REVIEW

Revie

Air pollution and health

Bert Brunekreef, Stephen T Holgate

The health effects of air pollution have been subject to intense study in recent years. Exposure to pollutants such as
airborne particulate matter and ozone has been associated with increases in mortality and hospital admissions due to
respiratory and cardiovascular disease. These effects have been found in short-term studies, which relate day-to-day
variations in air pollution and health, and long-term studies, which have followed cohorts of exposed individuals over
time. Effects have been seen at very low levels of exposure, and it is unclear whether a threshold concentration
exists for particulate matter and ozone below which no effects on health are likely. In this review, we discuss the
evidence for adverse effects on health of selected air pollutants.

December, 2002, marks the 50th anniversary of the great

Department for Environment, Food and Rural Affairs
smog event in London, UK. Stagnant weather conditions
(panel).
caused a sharp increase in the concentration of air
pollutants, and over several days, more than three times
A new era of air pollution research
as many people died than expected, leading to an
20 years ago, the era of successful abatement of
estimated excess death toll of over 4000. Concentrations
traditional air pollutants culminated in a voluminous
of sulphur
review of the
dioXide and smoke reached several thousands of health effects of ambient particulates.23 At concentrations
µg per m3.1 The London 1952 smog was not without seen in the late 1970s in the developed world, adverse
precedent—similar events occurred in the Meuse valley, health effects were then regarded as unlikely. In the two
Belgium, in 1930, and elsewhere.2,3 decades since then, however, air pollution has re-
Conditions have changed; effective legislation has emerged as a major environmental health issue. One
eliminated most of the air pollution of 50 years ago. Yet reason is that, although air pollution from combustion of
the 1952 London smog event keeps attracting the traditional fossil fuel is now present in much lower
attention of contemporary air pollution scientists. One concentrations than 50 years ago, other components have
question that remains important is the extent to which air gained prominence. Photochemical air pollution,
pollution affects life expectancy. The 1954 report1 characterised by high ozone concentrations during warm
suggests that death occurred mostly in individuals who and sunny weather, was found to occur not only in
were on the brink of death already, but if this were the places like Los Angeles and Mexico City, but also in
case, the death rate should have dropped sharply after large areas of Europe. OXides of nitrogen produced by
the episode. On the contrary, the death rate remained the ever rising number of motorised vehicles have
high for several months, and a recent re-analysis of the increased until recently. Airborne particles have changed
data indicates that the number of additional deaths due to size distribution and composition, altering their toXicity.
the episode was about 12 000. 4 Another question is that
of causality; although concentrations of sulphur dioXide From episodes to time-series studies
and black smoke were greatly increased during the In the 1980s, a few air pollution episodes related to long-
episode, sulphuric acid was thought to be the critical range transport from the east occurred in western
component. On this basis, bottles of ammonia were Europe. In 1985, increases in mortality and hospital
distributed among bronchitis patients so that they could admissions in Germany and temporary lung function
neutralise acids during episodes of air pollution.5 changes in the Netherlands were seen during one such
In this review, we have focused on recent studies that episode.24,25 Because concentrations of sulphur dioXide
have answered key questions that have arisen in the past and particulate matter were in the hundreds rather than
5 years. There is no shortage of recent reviews in this thousands of µg/m3 even under such episodic conditions,
field,6–22 and we have tried to add to, rather than health effects of short episodes became hard to detect.
duplicate, these reviews. For further reading, we have Attention shifted to weather-driven, day-to-day
also added a list of relevant websites. These give access variations in air pollution over long periods of time as
to compre- hensive reviews produced by organisations determinants of day-to-day variations in mortality,
such as the World Health Organisation, the European hospital admissions, and other public-health indicators.
Union, the US Environmental Protection Agency, Such time-series studies were
and the UK

Lancet 2002; 360: 1233–42 Search strategy and selection criteria

Institute for Risk Assessment Sciences, Utrecht University, The key words "air pollution and health" produce over
PO Box 80176, 3508 TD Utrecht, Netherlands 500 references a year from Medline alone. The present
(Prof B Brunekreef PhD); and RCMB Division, School of Medicine, review is therefore, of necessity, our selection of just
Southampton General Hospital, Southampton, UK some of the major studies. The selection is based on a
(Prof S T Holgate MD) systematic Medline search until early 2002, on more than
Correspondence to: Prof Bert Brunekreef 20 years of continuous research in the field, and on
(e-mail: [email protected]) participation in many advisory boards nationally and

