearney et al., 2005), and the number increased to 1.39 billion in 2010 (Mills et al., 2016). In 2019, there were 10.8 million deaths attributed to high systolic blood pressure (GBD 2019 Risk Factors Collaborators, 2020). Besides lifestyle risk factors, environmental factors, such as air pollution, may influence hypertension risk. More exposure to traffic may increase people’s risk of hypertension (Avila-Palencia et al., 2022). The combustion of fossil fuels releases a complex mixture of pollutants and toxics, which contribute to air pollution levels (Cosselman et al., 2015). Multiple epidemiological studies have demonstrated that increased levels of air pollution, when assessed using measures of particulate matter and nitrogen oxides, are associated with higher blood pressure (Chan et al., 2015; Yang et al., 2018). Possible mechanisms include pollutant-related oxidative stress and systemic inflammatory response (Donaldson et al., 2001).

Combustion by-products may play a role in the relationship between air pollution and hypertension. For example, diesel particulate matter from combustion of diesel fuels has been associated with blood pressure in some (Cosselman et al., 2012), but not other (Bangia et al., 2015) epidemiological studies. Polycyclic aromatic hydrocarbons (PAHs) are a group of chemicals produced by incomplete combustion of organic materials (e.g., coals and fossil fuels) (Holme et al., 2019). Multiple epidemiological studies have found that elevated levels of PAHs are associated with higher blood pressure (Bangia et al., 2015; Holme et al., 2019). Nitrogen dioxide (NO2) is considered as an air pollutant from fuel-burning, and some epidemiological studies have shown that NO2 is associated with higher blood pressure and risk of hypertension (Chan et al., 2015; Yang et al., 2018). Volatile organic compounds (VOCs) are a group of chemicals from anthropogenic sources such as gas stations and vehicle emissions exhaust (Delikhoon et al., 2018). Several epidemiological studies have focused on the association between VOCs and hypertension or high blood pressure, but the results are inconsistent (Männistö et al., 2015; Shin et al., 2015; Zhu et al., 2017).

In this study, we explored the association between fossil-fuel and combustion-related air pollution and hypertension. We first evaluated the association between individual combustion-related air pollutant and hypertension. Since ambient PAHs, VOCs, and NO2 are mainly from combustion, assessing their joint effect can more comprehensively evaluate the association between combustion-related air pollution and hypertension. To address the correlated nature of these air pollutants and more realistically model real-world exposures, we also evaluated the effect of the fossil-fuel and combustion-related pollutants mixture on hypertension. The prevalence of hypertension is different across racial/ethnic groups (Bennett et al., 2016) and regions of the U.S. (https://www.cdc.gov/bloodpressure/maps_data.htm), which may indicate the possibility of differential prevalence or type of exposures or other risk factors (Kershaw et al., 2013). We therefore considered whether associations varied by race/ethnicity or geographic region.

Methods

Study population

Participants included in this analysis are from the Sister Study, a prospective cohort study of the effects of environmental exposures on chronic diseases among women aged 35-74 dwelling in the United States (Sandler et al., 2017). A total of 50,884 breast cancer-free women whose sister had been diagnosed with breast cancer were enrolled from 2003-2009 (Sandler et al., 2017). The median age at enrollment was 55.6 years old, and 83.7% of the women are non-Hispanic White. Participants have relatively high socioeconomic status, with 74.2% of participants having household income more than $50,000 and 51% having a bachelor’s degree or higher (Sandler et al., 2017).

Information on demographic characteristics, residential address, socioeconomic status, personal anti-hypertensive medication use, and smoking status was collected through a computer-assisted telephone interview (Sandler et al., 2017). During a home visit, examiners measured participant’s height and weight as well as participants’ resting blood pressure following a standardized protocol (Sandler et al., 2017). In this cross-sectional study, we included 47,467 participants residing in the US after excluding those without complete data on included covariates (missing n=2997), blood pressure measurements or anti-hypertensive medication use (missing n=1005), and concentrations of pollutants at the enrollment residence (missing n=449) (NO2 data were only obtained for women living in continental US). This database used in this study was from Sister Study Data Release 7.2. The institutional review board of the National Institutes of Health oversees the Sister Study. All participants provided written consent.

Exposure assessment

Annual average NO2 concentration at each participant’s enrollment residence at the census-block level was derived based on monitoring data from 2005 using a satellite-based land-use regression model (Chan et al., 2015; Novotny et al., 2011). Briefly, atmospheric NO2 surface concentrations were predicted based on multivariate linear regression considering land-use characteristics and NO2 column measurements from the ozone monitoring instrument satellite sensor.

Other exposure data were from the Environmental Protection Agency (EPA)’s National Air Toxics Assessment (NATA) database, which comprehensively evaluates total annual average of ambient toxics concentrations for each census tract in the United States (Environmental Protection Agency, 2011). The National Emission Inventory (NEI) was first compiled based on releases from point and non-point sources, on-road mobile and non-road mobile sources, background sources, and secondary formation of toxics. EPA used data from NEI as input in air quality models to estimate ambient concentrations of air toxics (Environmental Protection Agency, 2011). Census tract number was assigned to each participant by their geocoded baseline residential address, and 2005 NATA data at the census-tract level was linked to each participant’s census tract number to represent their air pollution exposure. We selected the 2005 NATA database because 2005 falls within the baseline enrollment period (2003-2009) and a more refined dispersion model was used to estimate ambient concentrations for 2005 NATA compared to previous-released versions (Environmental Protection Agency, 2011).

Since this study focused on fossil-fuel and combustion-related air pollution, we selected polycyclic organic matter (PAHPOM) from the NATA database which represents the total concentration of 16 PAHs; we also included biphenyl and naphthalene as individual PAH. For VOCs, the combustion of fossil fuels, such as petroleum and gasoline, emits benzene and other hydrocarbons such as 1,3-butadiene, formaldehyde, acetaldehyde, and acrolein, and limits for these chemicals’ content in gasoline are included in the gasoline standard released by EPA (https://www.epa.gov/gasoline-standards/learn-about-gasoline). We also included diesel emissions levels to represent emissions from combustion via diesel engines.

Outcome assessment

Blood pressure was measured at the enrollment home visit based on procedures used in the US National Health and Nutrition Examination Survey. Sitting blood pressure was measured 3 times after a short rest using an alternate arm protocol (L-R-L). If three blood pressure measurements were obtained, we used the average of the last two measurements. If there were two or one blood pressure measurement, the last or single measurement was used (<2%). Hypertension was defined as: systolic blood pressure (SBP) ≥140 mm Hg or diastolic blood pressure (DBP) ≥90 mm Hg or taking anti-hypertensive medication. When considering continuous blood pressure as the outcome, we corrected the blood pressure measurements by adding 15 mm Hg to the measured SBP and 10 mm Hg to the measured DBP for those who reported taking antihypertensive medication. Correcting blood pressure measurements by adding constants to the measured blood pressure can reduce the bias caused by excluding participants taking medication (Tobin et al., 2005), and we added 10 mm Hg to DBP and 15 mm Hg to SBP based on a published meta-analysis focusing on the effect of antihypertensive medication (Baguet et al., 2007). We also did sensitivity analyses excluding participants who were taking anti-hypertensive medication and using the original blood pressure measurements without correction.

