Occupational Medicine 2005;55:425–431
doi:10.1093/occmed/kqi138
IN-DEPTH
REVIEW
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Point sources of air pollution
Andrew Kibble1 and Roy Harrison2
............................................................................................................................................................
Abstract
Many people live near point sources of air pollution such as industrial sites and waste disposal
operations and there are often suggestions of clusters of disease around such activities. Such alleged
clusters will generate significant public concern and media interest and in many cases will warrant
detailed investigation. However, the ability of current epidemiological methods to investigate such
clusters is limited, particularly with regard to obtaining reliable and accurate population exposure data.
In many cases, the key question is whether releases from a point source result in a significant increase in
exposure or whether other sources (background exposure) give rise to the dominant exposure. This
review considers some of the issues around point sources including methods of estimating exposure
and briefly discusses some of the epidemiological evidence linking respiratory disease and cancer with
specific industries such as coking works and incinerators.
............................................................................................................................................................
Key words
Air pollution; environmental exposure; epidemiology; point sources.
............................................................................................................................................................
Introduction
There is ample evidence that air pollution can have an
adverse effect on the health of susceptible people (i.e.
young children, the elderly and particularly those with
pre-existing respiratory disease) [1]. There are often
suggestions of clusters of disease around point sources of
pollution such as industrial sites and waste disposal
operations. Such alleged clusters will generate significant
public concern and media interest and in many cases will
warrant detailed investigation. Many clusters occur
purely by chance but, to the concerned lay people,
these clusters appear remarkable confirming their suspicions of a major health risk.
What are point sources?
Most air pollutants arise from a wide range of sources
such as industry, traffic or natural sources (e.g. sea-spray)
and many are ubiquitous in the atmosphere. Air
pollutants are classified into two distinct types: primary
and secondary pollutants.
Primary pollutants are released directly into the
atmosphere from specific sources of pollution, such as
industry or motor vehicles. Once emitted into the
atmosphere, some of these primary pollutants can be
altered by light energy, heat or the presence of other
chemicals (e.g. oxygen) to form secondary pollutants.
Ozone, for example, is a secondary pollutant formed as a
result of photolysis (the splitting of molecules by means of
1
Health Protection Agency, Chemical Hazards & Poisons Division (Birmingham),
Birmingham Research Park, 97 Vincent Drive, Edgbaston, Birmingham B15
2SQ, UK. Tel: +44 121 414 6547; e-mail: [email protected]
2
School of Geography, Earth & Environmental Sciences, The University of
Birmingham, Edgbaston, Birmingham B15 2TT, UK. Tel.: +0121 414 3494.
light energy) of nitrogen dioxide and subsequent reactions with volatile organic compounds (VOCs).
Many people live near point sources of air pollution,
that can be defined as a stationary location or fixed facility
from which pollutants are released [2]. Examples of point
sources include power stations, steel works, foundries,
incinerators, wood and pulp processors, paper mills,
refineries and chemical production. Most industrial point
sources release pollution into the atmosphere through
chimneys at a height sufficient to provide ample dilution
before the pollutants reach ground level. However,
certain meteorological conditions may prevent or reduce
the effectiveness of this dispersion and pollutants may
become trapped near the source and descend to ground
level where they may cause poor air quality. For example,
in 1998 high concentrations of sulphur dioxide were
recorded across much of the Midlands and South
Yorkshire as a result of emissions from coal-fired power
stations becoming trapped near ground level by a
combination of weather conditions including a very
stable atmosphere, low temperature and low winds [3].
Apart from point sources, air pollution may be
released from non-point sources such as mobile sources
(e.g. road traffic or aircraft emissions) and area sources
(e.g. emissions from many smaller stationary sources
such as homes, dry cleaners, petrol stations, etc.). For
many common air pollutants, motor vehicles are the
largest source of emissions in the UK, e.g. traffic
contributes about 49% of the total emission of oxides of
nitrogen [4].
It is also important to distinguish between point
sources and fugitive emissions. The latter typically relate
to pollutant releases from points in a process that are not
associated with a defined process stream such as a stack.
q The Author 2005. Published by Oxford University Press on behalf of the Society of Occupational Medicine.