THE LANCET • Vol 360 • October 19, 2002 • 1

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

Useful websites containing information on air pollution and health

WHO
European Union (EU)
WHO air quality guidelines for Europe, 2000
European commission air quality website (2001)
http://www.euro.who.int/document/e71922.pdf
http://europa.eu.int/comm/environment/air/
WHO air quality guidelines for Europe, 2000 (background
Air quality framework directive (2002)
documents)
http://europa.eu.int/comm/environment/air/ambient.htm
http://www.euro.who.int/air
Clean air for Europe programme (2001)
Transport, environment, and health, 2000
http://europa.eu.int/comm/environment/air/cafe.htm
http://www.euro.who.int/document/e72015.pdf
EU Particulate Matter position paper (1997)
US Environmental Protection Agency (EPA) http://europa.eu.int/comm/environment/air/pp_pm.pdf
National ambient air quality standards (NAAQS), 2001 EU ozone position paper (1999)
http://www.epa.gov/airs/criteria.html http://europa.eu.int/comm/environment/docum/pos_paper.pdf
EPA’s national ambient air quality standards: the standard
UK Department for Environment, Food and Rural Affairs
review/re-evaluation process (1997)
Air quality—what it means for your health (2001)
http://www.epa.gov/ttn/oarpg/naaqsfin/naaqs.html
http://www.defra.gov.uk/environment/airquality/airpoll/index.htm
Health and environmental effects of ground-level ozone (1997)
Expert panel on air quality standards. Airborne particles: what
http://www.epa.gov/ttn/oarpg/naaqsfin/o3health.html
is the appropriate measurement on which to base a standard?
Health and environmental effects of particulate matter (1997)
A discusson document (2001; 110 pp)
http://www.epa.gov/ttn/oarpg/naaqsfin/pmhealth.html
http://www.defra.gov.uk/environment/airquality/aqs/air_meas
Transportation and fuels (2002)
ure/index2.htm
http://www.epa.gov/air/transport/index.html
Air quality guide for ozone (1999) Other sources of information
http://www.epa.gov/airnow/consumer.html A thematic network on air pollution and health: funded by EU,
A guide to air quality and your health gives access to a network of research projects and information
(2000) on air pollution and health.
http://www.epa.gov/airnow/aqi_cl.pdf http://airnet.iras.uu.nl
Air quality criteria for particulate matter (third external review The Health and Clean Air Newsletter is an attempt to make
draft), July 2002 scientific information available to non-specialist readers,
http://cfpub.epa.gov/ncea/cfm/partmatt.cfm?ActType=default including reporters, without sacrificing accuracy.
http://healthandcleanair.org

sporadic until about 1990,26,27 but have increased

years of observation, with more consistent effects on
exponentially since then. These studies have several
lung cancer, in addition to non-malignant
advantages: weather-driven variations in air pollution
cardiopulmonary deaths, than in the original paper.32
concentrations generate large contrasts in exposure over
Case-control studies also continue to provide evidence of
time; populations serve as their own controls; and
a link between air pollution (especially from traffic) and
studies can often use routinely collected data. As a
lung cancer.33 Recent work has suggested that effects on
result, the number of deaths or hospital admissions
life expectancy are not uniformly distributed but depend
studied can easily be in the hundreds of thousands,
on factors such as education and antioXidant vitamin
leading to great statistical power to detect small increases
status,30,32,34 which implies that life expectancy could be
in adverse health effects of air pollution. Limitations of
reduced more in disadvantaged population groups. On
this approach include lack of individual characteristics;
the basis of the cohort study findings, stringent standards
an assumption that exposure is representative for large
for fine particulate matter have been proposed in the
populations by measurements made at central
USA. Because large investments would be needed to
monitoring sites; and an assumption that confounding by
improve air quality, the findings of the US cohort studies
long-term time trends, seasonal variations, weather, and
have received intense scrutiny (see section on disputing
influenza epidemics are controlled for by sophisticated
the evidence).
statistical methods. Until recently, no European cohort studies have
provided data on life shortening. A Dutch study suggests
Life shortening due to air pollution
that exposure to traffic-related air pollution is associated
Interest in health effects of air pollution became more
with cardiorespiratory deaths in much the same way as
intense after two US cohort studies suggested that
in the USA.35
exposure to fine particulate matter in the air was
associated with life shortening. 28,29 Both studies were based
Pollutants of current interest: ozone,
on observations from the late 1970s to late 1980s, when
particulates, nitrogen dioxide
air pollution concentrations were much lower than they
had been in the past. A third cohort study (AHSMOG) Now that the concentration of sulphur dioXide has
found significant effects of particulate matter with a decreased strikingly, attention has shifted to ozone,
diameter of less than 10 µm (PM10) on non-malignant nitrogen dioXide, and particulates. Before discussing these
respiratory deaths in men and women, and on lung-cancer pollutants in more detail, some qualification is needed to
mortality in male, non-smoking Seventh-Day Adventists.30 put our discussion in a wider global perspective. For
The effect on shortening life expectancy has been millions of people living in rural areas in developing
estimated at 1–2 years for realistic exposure contrasts,31 countries, indoor pollution from the use of biomass fuels
which is substantial compared with the effects of other occurs at concentrations that are orders of magnitude
lifestyle or environmental risk factors related to mortality. higher than currently seen in the developed world. 36–38
New observations from the largest study to date show Deaths due to acute respiratory infections in children
that the findings persist after inclusion of several more resulting from these exposures are estimated to be over

