REVIEW ARTICLE

Advancing environmental exposure

BY
J. ANNS NIBI SHERINA
180711001
ME (COMMUNICATION
SYSTEMS)
ABSTRACT
Awareness of the human health impacts of exposure to air pollution is
growing rapidly. For example, it has become evident that the adverse health
effects of air pollution are more pronounced in disadvantaged populations.
Policymakers in many jurisdictions have respon-ded to this evidence by
enacting initiatives that lead to lower concentrations of air pollutants, such
as urban traffic restrictions. In this review, we focus on the interplay between
advances in environmental exposure assessment and developments in policy.
We highlight recent progress in the granular measurement of air pollutants
and individual-level exposures, and how this has enabled focused local
policy actions. Finally, we detail an illustrative study designed to link
individual-level health-relevant exposures with economic, behavioral, bio-
logical, familial, and environmental variables.

INTRODUCTION
Awareness of the human health impacts of exposure to air pollution
is growing rapidly and is increasingly being translated into public
policy. A particular spur to action is that the effects of air pollution are
often most clearly evidenced among the most dis-
advantaged populations. For example, children in households with low-
socioeconomic status (SES) have worse health at birth than those with high
SES. This gradient between SES and health steepens during life, and various
health conditions such as respiratory disease can attribute to this relationship.
As a respiratory disease, childhood asthma can be aggravated and induced
by air pollution exposure. Those who are less well-off often live in areas
where the air contains higher levels of hazardous particulate matter (PM),
thus having a compound effect on health conditions later in life. While all
are of us are affected by air pollution exposure, the combination of
compromised baseline health and lower air quality can inflict even more
profound long-term health effects.
Given this sobering reality, policymakers in many countries and
communities have begun responding appropriately by controlling or banning
dangerous air pollutants, requiring cleaner heating fuels, restricting urban
traffic, etc. The political will for cleaner air presents us with a tremendous
opportunity. The more that factors determining pollutant exposures are
detailed, the better individuals and policymakers can mitigate elevated
exposure and adverse health effects. In this paper, we focus precisely on the
interplay between advances in environmental exposure assessment on the
one hand, and policy advances on the other.
In the past, pollution measurement has been relatively crude in terms of
compositional, spatial, and temporal resolution. There have been significant
improvements in the ability to conduct granular measurement of air
pollutants, and this has enabled more awareness and
more focused local policy actions, as addressed example, estimates by Public Health England
below. We out-line more recent progress and the have indicated that the exposure
most current practices in the ability to measure
pollution exposure at the individual level, rather
than in geographical terms. The level of practical
use of these personal monitoring methods has not
yet been developed to the point of directly impacting
policy (e.g., by their application for ambient air to PM and nitrogen dioxide air pollution in
regulatory purposes), but the potential is high. We London resulted in 41,404 life years lost in 2010,
detail an illustrative study designed to turn that with an estimated associated annual financial
potential into practice by measuring individuals’ impact of £3,653 million. Worldwide, WHO has
exposures over a pro-longed period of time, along estimated that each year, more than 7 million
with measures of factors that influence health premature deaths can be attributed to the combined
outcomes, including economic, behavioral, effects of indoor and outdoor air pollution. Overall,
biological, familial, and environmental contexts. 80% of the premature deaths attributed to outdoor
Though other parallel projects may well be in air pollution are caused by ischemic heart disease
planning elsewhere, we describe this initiative, and strokes, 14% are due to chronic obstructive
which is currently being applied in New York City, pulmonary disease or acute lower respiratory
as a pilot study for the method applied in a very infections, and 6% are due to lung cancer, primarily
from expo-sure to fine PM2.5. A recent Global
large metro-politan area. We expect that such Burden of Disease (GBD) study reported that 4.2
individual-level studies will deepen the million annual deaths were attributable to ambient
understanding of the health outcomes of pollution PM2·5, making it the fifth-ranking mortality risk
exposure, enable personal preventive actions, factor in 2015 globally, and that, if trends of
motivate collective adjustment, thereby enabling increasing levels of air pol-lution continue in low-
more effective policies. There are challenges in and middle-income countries, the increasing health
enabling policies that limit personal exposures, but burdens will be significant if a substantial reduction
the likely improvements in health and welfare are of pollution is not achieved. Compounding this
potentially massive. concern, a recent reanalysis of available air
pollution cohort studies has revealed that past GBD
estimates of the mortality effects, which relied on
Pollution and poverty the integrated exposure-response, have
Common outdoor “community”-based air pollutants
that have been identified as associated with
increasing risks of cardiovas-cular and respiratory
diseases, lung cancer, and early death, include ozone
(O3); fine PM2.5 and elemental carbon (EC) soot;
nitrogen dioxide (NO2); and sulfur dioxide (SO2).