Covariates

We used a directed acyclic graph to pick covariates adjusted in the model (figure S1). Body mass index (BMI) was calculated as a person’s weight in kilograms divided by the square of the person’s height in meters, and obesity was defined as BMI≥30 (World Health Organization). We adjusted for race/ethnicity (non-Hispanic White, other [Hispanic/Latina, non-Hispanic Black and other]), individual educational attainment (high school or less, some college, Bachelor’s degree or more), individual annual household income (≥$49,999, $50,000-$99,999, ≥$100,000), census-tract education level (high school or less, some college, Bachelor’s degree or more), census-tract income level (≤$49,999, $50,000-$99,999, ≥$100,000), and geographic region (Northeast, Midwest, South, West). Census-tract education and median income level were classified based on 2000 U.S. Census data. The most frequent education level of residents in each census-tract was used to characterize census-tract level education.

Statistical analysis

The exposure variables were categorized in quartiles, and the distribution is shown in table S1. We used Spearman rank correlation to calculate the correlation coefficients among air toxics and NO2. To investigate the association between fossil-fuel and combustion-related air toxics and NO2 and hypertension, we used log binomial regression considering hypertension as a binary outcome in relation to quartiles of air toxics/NO2 (first quartile as referent). We reported prevalence ratios (PRs) and 95% confidence intervals (CIs) for overall risk. We also estimated risks stratified by race/ethnicity and geographic region, and interaction terms were introduced to the models to investigate potential effect modification. We used the median of each quartile category for trend tests. When considering continuous blood pressure as the outcome, we used general linear models.

To explore the association between the mixture of fossil-fuel and combustion-related air toxics and NO2 and hypertension, we used a quantile-based g-computation approach to estimate the mixture PR, 95% CI when all exposure variables increase one quartile category simultaneously. We also evaluated the association between the mixture and hypertension in racial/ethnical subgroups.

We conducted the following stratification and sensitivity analyses. (1) Using a newer definition of hypertension (SBP≥130 mm Hg or DBP≥80 mm Hg) released by the American College of Cardiology and the American Heart Association (Whelton et al., 2018) well after study enrollment. (2) Since smoking is a major source of PAHs and VOCs but the relationship between smoking status and the development of hypertension is still unclear (even though smoking is related to acute increases in blood pressure) (Appel, 2022), we carried out an analysis additionally adjusting for smoking status and an analysis stratified by smoking status. (3) BMI might be a modifier or a mediator of the association between air pollution and hypertension, so we also stratified by BMI (BMI<30 and BMI≥30) but did not include BMI as a confounder. (4) Physical activity is a protective factor for hypertension. However, the causality of physical activity and air pollution is still unclear (Tainio et al., 2021). Physical activity might be a mediator or a modifier of the association between air pollution and hypertension. Thus, we carried out a stratified analysis but did not adjust for physical activity as a confounder in our main models. We stratified by total metabolic equivalent task (MET)-hours for exercise/sports per week (<10 and ≥10). MET-hours were calculated using self-reported information on frequency and intensity of exercise and sport activities.

Statistical significance was defined as p < 0.05 (two-tailed). Analysis was done using SAS 9.4 software (SAS Institution, Inc., Cary, NC). Quantile-based g-computation was done using R 3.6.1 package “qgcomp” (Keil et al., 2020).

Results

The characteristics of participants included in this analysis are shown in table 1. Among 47,467 participants, 15,905 (34%) had hypertension. Compared with non-hypertensive participants, women with hypertension were on average older, were more likely to self-report race/ethnicity other than non-Hispanic White (Hispanic/Latina, non-Hispanic Black, and other), had lower individual educational attainment and annual household income, and lived in areas with lower census-tract income and education level.

Table 1.

Characteristics of hypertensive and non-hypertensive participants in the Sister Study (N [%])

Characteristic Full population (N=47,467) Hypertensive (N=15,905) Non-hypertensive (N=31,562)
Age (years)
  ≤45 7487 (16) 900 (6) 6587 (21)
  46-50 7815 (16) 1619 (10) 6196 (20)
  51-55 9571 (20) 2880 (18) 6691 (21)
  56-60 9056 (19) 3448 (22) 5608 (18)
  61-65 6933 (15) 3292 (21) 3641 (12)
  >65 6605 (14) 3766 (24) 2839 (9)
Race/ethnicity
  Non-Hispanic White 40358 (85) 12831 (81) 27527 (87)
  Other (Hispanic/Latina, non-Hispanic Black, and other) 7109 (15) 3074 (19) 4035 (13)
Educational attainment
  High school or less 7122 (15) 3000 (19) 4122 (13)
  Some college 16041 (34) 5963 (37) 10078 (32)
  Bachelor’s degree or more 24304 (51) 6942 (44) 17362 (55)
Annual household income
  ≤$49,999 11726 (25) 5167 (32) 6559 (21)
  $50,000-$99,999 19535 (41) 6587 (41) 12948 (41)
  ≥$100,000 16206 (34) 4151 (26) 12055 (38)
Region
  Northeast 8042 (17) 2433 (15) 5609 (18)
  Midwest 13002 (27) 4455 (28) 8547 (27)
  South 15907 (34) 5743 (36) 10164 (32)
  West 10516 (22) 3274 (21) 7242 (23)
Census-tract income
  ≤$49,999 15593 (33) 6104 (38) 9489 (30)
  $50,000-$99,999 27930 (59) 8864 (56) 19066 (60)
  ≥$100,000 3944 (8) 937 (6) 3007 (10)
Census-tract education
  High school or less 25596 (54) 9553 (60) 16043 (51)
  Some college 6772 (14) 2238 (14) 4534 (14)
  Bachelor’s degree or more 15099 (32) 4114 (26) 10985 (35)
Body mass index (BMI)
  <30 33134 (70) 8344 (52) 24790 (79)
  ≥30 14317 (30) 7554 (48) 6763 (21)
  Missing 16 7 9
Smoking status
  Never smoked 26483 (56) 8394 (53) 18089 (57)
  Past smoker 17024 (36) 6193 (39) 10831 (34)
  Current smoker 3957 (8) 1316 (8) 2641 (8)
  Missing 3 2 1
MET-hoursa
  <10 25302 (53) 9672 (61) 15630 (50)
  ≥10 22135 (47) 6226 (39) 15909 (50)
  Missing 30 7 23
Open in a new tab
a

Metabolic equivalent task (MET)-hours for exercise/sports per week

The median census-tract/census block concentrations based on each participant’s enrollment address were 0.48μg/m3 for diesel emissions, 1.76μg/m3 for acetaldehyde, 0.88μg/m3 for benzene, 1.88μg/m3 for formaldehyde, 10.1μg/m3 for NO2, 7.38×10−05μg/m3 for biphenyl, 4.12×10−02μg/m3 for naphthalene, 8.69×10−03 μg/m3 for PAHPOM, 5.61×10−02 μg/m3 for 1,3-butadiene, and 3.45×10−02 μg/m3 for acrolein (table S1). The correlations among fossil-fuel and combustion-related air toxics and NO2 were positive and mostly moderate or strong (correlation coefficients 0.2-0.9) (table S2).