2 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing
 REVIEW

2 million per year. In the largest cities of the developing admissions for asthma and chronic obstructive pulmonary
world, extreme exposures occur with both the traditional disease (COPD) among people older than 65 years were
and the modern variety of pollutants. 39,40 Ideally, lessons increased by 1·0% (0·4–1·5) per 10 µg/m3 PM10,44 and
learned in the developed world could help developing admissions for cardiovascular disease (CVD) were
countries follow a more sustainable, less polluting path increased by about 0·5% (0·2–0·8) per 10 µg/m3 PM10
to industrialisation and modernisation. However, the and by about 1·1% (0·4–1·8) per 10 µg/m3 black smoke,
available data suggest that the combined pressures of suggesting an important role for diesel exhaust.45 In
global competition and population increase leave little, if APHEA-1, daily mortality increased in six cities 46 by 2·9%
any, room to manoeuvre in this respect. for each 50 µg/m3 increase in the 1-h maximum ozone
concentration. Associations between nitrogen dio Xide and
Sources of ozone, particulate matter, and nitrogen dioxide mortality were also found, but these were sensitive to
Ozone is a strong oXidising agent formed in the adjustment for black smoke, suggesting that the nitrogen
troposphere through a complex series of reactions dioXid represented a mixture of traffic-related air
e
involving the action of sunlight on nitrogen dio Xide and pollution. The corresponding data from APHEA-2 are
hydrocarbons.41 Concentrations in city centres tend to be awaited.
lower than those in suburbs, mainly as a result of the In the USA, the National Mortality, Morbidity and Air
scavenging of ozone by nitric oXide originating from Pollution Studies (NMMAPS) focused on the 20 largest
traffic. metropolitan areas in the USA, home to 50 million
The major source of anthropogenic emissions of inhabitants, during 1987–94. All-cause mortality
nitrogen oXides into the atmosphere is the combustion of increased by 0·5% (0·1–0·9) for each 10 µg/m3 of PM10,
fossil fuels from stationary sources (heating, power a value close to the European results. 47,48 However, a
generation) and in motor vehicles.41 In ambient recent communication suggests that an unrecognised
software
conditions, nitric oXide is rapidly transformed into problem led to significant overestimation of this effect
nitrogen dioXide by atmospheric oXidants such as ozone. estimate, albeit without affecting its statistical significance
Particulate air pollution is a mixture of solid, liquid, or or sensitivity to co-pollutant adjustment; details can be
solid and liquid particles suspended in the air. The size found at http://www.healtheffects.org/Pubs/NMMAPS
of suspended particles varies, from a few nm to tens of letter.pdf (accessed July 11, 2002). Effects on hospital
µm. The largest particles (coarse fraction) are admissions were studied in ten cities with a combined
mechanically produced by attrition of larger particles. population of 1 843 000 individuals older than 65 years. 49
Small particles (<1 µm) are largely formed from gases, Effects of PM10 on COPD admissions were 1·5% (1·0–
the smallest (<0·1 µm, ultrafine) of which are formed by 1·9) and on CVD admissions 1·1% (0·9–1·3) per 10
nucleation resulting from condensation or chemical µg/m3 of PM10. There was also a weak effect of ozone
reactions that form new particles. In practical terms, a on mortality in the summer but not winter. After
distinction is made between PM10 (“thoracic” particles adjustment for ozone and PM10, none of the other
smaller than 10 µm in
diameter that can penetrate into the lower respiratory gaseous pollutants (including nitrogen dioXide) were
system), PM2·5 (“respirable” particles smaller than 2·5 associated with mortality.
µm that can penetrate into the gas-exchange region of the study covered a population of
lung), and ultrafine particles smaller than 100 nm which 38 million living in eight European cities, which were
contribute little to particle mass but which are most studied for 3–9 years in the early to mid 1990s.44 Hospital
abundant in terms of numbers and offer a very large
surface area, with increasing degrees of lung penetration.