All rights reserved. For Permissions, please email: [email protected]
426 OCCUPATIONAL MEDICINE
Examples of fugitive releases include losses from pipe
work or wind-blown dust from stockpiles. Occasionally
these releases can be of significance at a local level.
Releases from point sources can include complex
mixtures of substances and the pollutants released will be
dependent on process input materials, type of process,
etc. Examples are given in Table 1 which lists the
pollutants requiring regulation under Pollution Prevention and Control [5].
A summary of the effect of exposure to these chemicals
can be found elsewhere [1,6,7]. In all cases, the key
question relates to the extent of exposure of the public to
these chemicals. Where pollution abatement procedures
are operated as prescribed, then exposure from point
source emissions should be very low.
People living near point sources can be exposed
through a number of pathways depending on the point
source and the type of release. For most pollutants
inhalation will be the main source of exposure but for
some, indirect pathways such as food will assume greater
importance. Pollutants may enter the food chain through
deposition into soil or onto plants and subsequent uptake
in crops and farm animals. Exposure via the food chain
may be a potential problem if locally grown or reared
produce is important to the diet of local people where
groups such as local allotment owners and farmers may
need particular consideration. Such processes are most
important for persistent organic pollutants such as dioxins
and polychlorinated biphenyls (PCBs), which because of
their hydrophobicity and extreme persistence, accumulate
in the lipid reservoirs of animals consuming those plants.
Approximately 95% of human exposure to dioxins is
estimated to occur through the diet with the consumption
of fatty foods being the predominant source [8].
However, many of the pollutants arise from a wide
range of sources and are hence ubiquitous in the
environment [9]. So, releases from a point source may
result in only a negligible increase in exposure.
Table 1. Indicative list of pollutants (to air) requiring regulation
under Pollution Prevention and Control [5]
† Sulphur dioxide and other sulphur compounds
† Oxides of nitrogen and other nitrogen compounds
† Carbon monoxide
† Volatile organic compounds (including benzene, 1,3-butadiene)
† Metals and their compounds (e.g. lead, arsenic, cadmium,
nickel)
† Dust (e.g. particulate matter)
† Asbestos (suspended particulates, fibres)
† Chlorine and its compounds
† Fluorine and its compounds
† Cyanides
† Substances and preparations that have been proved to possess
carcinogenic or mutagenic properties or properties that may
affect reproduction via the air
† Polychlorinated dibenzodioxins and polychlorinated
dibenzofurans
When investigating possible effects of pollutants
emitted by a point source, it is highly desirable to have
some direct measurement of exposure. Personal or
biological monitoring may be helpful in establishing
current levels of exposure or estimating dose levels.
Personal air monitors that can be worn during normal
day-to-day activities are almost never used to measure
environmental exposure despite the fact that they can
provide actual data on individual exposure. Biomarkers
can help demonstrate that exposure has occurred and can
be used to help identify exposed populations for
investigation, e.g. urinary thioether assays can be used
as biomarkers for electrophilic chemicals such as polycyclic aromatic hydrocarbons (PAHs) [10]. Biomarkers
can also provide an estimate of past exposure provided
the pollutant under investigation has a long half life in the
body and is relatively easily detected (e.g. dioxins).
In the absence of biological measurements or personal
monitoring, exposure has to be estimated through some
other method. Typically these are:
† use of residence or proximity as an indicator of
exposure;
† environmental measurements, e.g. air monitoring;
† dispersion modelling.
Direct measurements of exposure from industry are
seldom made and often the distance from the source is
used as a proxy for exposure. This approach assumes that
exposure decreases with increasing distance from the
source. Typically a series of concentric rings are drawn
around the source to help identify potentially exposed
populations (Figure 1). Depending on the source, the
type of exposure and the population under investigation,
these rings can have varying radii but typically start in the
region of 0.5 – 1 km and may increase to 5– 10 km. Within
each concentric circle, every individual is considered to
have the same degree of exposure.