As shown in table 2, comparing the fourth quartile of exposure with the first, we observed a modestly higher prevalence of hypertension for diesel emissions (PR=1.05, 95%CI=1.01,1.08; ptrend=0.02), 1,3-butadiene (PR=1.04, 95%CI=1.00,1.07; p trend=0.02), acetaldehyde (PR=1.08, 95%CI=1.04,1.12; p trend<0.001), benzene (PR=1.05, 95%CI=1.02,1.08; p trend=0.002), formaldehyde (PR=1.08, 95%CI=1.04,1.11; p trend<0.001), and NO2 (PR=1.08, 95%CI=1.05,1.12; p trend><0.001), with increasing trends; we did not find associations for biphenyl (PR=1.02, 95%CI=0.99,1.06; p trend=0.3) or PAHPOM (PR=1.00, 95%CI=0.97,1.04; p trend=0.7). Comparing quartile 3 with quartile 1, we also observed a higher prevalence of hypertension for naphthalene (PR=1.05, 95%CI=1.02,1.09; p trend=1.0) and acrolein (PR =1.04, 95%CI=1.01,1.07; p trend=0.9), but the exposure-related trends were non-monotonic.

Table 2.

Fossil-fuel and combustion-related air pollution and overall risk of hypertension in the Sister Study (N=47,467)

N PR (95%CI)
Biphenyl Quartile 1 11867 1 (referent)
Quartile 2 11867 1.02 (0.99, 1.05)
Quartile 3 11867 1.00 (0.97, 1.04)
Quartile 4 11866 1.02 (0.99, 1.06)
p for trend 0.3
Naphthalene Quartile 1 11865 1 (referent)
Quartile 2 11869 1.03 (1.00, 1.06)
Quartile 3 11868 1.05 (1.02, 1.09)
Quartile 4 11865 1.01 (0.98, 1.04)
p for trend 1.0
PAHPOM Quartile 1 11868 1 (referent)
Quartile 2 11866 0.99 (0.96, 1.02)
Quartile 3 11867 1.01 (0.98, 1.04)
Quartile 4 11866 1.00 (0.97, 1.04)
p for trend 0.7
Diesel emissions Quartile 1 11865 1 (referent)
Quartile 2 11869 1.03 (0.99, 1.06)
Quartile 3 11867 1.03 (1.00, 1.07)
Quartile 4 11866 1.05 (1.01, 1.08)
p for trend 0.02
1,3-Butadiene Quartile 1 11863 1 (referent)
Quartile 2 11870 1.01 (0.98, 1.05)
Quartile 3 11867 1.05 (1.02, 1.09)
Quartile 4 11867 1.04 (1.00, 1.07)
p for trend 0.02
Acetaldehyde Quartile 1 11865 1 (referent)
Quartile 2 11868 1.03 (0.99, 1.06)
Quartile 3 11867 1.05 (1.02, 1.09)
Quartile 4 11867 1.08 (1.04, 1.12)
p for trend <0.001
Benzene Quartile 1 11868 1 (referent)
Quartile 2 11865 1.02 (0.99, 1.05)
Quartile 3 11868 1.04 (1.01, 1.07)
Quartile 4 11866 1.05 (1.02, 1.08)
p for trend 0.002
Acrolein Quartile 1 11867 1 (referent)
Quartile 2 11867 1.02 (0.99, 1.06)
Quartile 3 11866 1.04 (1.01, 1.07)
Quartile 4 11867 1.00 (0.97, 1.04)
p for trend 0.9
Formaldehyde Quartile 1 11867 1 (referent)
Quartile 2 11866 1.04 (1.00, 1.07)
Quartile 3 11866 1.05 (1.01, 1.08)
Quartile 4 11868 1.08 (1.04, 1.11)
p for trend <0.001
NO2a Quartile 1 11798 1 (referent)
Quartile 2 11791 1.06 (1.02, 1.09)
Quartile 3 11812 1.10 (1.07, 1.14)
Quartile 4 11771 1.08 (1.05, 1.12)
p for trend <0.001
Open in a new tab

Models were adjusted for age, race/ethnicity, individual income and education, census-tract income and education, and region

a

Ambient NO2 concentrations were only obtained for participants living in continental United States (missing N=295).

After stratifying by race/ethnicity, the association of prevalence of hypertension with biphenyl (p interaction<0.001), diesel emissions (p interaction=0.02), 1,3-butadiene (p interaction=0.06), acetaldehyde (p interaction<0.001), acrolein (p interaction=0.004), and formaldehyde (p interaction<0.001) differed across races/ethnicities (table 3). Comparing the fourth quartile of exposure with the first, among participants reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) we observed increasing trends for biphenyl (PR =1.07, 95%CI=1.00,1.15; p trend=0.02), diesel emissions (PR=1.12, 95%CI=1.04,1.21; ptrend=0.003), l,3-butadiene(PR =1.09, 95%CI=1.01,1.17; p trend=0.003), acetaldehyde(PR =1.28, 95%CI=1.16,1.42; p trend<0.001), acrolein (PR =1.13, 95%CI=1.04,1.22; p trend=0.01), and formaldehyde(PR =1.26, 95%CI=1.14,1.39; p trend<0.001); these trends were not seen for non-Hispanic White participants (table 3). As shown in table S3, the median concentrations of all air toxics and NO2 were higher for participants self-reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) compared with non-Hispanic White participants. When we further investigated the associations among Black women, some of the associations were attenuated compared to the combined group of participants self-reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) but most of the directions of association did not change (table S4).

Table 3.