Main findings from epidemiological studies

Short-term studies
There have been abundant studies on the short-term
effects of air pollution on health, with emphasis on
mortality and hospital admissions. Panel studies have been
done in volunteers, which have provided data on health
endpoints such as respiratory and cardiovascular
symptoms, and objective measures of lung or cardiac
function on a daily or weekly basis. Large, collaborative
efforts are under way in Europe and the USA which will
be summarised.
In Europe, the APHEA (Air Pollution and Health: a
European Approach) studies have provided many new
insights. Initial studies were based on older data
(APHEA- 1),42 and in the late 1990s a new series of
studies (APHEA-2) was done which was able to make
use of data on the PM10 fraction. The APHEA-2
mortality study covered a population of more than 43
million people living in 29 European cities, which were
all studied for more than 5 years in the early-mid
1990s.43 From data involving 21 cities, the combined
effect estimate showed that all- cause daily mortality
increased by 0·6% (95% CI 0·4–0·8) for each 10 µg/m3
increase in PM10, with some heterogeneity to be
discussed later. The APHEA-2 hospital admission

THE LANCET • Vol 360 • October 19, 2002 • 3

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

Smaller-scale studies have been done on panels of

individuals with daily or weekly observations of their
exposure and health status.50,51 These studies provide
greater insight into the role of individuals’
characteristics, but are dependent on participants’
collaboration and provide relatively small data sets.
These shortcomings make the results less reliable,
especially with respect to uncontrolled confounding by
incompletely modelled long- term time trends and
medication use. One large-scale collaborative study, the
PEACE (Pollution Effects on Asthmatic Children in
Europe) study, has been completed in 28 regions of
Europe. It failed to show effects of particulate
matter and nitrogen dio Xide on lung function and acute
symptoms.52,53 Because this was a winter-time study, no
effects of ozone were assessed. Similar, smaller- scale
studies have documented acute effects on lung function
as well as acute symptoms in children and adults. One
interpretation of these findings is that the inherent
limitations of panel studies make them less suitable to
detect subtle effects of air pollution than the large-scale
mortality and hospital admission studies.
In both short-term and long-term studies, air pollution
has an effect on cardiac deaths and hospital admissions in
addition to respiratory effects. In Augsburg, Germany,
during the 1985 air pollution episode, intriguing
epidemiological observations were made serendipitously
during a large cardiovascular study, the MONICA project.
In retrospect, plasma viscosity, as well as heart rate and
concentrations of C-reactive protein, were increased
during the episode,54–56 all of which can contribute to an
increased risk of cardiovascular events. Studies in Boston,
MA, USA, showed that nitrogen dio Xide and PM2·5 were
associated with life-threatening arrhythmia leading
to

4 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing
 REVIEW

therapeutic interventions by an implanted cardioverter intercellular adhesion molecule 1).74 These molecules
defibrillator,57 and that PM2·5 concentrations were increase neutrophil recruitment into the airways and
higher in the hours and days before onset of myocardial alveoli and activate them for mediator secretion and the
infarction in a large group of patients. 58 Cardiovascular capacity to cause tissue damage.75,76
events are now being studied in panel studies to The large interindividual differences in responsiveness
ascertain more precisely the role of various pollutants to inhaled ozone have an important genetic basis, as
and mechanisms.10 recently shown in mice77,78 and human beings.79 Candidate
genes have included those for TNFα, manganese
Long-term studies
superoXide dismutase, glutathione pero Xidase, NAD(P)
In addition to cohort studies on mortality, air pollution
quinone oXidoreductase, and glutathione S transferases.
effects on morbidity endpoints have been studied. Most of
This finding emphasises the importance of locally
these have been cross-sectional, and assume that current
available antioXidants such as uric acid, albumen,
air pollution exposure is sufficiently representative of long-
reduced glutathione, vitamin C, and vitamin E present in
term, previous exposure to make a plausible link with
the lung lining fluid and epithelial barrier 71,80 in protecting
current health status. A series of Swiss studies on the
the lung against ozone, and the protective effect exerted
association between air pollution and health among
by diets
children and adults stand out for their careful design and supplemented with antioXidants.81 An inflammatory
conduct. Although Switzerland is a small country, response induced in the lung on exposure to ozone,
exposure contrasts are relatively large because of the together with increasing neuropeptide release from
mountainous terrain. In eight different communities, sensory neurons, contributes to the acute broncho-
lung function in adults was negatively associated with constrictor response and hyper-responsiveness seen in
PM10,
nitrogen dioXide, and sulphur dioXide,59 in association asthma on exposure to this pollutant,82,83 as well as to the
with symptoms of bronchitis but not asthma. 60 In children tolerance induced by repeated short-term ozone
from ten Swiss communities, the same pollutants were exposure.83,84 The significance of this induced tolerance
found to be associated with symptoms of bronchitis but on the recognised adverse effect of this pollutant on
not asthma or allergy. 61 The associations were seen at a exacerbations of asthma during the summer months is
range of PM10 concentrations of 10–33 µg/m3 only, not known. Studies have shown wide interindividual
which is well below the concentrations in many variability in responses to air pollutants. Although
European countries. The high correlation between genetic factors undoubtedly account for part of this
pollutants prevented separation of their individual inconsistency, other more subtle factors could be
effects. operating, such as the distribution of ventilation in the
In children living in 24 US and Canadian different lung compartments, and in the case of particles,
communities, significant associations were reported differences in regional deposition within the lungs.
between exposure to fine particles and their acidity and By contrast with ozone, little is known about the effects
lung function, symptoms of bronchitis, but again not
asthma.62–64
EXposure to particles has now also been related of nitrogen dioXide on normal and diseased lung. In-vitro
prospectively to reduced lung function growth in studies in animals and human beings confirm the
children,65 and even children relocating from high to low capacity of nitrogen dioXide to activate oXidant
pollution areas (or vice versa) were shown to experience pathways, albeit less potently than ozone. 72,85 The ensuing
changes in lung function growth that mirrored changes in inflammatory response also differs by enhancing the
exposure to particulate matter.66 recruitment of T lymphocytes and macrophages.86
Relation of variations in asthma prevalence to air However, one feature of nitrogen dioXide that might
pollution has been difficult;67,68 however, prospective contribute to exacerbations of respiratory disease is its
studies in California have recently suggested that some capacity to impair the function of alveolar macrophages
incident asthma cases69,70 could be related to ozone but and epithelial cells, thereby increasing the risk of lung
not infection.87,88 Although nitrogen
other pollutants. dioXide can also enhance airway responses to inhaled