This is a simple, easy to use and cost-effective
technique. However, the sole use of proximity as an
indicator of exposure leads to much misclassification. It is
an approach that takes no account of the influence of
meteorological conditions or process characteristics such
as stack height, efflux velocity, plume temperature, etc.
Furthermore, since the zones of influence can often be
very large there is the strong possibility that the exposed
community may be exposed to pollution from other
industries or sources or may not be exposed at all. The
inclusion of people who may not be exposed may dilute
any effect that may be estimated and might result in a true
greater effect downwind of the point source being missed.
Individuals will also move within and outside of these
zones (to work, school, etc) and many people will not
reside within the zone for most of the day.
It is inevitable that there will be a large degree of
variability in terms of exposure within each of these
A. KIBBLE AND R. HARRISON: POINT SOURCES OF AIR POLLUTION 427
Figure 1. Use of concentric circles to identify potential exposure populations, in this case people living near a foundry. Acknowledgement: Crown
Copyright Ordnance Survey. An Edina Digimap/JISC supplied service.
concentric circles and this approach may be able to do no
more than distinguish between possibly exposed and
unexposed populations. Even then, such broad classifications may not be valid. Despite its flaws, it is a
commonly used technique. In a review of 45 epidemiological studies of air pollution around point and nonpoint sources, 29% determined exposure solely on
proximity measures and most used a combination of
proximity and environmental measurements [11].
It is preferable to have some direct measurement of
exposure, ideally data on pollutant concentrations at
ground level. Emission data alone can be extremely useful
in identifying the pollutants concerned, but are not a
useful surrogate for environmental measurements
especially when considering personal exposures.
In the UK, ground level pollutant monitoring takes
place at many sites that can be classified according to their
location, e.g. city centre, urban background, industrial
and rural. There are three basic ways of monitoring
pollution at ground level: automatic point monitors that
provide real time measurements, semi-automatic (or
active) samplers that involve the collection of pollutant
samples by physical or chemical means for subsequent
analyses and passive samplers such as diffusion tubes.
The UK has two automated monitoring networks. The
Automatic Urban and Rural Air Quality Monitoring
Network (AURN) can provide considerable data on
background levels of air pollution and ‘hotspot’ monitoring at urban roadsides and, occasionally, around point
sources [9]. In addition, the Ambient Hydrocarbon Air
Quality Network monitors approximately 25 VOC
species at various urban and rural locations [9].
Where such automated monitoring sites are located
near point sources, they can provide valuable data on likely
levels of exposure, but in many cases monitoring sites may
not be advantageously located or are unable to measure
specific pollutants of concern. In such cases, semiautomatic or passive samplers can be located around or
downwind of point sources. Such instruments can be
useful in highlighting ‘hotspots’ of pollution, but since they
take less frequent measurements, detailed monitoring may
be required to better quantify local concentrations.
Useful data on background exposures can also be
obtained from maps showing levels of air pollution on a
variety of geographical scales, e.g. NO2 is typically
mapped on a 1 £ 1 km spatial scale [12].
Apart from atmospheric data, environmental monitoring of soil, vegetation and water can also help identify
exposed communities. Analysis of soil and vegetation
downwind of a point source can often prove to be a good
indication of exposure. Following the release of a large
quantity of dioxin from an accident at a pesticide plant in
Seveso, Italy, the extent and level of dioxin contamination
in soil in the prevailing wind direction was used to
identify the most exposed areas [13]. Subsequent analysis
of dioxin levels in the plasma of people from these
affected areas showed that body burden was closely
correlated with levels of environmental contamination.
428 OCCUPATIONAL MEDICINE
Another approach is to use computer models to
predict exposure. The use of air dispersion models is a
long established and well-accepted methodology for
regulating emissions to atmosphere from major industries. Such models require information on the release
rates of pollutants, together with data on the height of
release and meteorological data such as temperature,
wind direction and wind speed to predict ground level
concentrations. Most models typically predict the maximum ground level concentration over the short- and
long-term around point sources including concentrations
within the nearest area of housing (Figure 2). As a result,
the exposed population can be more accurately defined
than through the use of proximity as a surrogate.