Fossil-fuel and combustion-related air pollution and risk of hypertension stratified by race/ethnicity in the Sister Study

Non-Hispanic White (N=40,358)
 Other (Hispanic/Latina, non-Hispanic Black and other) (N=7,109)
 p for interaction
N PR (95%CI) N PR (95%CI)
Biphenyl Quartile 1 10649 1 (referent) 1218 1 (referent)
Quartile 2 10055 1.03 (1.00, 1.07) 1812 0.97 (0.90, 1.05)
Quartile 3 9963 0.97 (0.94, 1.01) 1904 1.07 (1.00, 1.16)
Quartile 4 9691 1.00 (0.96, 1.04) 2175 1.07 (1.00, 1.15)
p for trend 0.7 0.02 <0.001
Naphthalene Quartile 1 10368 1 (referent) 1497 1 (referent)
Quartile 2 10281 1.02 (0.98, 1.06) 1588 1.04 (0.97, 1.12)
Quartile 3 10213 1.03 (0.99, 1.07) 1655 1.11 (1.03, 1.19)
Quartile 4 9496 0.98 (0.94, 1.03) 2369 1.09 (1.01, 1.17)
p for trend 0.3 0.04   0.3
PAHPOM Quartile 1 10390 1 (referent) 1478 1 (referent)
Quartile 2 9971 0.99 (0.96, 1.03) 1895 0.97 (0.90, 1.04)
Quartile 3 9797 0.99 (0.96, 1.03) 2070 1.05 (0.98, 1.12)
Quartile 4 10200 0.99 (0.95, 1.03) 1666 1.06 (0.98, 1.14)
p for trend 0.8 0.05   0.2
Diesel emissions Quartile 1 10779 1 (referent) 1086 1 (referent)
Quartile 2 10534 1.02 (0.99, 1.06) 1335 1.06 (0.97, 1.15)
Quartile 3 9887 1.01 (0.97, 1.05) 1980 1.10 (1.02, 1.19)
Quartile 4 9158 1.01 (0.98, 1.06) 2708 1.12 (1.04, 1.21)
p for trend 0.7 0.003   0.02
1,3-Butadiene Quartile 1 10698 1 (referent) 1165 1 (referent)
Quartile 2 10419 1.01 (0.97, 1.05) 1451 0.98 (0.91, 1.07)
Quartile 3 9917 1.02 (0.99, 1.06) 1950 1.12 (1.04, 1.20)
Quartile 4 9324 1.01 (0.97, 1.06) 2543 1.09 (1.01, 1.17)
p for trend 0.5 0.003   0.06
Acetaldehyde Quartile 1 11074 1 (referent) 791 1 (referent)
Quartile 2 10510 1.02 (0.98, 1.06) 1358 1.16 (1.04, 1.29)
Quartile 3 9740 1.02 (0.99, 1.06) 2127 1.26 (1.14, 1.39)
Quartile 4 9034 1.04 (1.00, 1.08) 2833 1.28 (1.16, 1.42)
p for trend 0.05 <0.001 <0.001
Benzene Quartile 1 10702 1 (referent) 1166 1 (referent)
Quartile 2 10301 1.01 (0.97, 1.05) 1564 1.03 (0.95, 1.12)
Quartile 3 9838 1.02 (0.98, 1.06) 2030 1.09 (1.01, 1.17)
Quartile 4 9517 1.02 (0.98, 1.06) 2349 1.11 (1.03, 1.19)
p for trend 0.3 0.002   0.08
Acrolein Quartile 1 10861 1 (referent) 1006 1 (referent)
Quartile 2 10295 1.00 (0.97, 1.04) 1572 1.10 (1.01, 1.19)
Quartile 3 9742 1.01 (0.97, 1.05) 2124 1.14 (1.06, 1.24)
Quartile 4 9460 0.97 (0.93, 1.01) 2407 1.13 (1.04, 1.22)
p for trend 0.09 0.01   0.004
Formaldehyde Quartile 1 11138 1 (referent) 729 1 (referent)
Quartile 2 10589 1.03 (0.99, 1.07) 1277 1.16 (1.04, 1.29)
Quartile 3 9722 1.01 (0.97, 1.05) 2144 1.25 (1.13, 1.38)
Quartile 4 8909 1.04 (1.00, 1.08) 2959 1.26 (1.14, 1.39)
p for trend 0.08 <0.001 <0.001
NO2a Quartile 1 10573 1 (referent) 1225 1 (referent)
Quartile 2 10322 1.07 (1.03, 1.11) 1469 1.01 (0.93, 1.09)
Quartile 3 9956 1.10 (1.06, 1.14) 1856 1.08 (1.01, 1.17)
Quartile 4 9290 1.05 (1.01, 1.10) 2481 1.15 (1.06, 1.23)
p for trend 0.01 <0.001   0.09
Open in a new tab

Models were adjusted for age, individual income and education, census-tract income and education, and region

a

Ambient NO2 concentrations were only obtained for participants living in continental United States (missing N=295).

In an analysis stratified by region, we found interaction between geographic region and diesel emissions (p interaction=0.02), acetaldehyde (p interaction<0.001), and formaldehyde (p interaction<0.001) (table S5). For diesel emissions, there was an increasing trend with a significant higher PR in the fourth quartile in the West (PR Q 4 vs 1=1.17, 95%CI=1.09,1.25; p trend<0.001), but not in other regions. The concentration of diesel emissions was relatively higher in the West (median=0.55, Q1=0.19, Q3=1.29) compared with other regions (table S6). We observed elevated PRs for hypertension in the South associated with acetaldehyde (PR Q4 vs 1=1.21, 95%CI=1.12,1.30; p trend<0.001) and formaldehyde (PRQ4 vs 1=1.15, 95%CI=1.07,1.25; p trend<0.001) but not in other regions (table S5), and the concentrations of acetaldehyde (median=2.07, Q1=1.74, Q3=2.38) and formaldehyde (median=2.19, Q1=1.84, Q3=2.50) in the South were highest among the four regions (table S6).

The joint PR for increasing all nine fossil-fuel and combustion-related air toxics and NO2 by one quartile was 1.02 (95%CI=1.01,1.04) (table 4). After stratifying by race/ethnicity, the mixture PR was 1.08 (95%CI: 1.04,1.12) for participants reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) and 1.01 (95%CI: 0.99,1.02) for non-Hispanic White participants (table 4). NO2 had a relatively greater positive weight in the mixture effect, and acrolein had a relatively greater negative weight (table S7). The joint effect among Black women was lower but in the same direction compared with the joint effect among women of other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) (data not shown).

Table 4.

Quantile-based g-computation analysis of fossil-fuel and combustion-related air pollution and hypertension

N Mixture OR (95%CI)
Fossil-fuel and combustion-related air toxics + NO2a Overall 47172 1.02 (1.01, 1.04)
Non-Hispanic White 40141 1.01 (0.99, 1.02)
Other (Hispanic/Latina, non-Hispanic Black and other) 7031 1.08 (1.04, 1.12)
Open in a new tab

Models were adjusted for age, race/ethnicity (not in models stratified by race), individual income and education, census-tract income and education, and region

a

Ambient NO2 concentrations were only obtained for participants living in continental United States (missing N=295), complete case analysis conducted with only those with NO2 concentrations included in the mixture.