Mechanisms proinflammatory genes that code for cytokines (eg,

Chamber studies provide a method by which to pursue granulocyte- macrophage colony-stimulating factor,
the acute mechanisms of individual air pollutants, but do tumour necrosis factor α [TNFα], and interleukin 1β),
not reproduce either the mixtures or temporal variation chemokines that attract neutrophils (eg, interleukin 8,
that occur in natural exposures. Although individual air neutrophil activating protein 78, and Gro α), and
pollutants can exert their own specific individual to Xic adhesion molecules (eg,
effects on the respiratory and cardiovascular systems,
ozone, oXides of nitrogen, and suspended particulates all
share a common property of being potent oXidants,
either through direct effects on lipids and proteins or
indirectly through the activation intracellular oXidant
pathways.71
Animal and human in-vitro and in-vivo exposure
studies have demonstrated the powerful o Xidant capacity
of inhaled ozone with activation of stress signalling
pathways in epithelial cells72 and resident alveolar
inflammatory cells.73 This mechanism involves activation
of the transcription factor nuclear factor (NF) nB and its
translocation to the nucleus. There it binds to DNA
consensus sequences in the promoters of

THE LANCET • Vol 360 • October 19, 2002 • 5

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

allergens in asthmatic individuals, 89 short-term exposure

remains relatively benign. Little is known about the
long- term effects of nitrogen dioXide in human beings,
but in rodents, prolonged exposure to either ozone or
nitrogen dioXide results in destruction of peripheral
airways.
The increase in respiratory and cardiovascular
morbidity and mortality that has been epidemiologically
linked to inhaled particulates has, until recently, defied
a mechanistic explanation. In-vitro and in-vivo studies
in animals and human beings have revealed potent
proinflammatory effects involving lung epithelial cells 90
and alveolar macrophages. 91 Both directly and through
uptake into epithelial cells 92 and macrophages,93,94
oXidant pathways are activated95,96 with the downstream
consequences of stimulating cytokine and mediator
release,97 resulting in extensive neutrophil migration,
but also T lymphocyte recruitment and activation.98,99 On
reaching the bone marrow, cytokines and chemokines
released from the lung stimulate egression of
neutrophils and their precursors into the circulation.100 In
the short term there is acute tissue damage with
activation of the epidermal-growth-factor receptor
(EGFR) pathway, and evidence for organ-repair
responses.101 This reaction is partly due to surface
processing of EGFR ligands such as heparin-binding
EGF-like growth factor (HB-EGF) and

6 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing
 REVIEW

oXidant-induced transactivation of the EGFR. 102 If this

cycle of damage and repair continues, epithelial mucus Maximum concentration allowed when averaged
over time*
metaplasia results, as does ongoing cytokine and
chemokine secretion that contributes to airway 1h 8h 24 h 1 year
inflammation. Ozone (µg/m )
3