The accuracy of dispersion models is heavily dependent on the quality of the input data. However, it is
generally considered that atmospheric dispersion models
can often provide a good approximation to ground level
concentrations, particularly over the long term. The best
dispersion models have been extensively validated by
comparing their predictions against concentrations
measured downwind of pollution sources. In an ideal
situation where the release rate is well known and the
local topography uniform and fairly flat, dispersion
models will typically give predictions within a factor of
two of true concentrations [14].
It is important to obtain as much accurate information
on potential exposures as possible, as poor quality
monitoring and modelling can be very misleading [15].
Detailed exposure assessment should not be undertaken
unless the health concerns are properly defined and there
is some element of biological plausibility (i.e. that
emissions from the point source could cause that specific
condition). Measurements should be concentrated in
those areas where ill-health is alleged, focus on pollutants
emitted by the source under investigation and comply
with recognized standards [15]. Poor quality data can
weaken any study, may raise expectations in the local
community and may incur unnecessary costs.
Health studies
There is considerable epidemiological evidence to
demonstrate that, at the population level, increases in
the levels of air pollution are related to increases in
hospital admissions and mortality [1]. However, the
ability of current epidemiological methods to fully
investigate point sources is more limited. Many of the
studies examining possible health effects are retrospective
(i.e. they were triggered by complaints of apparent
‘clusters’ of ill-health in areas around point sources)
and employ routinely collected data such as cancer
registrations, birth and death records. Such observational
studies can provide evidence of association between a
health outcome and an environmental pollutant; however, they are not able to demonstrate a cause and effect
relationship. The interpretation of these findings is also
crucially dependent on well-known limitations, including
possible sources of bias and confounding, together with
the ever-present difficulty in obtaining reliable and
accurate population exposure data.
A common problem relates to the small number of
cases that may be under investigation. Most studies
around point sources examine ‘small-areas’ where the
numbers (exposed or actual cases) involved are small,
Figure 2. Dispersion modelling of sulphur dioxide concentrations around a hypothetical municipal solid waste incinerator.
A. KIBBLE AND R. HARRISON: POINT SOURCES OF AIR POLLUTION 429
thus reducing the statistical power of the study to
distinguish between chance occurrences. Very often,
even in well-designed studies, there will be a high
probability that any reported effect might be due to
chance alone. It is generally accepted that single site
studies of effects of air pollutants on health are unlikely to
have sufficient statistical power to confirm or refute
assertions of effects and there is a significant risk that the
results of such investigations will be impossible to
interpret [15].
A problem is that many studies examine industries
that operated some time in the past, particularly where
the effect under consideration is cancer, which can have
a noticeable lag period between exposure and the onset
of disease (typically many years). As a result, it may be
very difficult to obtain accurate information on past
exposure and any data that is collected may not be
relevant to more modern facilities that will have a
significantly different exposure profile. For example,
until the mid-1990s, incinerators in the UK were fitted
with rather rudimentary emission controls and, therefore, emitted quite significant amounts of air pollutants.
Newly constructed incinerator plants have to meet much
stricter controls on emissions and are significantly
cleaner [16]. Therefore, when reviewing studies of
point sources it is important to know the details of the
period of operation.
Despite such limitations there have been some
studies around point sources with many investigating
cancer and respiratory disease. In the UK, coking
works have been implicated as causes of respiratory
disease in local communities [17– 20]. Aylin et al. [17]
reported that there was a significantly higher risk of
hospital admissions for respiratory disease and asthma
among children living near a coke works at Teesside.
However, when hospital admissions from all of their
six study areas were combined, no association between
living near coke works and respiratory illnesses could
be made. Furthermore, since Teesside contains a large
number of industries it was not possible to directly link
respiratory ill-health with any particular type of
industry.