Recently, increased attention has also been directed
toward ultrafine par-ticles, defined as particles with
≤0.1 µm in aerodynamic diameter, as they also have
been associated with adverse health effects, largely
independent of PM2.5 mass. Elevated exposures to
PM2.5 are especially widespread, with among the
most dan-gerous of these apparently being from
fossil fuel combustion, especially coal burning,
likely because of these particles’ small size and
especially toxic composition. These exposures have
been estimated to shorten adult life expectancy by
1–2 years, while also damaging long-term cardiac
and pulmonary health. Recent evidence also
implicates exposure to a range of adverse health
outcomes, including neurological diseases, cardio-
metabolic diseases, and renal disease.
Understanding of the widespread impact of air
pollution on human health, and the associated
economic impact, has improved in recent years. For
been underestimating the effects of outdoor air Other studies expand on the link between SES
pollution on human mortality. status and health effects of pollution. As reported for
Of the health problems known to be worsened by Sweden, fluctuations in pollution caused by
air pollution, asthma is among the most widespread, atmospheric inversions were used as indi-cators of
especially among the children. A study looking at pollution effects on health, while considering the
asthma prevalence and health care in the U.S. found role of SES. Inversions are transient episodes when
that, in 2008, persons with asthma missed 10.5
million school days and 14.2 million work days due the usual gradient of near surface air temperature
to their asthma and that, in 2007, there were 1.75 (cooling with increasing elevation) reverses,
million asthma-related emergency department visits resulting in an atmospheric trapping of pollutants.
and 456,000 asthma hospitalizations. A majority of The authors used information on pollution, weather,
childhood asthma is estimated to be induced by and precipitation generated by NASA and the
environmental exposures. In 1 year alone, the cost Swedish Environmental Research Institute, as well
of childhood exposure to environmental as pollution measured at fixed sites (typically
contaminants resulting in asthma was estimated to
be close to 2 billion USD. Several recent reviews located in town centers in Sweden). They acquired
have found evidence for a link between air pol- health data from Statistics Sweden, as well as
lution and incidence of new cases of child and adult inpatient and outpatient data covering all those of
asthma, however, the evidence has still been age 18 or under. The authors found that the
considered by some as yet insufficient proof of a worsening of pollution associated with atmospheric
causal link. inversions increased health care visits for
Poverty is known to be a critical factor in respiratory illnesses, and that the impact on children
susceptibility to pollution-related asthma; a 1999 from high-income households was sig-nificantly
study in Los Angeles found that the air pollution- lower than that on children in low-income
related asthma admissions were higher for families households, indicating that baseline health played a
of lower income, independent of health insurance
coverage. Similarly, an analysis of records of daily key role. While the impact was virtually identical
hospital admissions and air pollution in New York across income groups for children in poor health,
City (NYC), estab-lished that an apparent difference among children in good health the impact of
between the races in the risk of hospital admissions pollution was much larger for low and medium
was explained by consideration of lower SES, income than for high-income families.
where the poor and the working poor were found to
have higher risks per amount of pollution exposure
than wealthier individuals, irrespective of race. Pollution policy at the global and national level
Also, a recent US study of some 60 million older Air pollution exposure is a global health concern:
adults found that the mortality effects from long- some 90% of the world’s population is breathing
term air pollution exposure in the elderly Medicare outdoor air that falls below the World Health
population was greater amongst those eligible for Organization (WHO) air quality guidelines. In the
Medicaid (i.e., those with lower income) than others US alone, more than 125 million people are
in that population, again indicating that SES is a estimated to live in communities with unhealthy air.
primary risk factor that increases susceptibility to In the developing world, while accelerated
adverse health outcomes of air pollution. economic development and indus-trialization has
led to a reduction of poverty, it has also resulted in
greater economic disparities and extreme poor air has resulted in positive action, differences in air
qualities, especially in megacities. There are many quality across individual neighborhoods, and at
health risks and eco-nomic impacts associated with ground-level, are not captured by such coarsely
such pollution exposures, including a diminished distributed and elevated air monitors.
lifelong economic productivity that results from
permanent reduction in physical health and
cognitive cap-ability, compounded by increased
costs of health care and early morbidity. The chemical constituents and sources of
Knowledge of the damage that pollutants impose pollution are also often studied with low specificity.