Although knowledge of the respiratory actions of

particulates has grown, an understanding of how WHO ·· 120 ·· ··
particulates increase the risk of cardiovascular events has EPA 235 157 ·· ··
(proposed)
proven much more difficult to achieve. Again, through EU ·· 120 (2010) ·· ··
activation of stress signalling pathways from the epithelium
to the lung microvessels, factors that influence blood Nitrogen dioxide (µg/m )
3

WHO 200 ·· ·· 40
clotting are generated. Increased concentrations of EPA ·· ·· ·· 100
fibrinogen and platelets, and sequestration of red blood EU 200 ·· ·· 40 (2010)
(2010)
cells in the lung mass103 have also been detected in relation
to particulate pollution, but, along with increasing the A key question is whether threshold concentrations exist
risk of cardiac arrhythmia,10 their significance in the below which air pollution has no effect on population
cardiovascular events linked to particle pollution remains to health. If such a threshold could be identified, no
be established.11 additional
Diesel particulates and ozone have been shown to
increase the synthesis of the allergic antibody IgE in
animals104 and human beings,105 which would increase
sensitisation to common allergens.106 By interacting
together and with other environmental factors,
particulates and gaseous air pollutants can have long-
term effects on allergic individuals. Although pollutants
are unlikely to be able to interact to enhance proallergic
and inflammatory responses, no studies have investigated
mixtures. Similarly, long-term exposure to pollutant
mixtures can have tissue- damaging effects which could
be irreversible.13 The recent recognition that ultrafine
particles (mass median diameter
<0·1 µm) are more toXic when inhaled than PM10 suggests
that their ability to be absorbed into tissues and the
circulation, and their greatly increased surface area,
might be important factors in determining
cardiopulmonary toXicity.107

Air quality guidelines and standards

Several guidelines and standards exist for ozone,
nitrogen dioXide, and particulate matter in ambient air.
The table lists the most recent air quality guidelines and
standards recommended by WHO, the US Environmental
Protection Agency, and the European Union (EU). The
EU standards are targets to be reached in 2005 or 2010.
The most remarkable difference lies in the annual value
for nitrogen dioXide. The WHO and EU value is only
40% of the US value.
WHO has not proposed guidelines for particulate matter,
arguing that it was unable to define a threshold below
which no adverse effects are expected. Instead, dose-
response information was provided to help policy makers
decide when setting a standard. The effect estimates
given in the table were based on information as it was
available until 1996. As described earlier, the newest
large-scale studies43,48 tend to show somewhat smaller
effects per unit particulate matter. For particulate matter,
the proposed US standards for PM2·5 are not very
different from the EU 2010 annual average and the 2005
24-h average for PM10, considering that PM2·5 usually
comprises about 60–70% of the PM10 concentration, and
considering the number of exceedances allowed in the
24-h EU standard.
All guidelines and standards mentioned in the table are
subject to periodic revision when new scientific
information becomes available. WHO has just recently
started a process to re-evaluate the guidelines for these
three pollutants.

Current issues
Thresholds

THE LANCET • Vol 360 • October 19, 2002 • 7

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

PM10 (µg/m3)
WHO (mortality relative ·· ·· 1·007
1·10 risk per 10 µg/m3)†
EPA ·· ·· 150 50
EU ·· ·· 50‡ (2005), 40 (2005),
50§ (2010) 20 (2010)
PM2·5 (µg/m3)
WHO (mortality relative ·· ·· 1·015
1·14 risk per 10 µg/m3)†
EPA ·· ·· 65 15
(proposed) (proposed)
For details see http://www.euro.who.int/document/e71922.pdf,
http://www.epa.gov/airs/criteria.html, and http://www.irceline.be. *Short
averaging times are used when the guideline was developed to prevent acute
effects, long averaging times to prevent long-term effects. †No guideline
value for particulate matter was given because no threshold concentration
was identified below which no effects on health were expected. Relative risk
estimates were provided to help policy makers set standards based on
quantitative dose-response information. ‡To be exceeded on no more than
35 days per year. §To be exceeded on no more than 7 days per year.
WHO, US Environmental Protection Agency (EPA), and
European Union (EU) air quality guidelines and standards
for ozone, nitrogen dioxide, and particulate matter

public-health benefits would be expected from bringing

air pollution concentrations far below this level.
Theoretical and empirical work has been done to shed
light on this issue.108,109 In an analysis of NMMAPS data,
no evidence was found for a threshold for PM10 and
daily all-cause and cardiorespiratory mortality. 109 By
contrast, a threshold of about 50 µg/m3 was estimated
for non-cardiorespiratory causes of death, illustrating
the specificity of the approach. Earlier analyses
restricted the analysis to concentrations below a certain
value.37,110 These analyses suggest that a threshold for
acute effects of ozone on lung function changes must lie
well below 100 µg/m3 as an hourly maximum.