An excess of respiratory problems has also been
reported in the community living around the Monkton
coking works in South Tyneside, UK [18]. However, the
study used GP data to quantify local health problems,
which is open to interpretation bias and differences in GP
diagnostic practices, which could have influenced the
results of this study [21]. The study also attempted to link
exposure to sulphur dioxide with GP attendance on the
same day, which is not plausible as the time it takes for an
exacerbation to occur following exposure and for the
individual to be able to get an appointment at their
general practitioner surgery is very unlikely to be
encompassed within the space of a few hours.
A small excess in mortality from cardiovascular and
respiratory causes among residents near coke works has
also been reported but any effect would probably be outweighed by the effects of socio-economic factors (i.e.
deprivation) or bias resulting from inexact population or
exposure estimates [19,20].
Communities living around steel industries have also
been investigated. Findings from the US state of Utah,
where hospital admissions, symptoms and primary care
attendances fell when a local steel mill was closed only to
rise again on re-opening, suggest that local point sources
can plausibly cause health effects [22 –24]. This study
used time series methodology that is powerful in
detecting effects because a large population is followed
over time. Similarly in Holland, a panel study of adults
with respiratory conditions living near a steel mill showed
that both particulate matter (as PM10) and iron content
of the air were related to changes in lung function,
symptoms and treatment use on a day-to-day basis [25].
Both studies appear to provide good evidence of point
source pollution causing health effects in both children
and adults.
The health of the population of the coastal Australian
town of Lake Munmorah, that is located near two power
stations, has also been extensively investigated [26– 28].
Two studies consistently reported evidence of an increase
in respiratory illness, including asthma, in children in
Lake Munmorah when compared with a control population [27 – 28]. However, it is not possible to determine
whether this increase was the result of differences in air
quality and/or genetic factors between the two communities. Confounding factors such as exposure to other
pollutants, socio-economic factors, genetics and atopy
might explain all or some of these findings.
Several studies have examined possible adverse effects
on respiratory health among people living near incinerators. The most credible studies are those that
examined the respiratory health of six communities in
North Carolina, USA, three of which were exposed to
emissions from biomedical, municipal or hazardous
waste incinerators [29 – 32]. Despite detailed investigations, there were no significant differences in the
concentration of particulate pollution (PM10) or in the
overall respiratory health in those communities near the
incinerators relative to comparison communities.
It has been hypothesized that exposure to dioxins and
furans (either directly via inhalation or indirectly via the
food chain) are responsible for some cancers in communities around incinerators. Despite a number of studies,
there is not, to date, any consistent or convincing
evidence of a link with incineration [33 – 41]. In the
UK, the large epidemiological studies by Elliott et al.
[33 – 35] examined an aggregate population of 14 million
people living within 7.5 km of 72 municipal solid waste
incinerators. Despite the inclusion of incinerators with
430 OCCUPATIONAL MEDICINE
emissions of potential carcinogens orders of magnitude
higher than would occur from today’s modern incinerators, no excess of cancers could be found once socioeconomic confounding was taken into account. Given
that the emissions of dioxins and furans from modern
incinerators are orders of magnitude lower than from
older incinerators, it can be said with some confidence
that impacts on cancer rates in local people will not be
significant.
Despite the results from many of these studies being
tentative and open to question, there is clear evidence of
point sources causing ill-health. Good examples of this
are the outbreaks of asthma attacks seen in populations
living near castor bean processing factories [42 – 43] and
the epidemics seen in Barcelona on days of soya bean
unloading at the docks [44]. All these studies confirm
that local point source emissions can have an impact on
public health if the conditions for high exposure are right
and the individual is susceptible to the effects of the
exposure. It is worth noting that the affected individuals
in these outbreaks had all been recurrently exposed over a
long period thus becoming sensitized to the allergens in
bean dust and then, on a particular day, the high exposure
triggered a severe attack.
Conclusion
While it is well documented that raised ambient
concentrations of air pollution can have an effect on
susceptible people, it is very difficult to predict the likely
size of such effects from a single point source. The reality
is that current epidemiological methods cannot conclusively relate effects with specific emissions and many
studies suffer from poor estimates of exposure or small
numbers. Any effect at a local level is likely to be very
small.
References
1. Department of Health. Committee on the medical effects of
air pollutants. Quantification of the Effects of Air Pollution on
Health in the United Kingdom. London: HMSO, 1998.