In analyses using corrected continuous blood pressure as the outcome variable, comparing the fourth quartile of exposure with the first, we found that acetaldehyde (β=0.68, 95%CI=0.24,1.12; p trend=0.002), formaldehyde (β=0.75, 95%CI=0.31,1.20; p trend<0.001), and NO2 (β=0.86, 95%CI=0.43,1.30; p trend<0.001) were associated with higher SBP (table S8). The results in sensitivity analyses were largely similar (table S8). For acetaldehyde and formaldehyde, when using original blood pressure, the β estimate was still positive although there is no longer statistical significance and when excluding participants taking anti-hypertensive medication the association was near null. For NO2, the association when excluding participants taking anti-hypertensive medication was also attenuated.

Results did not change in sensitivity analysis using the updated definition of hypertension, additionally adjusting for smoking status, or stratifying by BMI and MET-hours (table S9 and table S10). In stratified analysis by smoking status, we found that the association between fossil fuel and combustion-related air pollution and hypertension was more apparent in never smoking participants and past smokers (tables S9).

Discussion

In this study, we found that fossil fuel and combustion-related air pollution was associated with higher risk of hypertension. We also observed an interaction between race/ethnicity and combustion-related air pollution, with more pronounced association for women with self-reported races/ethnicities (Hispanic/Latina, non-Hispanic Black and other) other than non-Hispanic White. The results of our study provide insight about components of air pollution from combustion of organic materials that may be important for hypertension risk as well as the role of the exposure mixture.

NO2, PAH, and Diesel Emissions

In an earlier study of air pollution and hypertension in the Sister Study, we found that exposure to particulate matter ≤2.5μm in diameter and NO2 were associated with higher blood pressure (Chan et al., 2015), which is consistent with this study. A meta-analysis of epidemiological studies also concluded that particulate matter and NO2 are associated with risk of hypertension and high blood pressure (Yang et al., 2018).

According to a study analyzing constituents of particulate matter, about a third of air pollution arises from gasoline, diesel, or coal combustion (Xue et al., 2019). PAHs are a group of chemicals that have multiple carbon rings and usually result from incomplete combustion of organic materials including gasoline and diesel. There are several epidemiological studies of the relationship between diesel exhaust or PAHs and hypertension with inconsistent results. A cross-sectional study observed a dose-dependent, positive association between PAHs and hypertension but not an association between diesel particulate matter and hypertension (Bangia et al., 2015). In our study, we did not find significant associations between PAHPOM, biphenyl, or naphthalene and hypertension in the full population, but we did find a positive association among women of other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other). An experimental study in humans using a crossover design showed that inhalation of diesel emissions can increase systolic blood pressure, but not diastolic blood pressure (Cosselman et al., 2012). In this study, we observed positive association of diesel emissions with prevalence of hypertension. We also observed a slightly increased diastolic blood pressure associated with diesel emissions. Possible mechanisms might involve the effect of exposure to PAHs and diesel emissions on the autonomic nervous system (Peretz et al., 2008; Zhang et al., 2008).

1,3-Butadiene

A cohort study observed that occupational exposure to 1,3-butadiene is associated with increased mortality from cardiovascular disease among Black male workers (Divine, 1990). However, the results of epidemiological studies of ambient exposure to 1,3-butadiene and hypertension are inconsistent. A cross-sectional study among adolescents and young adults (12-30 years old) found no association between 1,3-butadiene exposure and SBP (Lin et al., 2020). Another epidemiological study among adults aged 19-80 found that butadiene related VOCs were associated with decreased DBP (Shin et al., 2015). The different ages of populations in these two studies could account for the different results. There are also some studies focusing on pregnant women. Exposure to ambient 1,3-butadiene was not associated with gestational hypertension (Zhu et al., 2017), but was associated with increased risk of high blood pressure (Männistö et al., 2015). In our study, we observed a positive association between 1,3-butadiene and prevalence of hypertension in the full population, but did not find associations with continuous blood pressure. After stratifying by race/ethnicity, we found a higher risk among participants self-reporting races/ethnicities other than non-Hispanic White (Hispanic/Latina, non-Hispanic Black and other). The possible mechanisms of the association between 1,3-butadiene and hypertension might involve oxidative stress and DNA damage, which is supported by both experimental and epidemiological studies (Arayasiri et al., 2010; Erexson and Tindall, 2000).

Benzene

Three cross-sectional studies showed that those classified as having been exposed to benzene had a higher prevalence of hypertension, but these studies have relatively small sample sizes and participants were only divided into exposed and unexposed groups (Kasemy et al., 2019; Kotseva and Popov, 1998; Wiwanitkit, 2007). Benzene exposure among pregnant women was associated with increased risk of high blood pressure (Männistö et al., 2015), but not associated with gestational hypertension (Zhu et al., 2017). In our study, we linked benzene levels to participants’ residence at the census-tract level and observed an increased prevalence of hypertension associated with benzene exposure, with greater risk among participants with other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other). The mechanism for the association between benzene exposure and high blood pressure is still unclear; one possibility may be the suppression of circulating angiogenic cells which are sensitive to inhaled pollutants and are a predictive factor for cardiovascular diseases (Abplanalp et al., 2017).

Acrolein

Acrolein is an unsaturated aldehyde that exists in the air; the sources of airborne acrolein are the combustion of petroleum related materials (e.g., gasoline, diesel and wood burning) (Henning et al., 2017). Experimental studies show that in hypertensive rats, exposure to acrolein is associated with increased blood pressure, but no association was observed in normotensive rats (Perez et al., 2013). In an epidemiological study, the prevalence of hypertension among people with different urinary 3-hydroxyproplmercapturic acid (a biomarker representing acrolein exposure) levels was not significantly different (DeJarnett et al., 2014). Another epidemiological study shows that among people with moderate to high cardiovascular diseases risk, acrolein is positively associated with SBP, with Black participants having a larger association (McGraw et al., 2021). In our study, we found an increased prevalence of hypertension in the third quartile of acrolein exposure, but the trend was not monotonic in the full population. In women reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other), the association was stronger.

Acetaldehyde and Formaldehyde

Acetaldehyde and formaldehyde are main components of VOCs, and usually have the same emission sources (Delikhoon et al., 2018). In our study, we found a strong correlation between acetaldehyde and formaldehyde, which corroborates the evidence that these two air toxics have similar sources. Some experimental studies in rats (Strubelt et al., 1990; Takeshita. et al., 2009) and humans (Kupari et al., 1983) indicate that acetaldehyde and formaldehyde exposure is associated with health effects on the cardiovascular system, such as impaired left ventricular function. In these studies, the exposures were through ingestion or infusion, not from inhalation. Urinary acetaldehyde and formaldehyde in humans are highly correlated with ambient acetaldehyde and formaldehyde, suggesting airborne exposure might be a major source of environmental exposure (Tunsaringkarn. et al., 2012). In our study, we found that ambient air acetaldehyde and formaldehyde exposure was associated with higher prevalence of hypertension, with greater risk among participants self-reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black and other).