Displacement of daily mortality and hospital admissions

Although time-series studies have shown a link
between day-to-day variations in air pollution
concentrations and daily deaths and hospital
admissions, by how many days, weeks, or months
such events are brought forward is unclear. If deaths
occur just a few days earlier than they would have
occurred anyway, the public-health significance of these
associations would be much less than if life
expectancy is being reduced by months or years.111
Recent analyses show that, in the time-series studies,
there is no evidence that deaths or hospital
admissions are being brought forward by just a few
days.112–115 On the contrary, effect estimates increase
with increasing duration of exposure to air pollution,
suggesting that cumulative exposures have stronger
effects on mortality than can be extracted from
associations between day-to-day variations in air
pollution and deaths. Previous data have shown that
many of the deaths associated with air pollution
were occurring outside hospital, which also supports
the suggestion that these patients were often not
terminally ill.116

8 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing
 REVIEW

Relation between ambient and personal exposure to health endpoints.123 However, the number of ultrafine
particulate matter, nitrogen dioxide, and ozone in short- particles in the air is often poorly correlated with PM2·5,
term and long-term studies so ultrafine particles are unlikely to explain much of the
In developed countries, people spend more than 80% of
association between particulate matter mass and health
their time indoors, and most of that time in their own endpoints.124
home. Indoor pollutant concentrations can differ from Specific components related to traffic exhaust,
outdoor concentrations because of ventilation patterns especially diesel combustion products, could be
and indoor sources. Therefore whether measurement of important.125 The APHEA and NMMAPS analyses have
air pollution outdoors is a valid method of assessing suggested that particulate matter effects are larger in areas
exposure of the population has been questioned. The with high nitrogen dioXide (ie, traffic density)43 or in areas
day- to-day variation in personal exposure to particulate with high emissions of particulate matter from highway
matter correlates with the day-to-day variation in vehicles and diesel locomotives.126 An intriguing
ambient concentrations of particulate matter; observation from a time-series study is that individuals
correlations are especially high in the case of fine particles who live on the main roads of Amsterdam have much
(PM2·5).117 One study that addressed this issue higher relative risks of death than people who live away
simultaneously for particulate matter and gaseous from the main roads, when analysed with data from the
components found that there was no association between same background air pollution monitoring station. 127 This
the day-to-day variation in personal exposure to nitrogen finding clearly suggests that traffic-related particulate
dioXide, sulphur dioXide, and ozone and ambient matter is involved in explaining increases in mortality on
measurements of these three
components.118 Ambient PM2·5, nitrogen dioXide, high pollution days.
sulphur dioXide, and ozone were closely associated with Factor analysis suggested that particles from mobile and
personal PM2·5, which strongly suggests that gaseous and coal combustion sources, but not coarse particles
PM2·5 concentrations outdoors act as a surrogate for originating from the earth’s crust, are responsible for
personal exposure to PM2·5.118 effects on mortality.128 Studies from the Utah Valley,
Few studies have addressed the question of whether USA, have linked transition metal content of particulate
spatial variations in long-term average outdoor matter with particulate matter toXicity in human
concentrations are reflected in similar variations in long- bronchial instillation studies.129
term personal exposures. For PM2·5 and nitrogen Much attention has been devoted to separating the
dioXide regional and local-scale spatial variations in effects of fine particulate matter (PM2·5) and coarse
,
outdoor concentrations have been shown to be reflected mass. Initial observations that fine particulate matter was
in similar variations in personal exposures. 119,120 The US associated with mortality whereas coarse mass was not 130
AHSMOG study and the recent Dutch cohort study have have been corroborated in several recent studies, 131,132 but
used estimates of small-scale spatial variations in air others have found independent effects of coarse mass on
pollution exposure to their advantage, 30,121 suggesting that hospital admissions for asthma,133 and one study was
further improvements can be expected when such unable to separate effects of fine particulate matter from
within- community differences in exposure are taken those of coarse mass.134
into account. Support for causality also comes from studies on
effects of reducing air pollution on health outcomes. One
Causality of the best examples is a labour dispute that shut down a
Knowledge of which pollution components are large steel mill in the Utah Valley for 14 months in
responsible for any health effects observed in 1987. Ambient particulate matter concentrations as well
epidemiological studies is of obvious importance. For as respiratory hospital admissions were clearly decreased
ozone, the situation is relatively simple. A large during the strike, only to increase to prestrike levels after
experimental database exists to document that ozone has the dispute ended.135 Mortality was decreased as well.136
In
significant biological effects at ambient concentrations. combination with the recent toXicological studies of
In
addition, the correlation between ozone and other particulate matter collected before, during, and after the
pollutant concentrations in outdoor air is often low, so strike, the Utah Valley example provides strong
the effects of ozone and other pollutants can be separated evidence of a causal relation between exposure to
relatively easily. ambient particulate matter and mortality and morbidity.
For nitrogen dioXide, the situation is more complex. Other examples include reductions in acute-care visits
Evidence for biological effects at ambient concentrations and hospital admissions for asthma in Atlanta, GA,
is much weaker than for ozone. In outdoor air, nitrogen USA, in conjunction with reduced air pollution due to
dioXide is often highly correlated with other combustion traffic measures taken during the 2000 Olympic
products, notably fine particulate matter, and to personal games,137 and reductions in bronchitis in association
with reduced air
PM2·5, but not personal nitrogen dioXide.118 Our pollution over several years in the former German
assessment is that, in most circumstances, nitrogen Democratic Republic.138
dioXide serves as a surrogate for all traffic-related
combustion products. changes which may affect blood coagulability. Similarly,
Most complex is the question of which particulate some suggestive evidence from epidemiological studies
matter components or attributes are most important in points to ultrafine particles being related to respiratory
determining health effects. Many candidates have been
proposed. One is ultrafine particles—ie, particles of less
than 100 nm which can be found in very high numbers
(>100 000 cm3) near busy roads.122 The thinking is that
such particles can easily find their way from the lungs into
the bloodstream and then lead to systemic inflammatory