2. European Environment Agency. EEA multilingual environmental glossary. Available at http://glossary.eea.eu.int/
EEAGlossary
3. Environment Agency. Report into an Air Pollution Episode:
Sulphur Dioxide, 2 September 1998, Midlands and South
Yorkshire. Bristol: Environment Agency, 2000.
4. Department of the Environment Expert Panel on Air
Quality Standards. Nitrogen Dioxide. London: The Stationary Office, 1996.
5. The Pollution Prevention and Control (England and Wales)
Regulations 2000 SI 1973. Available at www.legislation.
hmso.gov.uk
6. World Health Organization Regional Office for Europe. Air
Quality Guidelines for Europe, 2nd edn. Copenhagen: WHO
Regional Office for Europe, 2000.
7. Expert Panel on Air Quality Standards. Available at http://
www.defra.gov.uk/environment/airquality/aqs/
8. MAFF Food Safety Directorate. Dioxins in food: UK
dietary intakes. Food Surveillance Information Sheet No. 71.
London: MAFF, 1995.
9. The UK National Air Quality Information Archive.
Available at http://www.airquality.co.uk/archive/index.php
10. Ardevol E, Minguillon C, Garcia G, et al. Environmental
tobacco smoke interference in the assessment of the health
impact of a municipal incinerator on children through
urinary thioether assay. Public Health 1999;113:295 –298.
11. Huang YL, Batterman S. Residence location as a measure
of environmental exposure: a review of air pollution
epidemiology studies. J Expo Anal Environ Epidemiol
2000;10:66– 85.
12. National Atmospheric Emissions Inventory (NAEI).
Available at http://www.naei.org.uk/
13. Bertazzi PA, Consonni D, Bachetti S, et al. Health effects of
dioxin exposure: a 20-year mortality study. Am J Epidemiol
2001;153:1031– 1044.
14. Williams ML. Atmospheric dispersal of pollutants and the
modelling of air pollution. In: Harrison RM, ed. Pollution:
Causes, Effects and Control, 3rd edn. Cambridge: The Royal
Society of Chemistry 1996:221 – 240.
15. Department of Health. Investigating the Health Impact of
Emissions to Air from Local Industry. London: Department of
Health, 2000, Available at www.doh.gov.uk
16. DEFRA. Guidance on: Directive 2000/76/EC on the Incineration of Waste. Available at http://www.defra.gov.uk/
environment/ppc/wasteincin/
17. Aylin P, Bottle A, Wakefield J, Jarup L, Elliott P. Proximity
to coke works and hospital admissions for respiratory and
cardiovascular disease in England and Wales. Thorax 2001;
56:228– 233.
18. Bhopal RS, Phillimore Moffat S, Foy C. Is living near a coke
works harmful to health? A study of industrial air pollution.
J Epidemiol Community Health 1994;48:237 –247.
19. Dolk H, Thakrar B, Walls P, et al. Mortality among
residents near cokeworks in Great Britain. Occup Environ
Med 1999;56:34 – 40.
20. Dolk H, Pattenden S, Vrijheid M, Thakrar B, Armstrong B.
Perinatal and infant mortality and low birth weight among
residents near cokeworks in Great Britain. Arch Environ
Health 2000;55:26 – 30.
21. Malmberg P. Coking to death [commentary]. Lancet 1994;
344:632.
22. Pope CA. Respiratory disease associated with community
air pollution and a steel mill, Utah Valley. Am J Public
Health 1989;79:623 – 628.
23. Pope CA. Respiratory hospital admissions associated with
PM10 pollution in Utah, Salt Lake, and Cache Valleys. Arch
Environ Health 1991;45:90 –97.
24. Pope CA. Particulate pollution and health: a review of the
Utah Valley experience. J Expo Anal Environ Epidemiol
1996;6:23– 34.
25. Dusseldorp A, Kruize H, Brunekreef B, Hofschreuder P,
de Meer G, van Oudvorst AB. Associations of PM10 and
airborne iron with respiratory health of adults living near
A. KIBBLE AND R. HARRISON: POINT SOURCES OF AIR POLLUTION 431
26.