Stratification Analyses according to Race/ethnicity and Geographic Region

In analyses stratified by race/ethnicity, we found that among women reporting races/ethnicities other than non-Hispanic White, the association between fossil fuel and combustion-related air toxics and hypertension was stronger. Both the higher risk of hypertension and higher exposure levels may contribute to the stronger association among these women. The prevalence of hypertension in the US varies across races/ethnicities, with the highest prevalence in Black/African American people (1.5 times higher than non-Hispanic White people); the prevalence in Hispanic/Latina people is similar to non-Hispanic White people (Bennett et al., 2016). Black, Hispanic, and Asian people are reported to have higher levels of stress than White people. Higher stress may lead to an increase in smoking, alcohol consumption, and physical inactivity, which may increase the risk of hypertension (Williams et al., 2003). There are also racial/ethnical disparities in exposure to environmental pollution. Hispanic and Black people are more likely to live in areas with higher levels of industrial air toxics compared with White people in the U.S. because of disproportionate siting for neighborhoods of racial and ethnic minorities or employment opportunities (Ash and Boyce, 2018; Beent, 1994; Zwickl et al., 2014). In our study, women self-reporting other races/ethnicities (Hispanic/Latina, non-Hispanic Black, and other) lived in areas with greater fossil-fuel and combustion-related air pollution. The higher levels of air pollution and the stronger association with hypertension suggests the need to focus pollution regulation efforts on reducing air pollution levels in areas where Black people, Hispanic people and people with other races/ethnicities live.

After stratifying by region, we observed that women residing in the West had higher prevalence of hypertension when exposed to the same quartile categories of diesel emissions and women living in the south had higher prevalence of hypertension when exposed to the same level of VOCs (acetaldehyde and formaldehyde). We found that the West had the highest level of diesel emissions and the South had the highest level of acetaldehyde and formaldehyde among all four regions (Northeast, Midwest, South, West), which may partly explain the differing risks of hypertension across the different regions.

Limitations and Strengths

There are limitations of our study. This is a cross-sectional analysis so we could not consider the temporal relationship between fossil fuel and combustion-related air pollution and hypertension. We used census-tract level of air toxics from the EPA NATA database and census-block level of NO2 from monitoring data linked to each participants’ enrollment address to represent their exposure level instead of individual exposure measurements, which may cause misclassification due to differential behavior and activity patterns or residential mobility. Since estimates from the NATA database and NO2 concentrations represent annual average air pollutant concentrations, short-term temporality was not evaluated in this study. In stratified analyses, results did not differ by duration of living in the current residence (>10 years and ≤10 years). We used 2005 NATA and NO2 data to represent participants’ baseline residence exposure because EPA suggests not combining exposure data across multiple versions of NATA, although enrollment spanned 2003-2009. We categorized each participant’s exposure level according to quartiles, which should minimize misclassification if the quartile rankings remain the same. The concentrations of some air pollutants decreased from 2003 to 2009 (https://www.epa.gov/air-trends), so air pollution exposure may have been underestimated for participants enrolled before 2005 and overestimated for participants enrolled after 2005. When we restricted to those enrolled in 2005 (27.8% of the women), the effect estimates were larger, suggesting that exposure misclassification may have led to underestimation of the association. The association between studied air toxics/pollutants and hypertension may not be linear. The use of categorical exposure variables partially addresses this. We further used quantile-based g computation to explore non-linear fit of the joint association and did not see obvious deviations from linearity. We did not consider other environmental factors, such as temperature and humidity. However, we used annual average concentrations of combustions-related air toxics/pollutants, which might be less affected by temperature variability. Participants in this cohort have relatively higher socioeconomic status than a general population sample, which may limit generalizability. Adjusting for or stratifying on measures of socioeconomic status did not, however, change results. Participants enrolled in this study are women and findings may not generalize to men. They are also at higher risk for breast cancer which could also limit generalizability to the extent that this enhanced risk was related to the exposure and outcome in this study. Hypertension has been shown to be a risk factor for breast cancer (Han et al., 2017), but women in this study were breast cancer free at the time of study enrollment. Furthermore, the prevalence of hypertension in this cohort was not greater than in a population-based US sample (Ostchega et al., 2020), suggesting that any association between hypertension and breast cancer or breast cancer susceptibility did not lead to hypertension-related selection in this study. Shared exposures may account for some of the familial breast cancer risk and there is growing evidence that breast cancer is associated with traffic-related air pollution (Gabet et al., 2021), raising the possibility of selection bias related to higher air pollution exposure. However, the participants who are now adults do not necessarily live close to their sisters with breast cancer, and as noted, the higher socioeconomic status of study participants likely led to lower overall air pollution exposure in this cohort

Our study also had strengths. We used measured blood pressure to define hypertension instead of using just self-reported information, which provided a more accurate assessment. Another strength is that we not only considered the effect of individual fossil fuel and combustion-related air toxics or NO2 on hypertension, but also considered the effects of the overall air toxics/NO2 mixture. Additionally, the population in this study is spread nationwide, so the results give us a nationwide and regional view of the association between combustion-related air pollution and hypertension.

Conclusions

In conclusion, we found that women residing in areas with higher levels of air pollution from fossil fuel combustion had higher prevalence of hypertension. Importantly, we found that associations between fossil fuel combustion-related air pollution and hypertension were stronger among women self-reporting races/ethnicities other than non-Hispanic White.

Supplementary Material

1

Highlights.

  1. A large-scale study considering combustion-related air pollution and hypertension.

  2. Diesel emissions, VOCs and NO2 were associated with higher risk of hypertension.

  3. Associations varied by race/ethnicity and geographic region.

  4. Air toxics/NO2 mixture was associated with hypertension.

  5. Joint effect of mixture was stronger among people other than non-Hispanic White.

Acknowledgments

Supported by the Intramural Research Program of the NIH, NIEHS (Z01ES044005) and China Scholarship Council (201906210460).