on health has been instrumental in policies and The label “fine particulate matter mass” is highly
global accords; the Paris Agreement being one nonspecific, even though the mass is comprised of
example. Recent progress in the scientific many different particles with differing compositions
understanding of pollution, and the air quality and sources that can have differing health
implications of anthropogenic climate change implications. For example, studies have been
mitigation policies, highlight the progress for carried out on differentiation of outdoor PM2.5 (fine
cleaner air can immediately benefit society at large. mass) exposures across the US by composition and
source category. In the US, these were then incorpo-
rated into a nationwide cohort study to differentiate
Improving air pollution policy at the local level the respective ischemic cardiovascular mortality
While there have been striking air quality policy impacts of PM2.5 from different sources, finding that
successes in the past, such as the US Clean Air Act, coal combustion particles were approximately five
there are limitations as to what can be achieved times the health risk for this outcome than other
when relying only on traditional central-site air particles. Hence, improving the study design
quality measurements that often deliver only requires not only finer measures of time and spatial
infrequent measures of just a few select pollutants. exposure resolution, but also to further distinguish
It is now known that air pollution, and its pollution constituents/ sources. These
composition, often varies dramatically over short differentiations are then linked to health out-comes,
distances and periods of time. Likewise, awareness helping guide the optimal public health strategies
has grown regarding the links between pollution and and respective regulatory efforts. In the next section,
poverty. The geographic link between asthma and we provide examples where finer resolution
poverty in NYC is one example, as indicated: (a) the monitoring strategies have been
spatial dis-tribution of asthma admissions rates by
children aged 5–14 (per 10,000 population) across
the 59 Community Districts in NYC during 2010–
2014; and (b) the percent of poverty in the
population in each of the same Districts in the same
time period. The spatial correlation between poor
health and poverty is apparent; a statistical analysis
of these data also indicates a sig-nificant association
between health and wealth in this case (r = 0.35, p =
0.01).
The asthma admissions rates vary drama-tically
across New York City. This suggests that it is
important to understand local patterns of pollution
exposures. Unfortunately, it has been common to
measure pollutant exposure levels to represent large
and highly heterogeneous areas (e.g., city or county-
wide). The fixed monitoring stations that are used in
the U.S. mainly for regulatory purposes are often
sparsely located and elevated on buildings. New
York City, for example, has 13 high-performance
regulatory air monitoring stations. While this level
of instrumentation has been of great value, and
adopted, and where air quality policies have been makers, developers, and residents to collaborate and
changed as a result. take action.
While not yet numerous, there are examples of Other approaches have employed land use
approaches that result in highly granular regression (LUR) methods to empirically model air
measurement of pollution over large areas. In pollution levels using both geographic variables and
London, for example, King’s College London regulatory stationary monitoring data. LUR models
“London Air” website provides hourly exposure that incorporate mobile sampling have also recently
estimates at a 20 m reso-lution across London for gained traction as a cost-effective approach to
NO2, O3, PM 10 (thoracic PM mass less than 10 µm model urban air quality. Mobile sampling consists
in diameter), and PM2.5, using models relying on of continuous monitors mounted on mobile
data from monitoring networks. With recent platforms (e.g., foot, bicycle, and cars) repeatedly
advancement and proliferation of low-cost sensor sampling travel routes, and then applying statistical
technology, networks of such sensors are also being approaches to predict air pollution concentration
deployed across urban locations to measure air levels. For example, efforts in Hong Kong,
pollution concentration levels. Examples include
Montreal, Minneapolis, and San Francisco have
deployments of a distributed network of sensors in
successfully modelled intra-urban air quality with
eight locations in Xi’an, China to measure and
locate pollution hotspots and 32 elec-trochemical high-predictive power and increased spatial
sensor nodes deployed over a 2-month period in resolution. For example, a Google Street View
Cambridge, UK, collecting high spatial and mapping vehicle, outfitted with air quality sensors,
temporal resolution data on carbon monoxide (CO) sampled every street in a 30-km2 area of Oakland,
concentrations. Researchers also recently carried CA in order to reveal urban air pollution patterns at
out sampling at 25 locations in Rochester, NY using 4–5 orders of magnitude greater spatial precision
low-cost sensors to estimate hourly concentrations than possible with regulatory central-site ambient
of PM2.5. As demonstrated by these studies, the low- monitoring. Vehicle-based mobile measure-ments
cost, ease of operation, and the high volume of data were employed to create LUR models to estimate
generated by sensor networks can allow for the spatial variation of PM2.5 and PM10 in the
streamlined deployment and maintenance of downtown area of Hong Kong. The “OpenSense”
multiple units, providing improved and detailed project in Zurich, Switzerland used mobile sensor
spatio-temporal coverage in a given area. Cities and nodes installed on top of public transport tram
government agen-cies around the world are vehicles in the city to create pollution maps with a
increasingly engaged in partnerships for the high spatial resolution for ultrafine particles and
application of sensor technologies, for example, the particle counts. This project aimed to use ultrafine
city of Chicago, in partnership with University of particle maps created with mea-surements from
Chicago and Argonne National Laboratory, recently mobile air pollution monitoring network to build
launched Array of Things (AoT), which consists of apps to minimize resident’s exposure by providing
a network of sensor boxes mounted on light posts to optimal pollution-avoidance travel routes.