27.
28.
29.
30.
31.
32.
33.
34.
a steel factory. Am J Respir Crit Care Med 1995;152:
1932– 1939.
Halliday JA, Henry RL, Hankin RG, Hensley MJ.
Increased wheeze but not bronchial hyper-reactivity near
power stations. J Epidemiol Community Health 1993;47:
282 – 286.
Henry RL, Abramson R, Adler JE, Wlodarcyzk J, Hensley
MJ. Asthma in the vicinity of power stations: I. A prevalence
study. Pediatr Pulmonol 1991;11:127– 133.
Henry RL, Bridgman HA, Wlodarczyk J, Abramson R,
Adler JA, Hensley MJ. Asthma in the vicinity of power
stations: II. Outdoor air quality and symptoms. Pediatr
Pulmonol 1991;11:134– 140.
Shy CM, Degnan D, Fox DL, et al. Do waste incinerators
induce adverse respiratory effects? An air quality and
epidemiological study of 6 communities. Environ Health
Perspect 1995;103:714– 724.
Lee J-T, Shy CM. Respiratory function as measured by
peak expiratory flow rate and PM10: six communities
study. J Expo Anal Environ Epidemiol 1999;9:293– 299.
Mohan AK, Degnan D, Feigley CE, et al. Comparison of
respiratory symptoms among community residents near
waste disposal incinerators. Int J Environ Health Res 2000;
10:63– 75.
Hu S-W, Hazucha M, Shy CM. Waste incineration and
pulmonary function: an epidemiologic study of six communities. J Air Waste Manag Assoc 2001;51:1185– 1194.
Elliott P, Shaddick G, Kleinschmidt I, et al. Cancer
incidence near municipal solid waste incinerators in Great
Britain. Br J Cancer 1996;73:702– 710.
Elliott P, Eaton N, Shaddick G, Carter R. Cancer incidence
near municipal solid waste incinerators in Great
Britain. Part 2: histopathological and case note review of
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
primary liver cancer cases. Br J Cancer 2000;82:
1103– 1106.
Elliott P, Hills M, Beresford J, et al. Incidence of cancer of
the larynx and lung near incinerators of waste solvents and
oils in Great Britain. Lancet 1992;339:854– 858.
Michelozzi P, Fusco D, Forastiere F, Ancona C, Dell’Orco
V, Perucci CA. Small area study of mortality among people
living near multiple sources of air pollution. Occup Environ
Med 1998;55:611 – 615.
Biggeri A, Barbone F, Lagazio C, Bovenzi M, Stanta G. Air
pollution and lung cancer in Trieste, Italy: spatial analysis
of risk as a function of distance from sources. Environ
Health Perspect 1996;104:750 – 754.
Knox EG, Gilman EA. Hazard proximities of childhood
cancers. J Epidemiol Community Health 1995;51:151 – 159.
Knox EG, Gilman EA. Migration patterns of children with
cancer in Britain. J Epidemiol Community Health 1998;52:
716– 726.
Knox EG. Childhood cancers, birthplaces, incinerators and
landfill sites. Int J Epidemiol 2000;29:391 – 397.
Viel J, Arveux P, Baverel J, Cahn S. Soft-tissue sarcoma and
non-Hodgkin’s lymphoma clusters around a municipal
solid waste incinerator with high dioxin emission levels.
Am J Epidemiol 2000;152:13 –19.
Ordman D. An outbreak of bronchial asthma in South
Africa, affecting more than 200 persons, caused by castor
bean dust from an oil-processing factory. Int Arch Allergy
1955;7:10– 24.
Figley KD, Elrod RH. Endemic asthma due to castor bean
dust. J Med Assoc Am 1928;90:79 – 82.
Anto JM, Sunyer J, Rodriguez-Roisin R, Suarez-Cevera M,
Vazquez L. Community outbreaks of asthma associated
with inhalation of soybean dust. N Engl J Med 1989;320:
1097– 1102.