Footnotes

Publisher's Disclaimer: 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.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Abplanalp W, DeJarnett N, Riggs DW, Conklin DJ, McCracken JP, Srivastava S, Xie Z, Rai S, Bhatnagar A, O’Toole TE, 2017. Benzene exposure is associated with cardiovascular disease risk. PLoS One 12, e0183602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Appel LJ, 2022. Smoking and hypertension.
  3. Arayasiri M, Mahidol C, Navasumrit P, Autrup H, Ruchirawat M, 2010. Biomonitoring of benzene and 1,3-butadiene exposure and early biological effects in traffic policemen. Sci Total Environ 408, 4855–4862. [DOI] [PubMed] [Google Scholar]
  4. Ash M, Boyce JK, 2018. Racial disparities in pollution exposure and employment at US industrial facilities. Proc Natl Acad Sci U S A 115, 10636–10641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Avila-Palencia I, Rodriguez DA, Miranda JJ, Moore K, Gouveia N, Moran MR, Caiaffa WT, Diez Roux AV, 2022. Associations of Urban Environment Features with Hypertension and Blood Pressure across 230 Latin American Cities. Environ Health Perspect 130, 27010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baguet JP, Legallicier B, Auquier P, Robitail S, 2007. Updated meta-analytical approach to the efficacy of antihypertensive drugs in reducing blood pressure. Clin Drug Investig 27, 735–753. [DOI] [PubMed] [Google Scholar]
  7. Bangia KS, Symanski E, Strom SS, Bondy M, 2015. A cross-sectional analysis of polycyclic aromatic hydrocarbons and diesel particulate matter exposures and hypertension among individuals of Mexican origin. Environ Health 14, 51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Beent V, 1994. Locally Undesirable Land Uses in Minority Neighborhoods: Disproportionate Siting or Market Dynamics? The Yale Law Journal 103, 1383–1422. [Google Scholar]
  9. Bennett A, Parto P, Krim SR, 2016. Hypertension and ethnicity. Curr Opin Cardiol 31, 381–386. [DOI] [PubMed] [Google Scholar]
  10. Chan SH, Van Hee VC, Bergen S, Szpiro AA, DeRoo LA, London SJ, Marshall JD, Kaufman JD, Sandler DP, 2015. Long-Term Air Pollution Exposure and Blood Pressure in the Sister Study. Environ Health Perspect 123, 951–958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cosselman KE, Krishnan RM, Oron AP, Jansen K, Peretz A, Sullivan JH, Larson TV, Kaufman JD, 2012. Blood pressure response to controlled diesel exhaust exposure in human subjects. Hypertension 59, 943–948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cosselman KE, Navas-Acien A, Kaufman JD, 2015. Environmental factors in cardiovascular disease. Nat Rev Cardiol 12, 627–642. [DOI] [PubMed] [Google Scholar]
  13. DeJarnett N, Conklin DJ, Riggs DW, Myers JA, O’Toole TE, Hamzeh I, Wagner S, Chugh A, Ramos KS, Srivastava S, Higdon D, Tollerud DJ, DeFilippis A, Becher C, Wyatt B, McCracken J, Abplanalp W, Rai SN, Ciszewski T, Xie Z, Yeager R, Prabhu SD, Bhatnagar A, 2014. Acrolein exposure is associated with increased cardiovascular disease risk. J Am Heart Assoc 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Delikhoon M, Fazlzadeh M, Sorooshian A, Baghani AN, Golaki M, Ashournejad Q, Barkhordari A, 2018. Characteristics and health effects of formaldehyde and acetaldehyde in an urban area in Iran. Environ Pollut 242, 938–951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Divine BJ, 1990. An update on mortality among workers at a 1,3-butadiene facility-preliminary results. Environ Health Perspect 86, 119–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Donaldson K, Stone V, Seaton A, MacNee W, 2001. Ambient particle inhalation and the cardiovascular system: potential mechanisms. Environ Health Perspect 109 Suppl 4, 523–527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Environmental Protection Agency, 2011. An Overview of Methods for EPA’s National-Scale Air Toxics Assessment, Office of Air Quality Planning and Standards, Research Triangle Park, NC.. [Google Scholar]
  18. Erexson GL, Tindall KR, 2000. Reduction of diepoxybutane-induced sister chromatid exchanges by glutathione peroxidase and erythrocytes in transgenic Big Blue mouse and rat fibroblasts. Mutat Res 447, 267–274. [DOI] [PubMed] [Google Scholar]
  19. Gabet S, Lemarchand C, Guenel P, Slama R, 2021. Breast Cancer Risk in Association with Atmospheric Pollution Exposure: A Meta-Analysis of Effect Estimates Followed by a Health Impact Assessment. Environ Health Perspect 129, 57012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. GBD 2019 Risk Factors Collaborators, 2020. Global burden of 87 risk factors in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 396, 1223–1249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Han H, Guo W, Shi W, Yu Y, Zhang Y, Ye X, He J, 2017. Hypertension and breast cancer risk: a systematic review and meta-analysis. Sci Rep 7, 44877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Henning RJ, Johnson GT, Coyle JP, Harbison RD, 2017. Acrolein Can Cause Cardiovascular Disease: A Review. Cardiovasc Toxicol 17, 227–236. [DOI] [PubMed] [Google Scholar]
  23. Holme JA, Brinchmann BC, Refsnes M, Låg M, Øvrevik J, 2019. Potential role of polycyclic aromatic hydrocarbons as mediators of cardiovascular effects from combustion particles. Environ Health 18, 74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kasemy ZA, Kamel GM, Abdel-Rasoul GM, Ismail AA, 2019. Environmental and Health Effects of Benzene Exposure among Egyptian Taxi Drivers. J Environ Public Health 2019, 7078024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J, 2005. Global burden of hypertension: analysis of worldwide data. Lancet 365, 217–223. [DOI] [PubMed] [Google Scholar]
  26. Keil AP, Buckley JP, O’Brien KM, Ferguson KK, Zhao S, White AJ, 2020. A Quantile-Based g-Computation Approach to Addressing the Effects of Exposure Mixtures. Environ Health Perspect 128, 47004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kershaw S, Gower S, Rinner C, Campbell M, 2013. Identifying inequitable exposure to toxic air pollution in racialized and low-income neighbourhoods to support pollution prevention. Geospat Health 7, 265–278. [DOI] [PubMed] [Google Scholar]
  28. Kotseva K, Popov T, 1998. Study of the cardiovascular effects of occupational exposure to organic solvents. Int Arch Occup Environ Health 71 Suppl, S87–91. [PubMed] [Google Scholar]
  29. Kupari M, Lindros K, Hillbom M, Heikkilä J, Ylikahri R, 1983. Cardiovascular effects of acetaldehyde accumulation after ethanol ingestion: their modification by beta-adrenergic blockade and alcohol dehydrogenase inhibition. Alcohol Clin Exp Res 7, 283–288. [DOI] [PubMed] [Google Scholar]