collect a host of real-time data on Chicago’s Deployment of improved measurement
environ-mental surroundings (e.g., air quality, technologies is now allowing policymakers to target
temperature, and wind speed) and urban activity. their mitigation and evaluation measures with far
The AoT aims to provide real-time, location-based greater precision. This happened, for example, in
data to the public, allowing researchers, policy- three Israeli cities, which resulted in switching from
heavier oil-fueled power plants to cleaner burning
natural gas, and
a corresponding reduction in mortality. Spatially known as EC soot, in particular. The study found
granular analysis of pollution has been carried out increased risk of cough, wheeze, and total symptoms
in London, resulting in seminal policy progress in to be associated with personal exposure to increased
traffic congestion pricing, while bans on diesel fuel diesel EC. The findings pinpointed the diesel “soot”
have been put in place in many European cities. fraction of PM2.5 as most responsible for pollution-
To illustrate how improved measurement and related asthma exacerbations among children living
policy interact, we cite one particular case in New near roadways. This helped identify the particular
York City, where a detailed study of patterns of diesel fuels that were most responsible, for which
pollution resulted in certain diesel fuels being substitutes were readily avail-able at only
banned. The 2005–2007 campaign involved the marginally higher prices. This led to calls for
chemical com-position analyses of pollution, reductions in these exposures, with resulting
carried out by the NYC Depart-ment of Health and government actions for less diesel traffic, and
Mental Hygiene (NYC DOHMH), as part of a limitations on idling vehicles near public schools in
program called the New York City Community Air NYC. These examples have documented how
Survey (NYCCAS). In this ongoing study, spatial improved science can lead to more effectively
variations in pollutant concentrations have been targeted public health action.
measured using 2-week averaged concentrations. Analysis of data from the NYC studies above have
These data were measured using street level shown the within-city variations in air pollution-
monitors that have been methodically rotated related premature mortality and morbidity at 42 city
among approxi-mately 150 sampling locations neighborhoods, as well as the benefits of the
around the city, incorporating unusually rich elimination of the low-quality fuel oil. Study results
chemical analysis, including PM2.5, nitrogen oxi- indicated that the complete phase out of the low-
quality building fuel oil reduced the annual average
des, black carbon, ozone, and sulfur dioxide.
level of PM2.5 by an estimated 0.71 μg/m3 citywide
Critically, the PM2.5 itself was analyzed for
(~0.5% of city-wide mean concentration). This was
constituents, such as aluminum, bromine, calcium, further estimated to have resulted in an annual
copper, iron, potassium, manganese, sodium, nickel, decrease of 290 premature deaths, 180 fewer
lead, sulfur, silicon, titanium, vanadium, and zinc. hospital admissions for respiratory and
The measured concentrations have been used, along cardiovascular complications, and 550 fewer
with population density and attributes of the emergency department visits for asthma. The largest
physical infrastructure, as inputs to a LUR model in reductions of air pollutant emissions were in more
order to estimate pollutant level and composition at affluent neighborhoods, but the greatest health
a spatial resolution of 300 m, resulting in the benefits were estimated to occur in the high-poverty
identification of sources. Once the rich details of neighborhoods. This disparity in effects is attributed
pollutant concentrations were established at the to the higher baseline morbidity and mortality rates
local level in New York City, it was then pos-sible at lower income levels. These estimates of the 2005–
for researchers to gain insight into the average 2007 human toll and cost of PM air pollution
exposures of those living near particular sources of exposures. The valuation shown was calculate with
pollution. As a result, in April of 2011, specifically BenMAP methodology developed by USEPA, using
citing the NYCCAS study, New York City the events as function of age group from studies
Department of Environmental Protection (DEP) set carried by NYCDOH. The citywide 3200 premature
a new rule to phase out the use of two highly death can be compared to the 290 decreases in
polluting forms of heating oil: those known as deaths as result of the switch to cleaner fuel.