  30. Lin CY, Lee HL, Jung WT, Sung FC, Su TC, 2020. The association between urinary levels of 1,3-butadiene metabolites, cardiovascular risk factors, microparticles, and oxidative stress products in adolescents and young adults. J Hazard Mater 396, 122745. [DOI] [PubMed] [Google Scholar]
  31. Männistö T, Mendola P, Liu D, Leishear K, Sherman S, Laughon SK, 2015. Acute air pollution exposure and blood pressure at delivery among women with and without hypertension. Am J Hypertens 28, 58–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McGraw KE, Riggs DW, Rai S, Navas-Acien A, Xie Z, Lorkiewicz P, Lynch J, Zafar N, Krishnasamy S, Taylor KC, Conklin DJ, DeFilippis AP, Srivastava S, Bhatnagar A, 2021. Exposure to volatile organic compounds - acrolein, 1,3-butadiene, and crotonaldehyde - is associated with vascular dysfunction. Environ Res 196, 110903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mills KT, Bundy JD, Kelly TN, Reed JE, Kearney PM, Reynolds K, Chen J, He J, 2016. Global Disparities of Hypertension Prevalence and Control: A Systematic Analysis of Population-Based Studies From 90 Countries. Circulation 134, 441–450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Novotny EV, Bechle MJ, Millet DB, Marshall JD, 2011. National satellite-based land-use regression: NO2 in the United States. Environ Sci Technol 45, 4407–4414. [DOI] [PubMed] [Google Scholar]
  35. Ostchega Y, Fryar CD, Nwankwo T, Nguyen DT, 2020. Hypertension Prevalence Among Adults Aged 18 and Over: United States, 2017–2018. Centers for Disease Control and Prevention National Center for Health Statistics. [Google Scholar]
  36. Peretz A, Sullivan JH, Leotta DF, Trenga CA, Sands FN, Allen J, Carlsten C, Wilkinson CW, Gill EA, Kaufman JD, 2008. Diesel exhaust inhalation elicits acute vasoconstriction in vivo. Environ Health Perspect 116, 937–942. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Perez CM, Ledbetter AD, Hazari MS, Haykal-Coates N, Carll AP, Winsett DW, Costa DL, Farraj AK, 2013. Hypoxia stress test reveals exaggerated cardiovascular effects in hypertensive rats after exposure to the air pollutant acrolein. Toxicol Sci 132, 467–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sandler DP, Hodgson ME, Deming-Halverson SL, Juras PS, D’Aloisio AA, Suarez LM, Kleeberger CA, Shore DL, DeRoo LA, Taylor JA, Weinberg CR, 2017. The Sister Study Cohort: Baseline Methods and Participant Characteristics. Environ Health Perspect 125, 127003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Shin HH, Jones P, Brook R, Bard R, Oliver K, Williams R, 2015. Associations between personal exposures to VOCs and alterations in cardiovascular physiology: Detroit Exposure and Aerosol Research Study (DEARS). Atmospheric Environment 104, 246–255. [Google Scholar]
  40. Spector D, Deroo LA, Sandler DP, 2011. Lifestyle behaviors in black and white women with a family history of breast cancer. Prev Med 52, 394–397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Strubelt O, Brasch H, Pentz R, Younes M, 1990. Experimental studies on the acute cardiovascular toxicity of formalin and its antidotal treatment. J Toxicol Clin Toxicol 28, 221–233. [DOI] [PubMed] [Google Scholar]
  42. Tainio M, Jovanovic Andersen Z, Nieuwenhuijsen MJ, Hu L, de Nazelle A, An R, Garcia LMT, Goenka S, Zapata-Diomedi B, Bull F, Sa TH, 2021. Air pollution, physical activity and health: A mapping review of the evidence. Environ Int 147, 105954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Takeshita D, Nakajima-Takenaka C, Shimizu J, Hattori H, Nakashima T, Kikuta A, Matsuyoshi H, Takaki M, 2009. Effects of formaldehyde on cardiovascular system in In Situ rat hearts. Basic & Clinical Pharmacology & Toxicology 105, 271–280. [DOI] [PubMed] [Google Scholar]
  44. Tobin MD, Sheehan NA, Scurrah KJ, Burton PR, 2005. Adjusting for treatment effects in studies of quantitative traits: antihypertensive therapy and systolic blood pressure. Stat Med 24, 2911–2935. [DOI] [PubMed] [Google Scholar]
  45. Tunsaringkarn T, Siriwong W, Prueksasit T, Sematong S, Zapuang K, Rungsiyothin A, 2012. Potential risk comparison of formaldehyde and acetaldehyde exposures in office and gasoline station workers. International Journal of Scientific and Research Publications 2. [Google Scholar]
  46. Whelton PK, Carey RM, Aronow WS, Casey DE Jr., Collins KJ, Dennison Himmelfarb C, DePalma SM, Gidding S, Jamerson KA, Jones DW, MacLaughlin EJ, Muntner P, Ovbiagele B, Smith SC Jr., Spencer CC, Stafford RS, Taler SJ, Thomas RJ, Williams KA Sr., Williamson JD, Wright JT Jr., 2018. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol 71, e127–e248. [DOI] [PubMed] [Google Scholar]
  47. Williams DR, Neighbors HW, Jackson JS, 2003. Racial/ethnic discrimination and health: findings from community studies. Am J Public Health 93, 200–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wiwanitkit V, 2007. Benzene exposure and hypertension: an observation. Cardiovasc J Afr 18, 264–265. [PubMed] [Google Scholar]
  49. World Health Organization, Body mass index - BMI, http://www.euro.who.int/en/health-topics/disease-prevention/nutrition/a-healthy-lifestyle/body-mass-index-bmi.
  50. Xue Q, Jiang Z, Wang X, Song D, Huang F, Tian Y, Huang-Fu Y, Feng Y, 2019. Comparative study of PM(10)-bound heavy metals and PAHs during six years in a Chinese megacity: Compositions, sources, and source-specific risks. Ecotoxicol Environ Saf 186, 109740. [DOI] [PubMed] [Google Scholar]
  51. Yang BY, Qian Z, Howard SW, Vaughn MG, Fan SJ, Liu KK, Dong GH, 2018. Global association between ambient air pollution and blood pressure: A systematic review and meta-analysis. Environ Pollut 235, 576–588. [DOI] [PubMed] [Google Scholar]
  52. Zhang HM, Nie JS, Wang F, Shi YT, Zhang L, Antonucci A, Liu HJ, Wang J, Zhao J, Zhang QL, Wang LP, Song J, Xue CE, Di Gioacchino M, Niu Q, 2008. Effects of benzo[a]pyrene on autonomic nervous system of coke oven workers. J Occup Health 50, 308–316. [DOI] [PubMed] [Google Scholar]
  53. Zhu Y, Zhang C, Liu D, Ha S, Kim SS, Pollack A, Mendola P, 2017. Ambient Air Pollution and Risk of Gestational Hypertension. Am J Epidemiol 186, 334–343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zwickl K, Ash M, Boyce JK, 2014. Regional variation in environmental inequality: Industrial air toxics exposure in U.S. cities. Ecological Economics 107, 494–509. [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS1843757-supplement-1.docx (266.1KB, docx)

RESOURCES