Number 6 and Number 4.
However, while the broad correlation between Conclusions
adverse respiratory health outcomes and residential This paper outlines the current state of research on
proximity to traffic pollution is known in New York spatial mapping of air pollution exposures, and the
City and elsewhere, we are still learning about the quantitative analysis of their human health effects.
individual contributions of different pollu-tants. In a It highlights the interaction of science with policy
panel study among inner-city children with asthma,
a study was used to evaluate the associations of and the many places advances in scientific
adverse asthma outcome incidences with increased understanding have produced appropriate policy
personal exposure to PM2.5 as a whole, and the responses. We outline newly feasible approaches to
diesel-related carbonaceous fraction of PM2.5 measuring exposures and health outcomes over the
course of the human life cycle in their economic,
geographic, familial, and social context.
 AARP diet and health cohort. Environ. Health
Perspect. 124, 484–490 (2015).
As such extensive new data becomes available, 3. Deschênes, O., Greenstone, M. & Shapiro, J. S.
policymakers will be tooled to pinpoint and Defensive Investments and the demand for air
eliminate key environmental hazards. Such policies quality: evidence from the NOx budget program.
are particularly necessary to protect those who are Am. Econ. Rev. 107, 2958–2989 (2017).
most vulnerable, including the young, the old, and 4. Sunyer, J. et al. Association between traffic-
the less well-off. related air pollution in schools and cognitive
We have documented examples of ways in which development in primary school children: a
scientific progress has resulted in public policy prospective cohort study. PLOS Med. 12,
improvements, and the opportunities and technical e1001792 (2015).
feasibility of increased assessment of individual 5. Chen, J.-C. et al. Ambient air pollution and
exposures and health effects. We end with a descrip- neurotoxicity on brain structure: evidence from
tion of a prototype study designed for the women’s health initiative memory study. Ann.
assessment of air pollution exposures and effects Neurol. 78, 466–476 (2015).
over an individual’s life cycle. It is possible that 6. Kheirbek, I. et al. The public health benefits of
such air pollution research will have policy out- reducing fine particulate matter through
comes as strong as those that followed research on conversion to cleaner heating fuels in New York
the health impacts of smoking. In that case,
City. Environ. Sci. Technol. 48, 13573–13582
scientific awareness of its damaging effects gave
rise to public health initiatives that resulted in (2014). New York City recently issued
dramatic declines in smoking-related deaths in the regulations to reduce emissions resulting from
developed world. Along these lines, the recent First burning heating oil. This paper highlights the
WHO Global Conference on Air Pollution and impact of such science-driven policies by
Health in Geneva, Switzerland (30 October–1 exploring the potential health benefits of these
November 2018) highlighted the need to make regulations, finding that although the reductions
commitments to air quality around the globe. By are greatest in low-poverty neighborhoods, the
personalizing benefits are estimated to be greatest in high-
T
poverty neighborhoods due to higher baseline
he exposures and impacts, such studies can
contribute to this advancement of environmental mortality rates.
health science and policy, and thereby have the 7. Matte, T. D. et al. Monitoring intraurban spatial
potential to engender more optimal societal changes patterns of multiple combustion air pollutants
to the benefit of everyone, perhaps particularly for in New York City: design and implementation.
society’s most adversely affected individuals and J. Expo. Sci. Environ. Epidemiol. 23, 223–231
groups. (2013).
8. World Health Organization. Poverty and Health.
References (2003).
1. Kheirbek, I., Haney, J., Douglas, S., Ito, K. & 9. Shmool, J. L. C. et al. Area-level socioeconomic
Matte, T. The contribution of motor vehicle deprivation, nitrogen dioxide exposure, and term
emissions to ambient fine particulate matter birth weight in New York City. Environ. Res.
public health impacts in New York City: a 142, 624–632 (2015).
health burden assessment. Environ. Health 15, 10. Bell, M. L. & Ebisu, K. Environmental
89 (2016). inequality in exposures to airborne particulate
2. Thurston, G. D. et al. Ambient particulate matter matter components in the United States.
air pollution exposure and mortality in the NIH- Environ. Health Perspect. 120, 1699–704
(2012).