THE LANCET • Vol 360 • October 19, 2002 • 9

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

Disputing the evidence

In view of the potentially large costs and benefits
associated with abatement of air pollution, questions
surrounding the relation between air pollution and
health have been an area of fierce debate in the past
decade. The early time-series studies have been
criticised for their analytical approach and inadequate
control for confounding by weather variables etc,
whereas the US cohort studies have been criticised for
inadequate confounder and co-pollutant control.
Reanalyses of such studies were done to investigate
whether findings depended more on analytical
approach than on robust

10 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing
 REVIEW

relations. The USA-based Health Effects Institute, itself

6 Goldsmith CA, Kobzik L. Particulate air pollution and asthma: a
a partnership between (automobile) industry and review of epidemiological and biological studies. Rev Environ Health
government, has taken a leadership role in this debate by 1999; 14: 121–34.
commissioning two reanalysis projects—one to 7 Koenig JQ. Air pollution and asthma. J Allergy Clin Immunol 1999;
reanalyse the Philadelphia time-series study,139 which 104: 717–22.
had generated much of the debate on short-term effects, 8 Nicolai T. Air pollution and respiratory disease in children: what is
the clinically relevant impact? Pediatr Pulmonol Suppl 1999; 18: 9–13.
and another to reanalyse the two main US cohort studies. 9 Pope CA III. Mortality and air pollution: associations persist with
These reanalyses were done by independent researchers continued advances in research methodology. Environ Health Perspect
who were granted access to all original data and largely 1999; 107: 613–14.
confirmed the findings of the original studies.140,141 In 10 Donaldson K, Gilmour MI, MacNee W. Asthma and PM10.
addition, the reanalysis revealed new insights into the Respir Res 2000; 1: 12–15.
11 Gordon T, Reibman J. Cardiovascular toXicity of inhaled
role of weather variables in time-series studies, 142,143 and ambient particulate matter. Toxicol Sci 2000; 56: 2–4.
new methods for enabling analysis of spatial association 12 Nyberg BF, Pershagen G. Epidemiologic studies on the health effects
between air pollution, mortality, and potential of ambient particulate air pollution. Scand J Work Environ Health
confounding variables.144 The reanalyses themselves have 2000; 26 (suppl 1): 49–89.
served as role models on how to provide policy makers 13 Pope CA 3rd. Epidemiol fine particulate air Pollut Hum health:
biologic mechanisms and who’s at risk? Environ Health Perspect
with the best possible scientific evidence.145 2000; 108 (suppl 4): 713–23.
14 Takafuji S, Nakagawa T. Air pollution and allergy. J Investig Allergol
Concluding remarks Clin Immunol 2000; 10: 5–10.
An excess risk of death of “0·5% per 10 µg/m3 PM10” 15 Dockery DW. Epidemiologic evidence of cardiovascular effects of
requires some translation before the effect on public particulate air pollution. Environ Health Perspect 2001; 109 (suppl 4):
health becomes clear. For the Netherlands (16 million 483–86.
16 Donaldson K, Stone V, Seaton A, MacNee W. Ambient particle
inhabitants, about 140 000 deaths per year, and an inhalation and the cardiovascular system: potential mechanisms.
average PM10 concentration of >30 µg/m3), the number Environ Health Perspect 2001; 109 (suppl 4): 523–27.
of deaths attributable to day-to-day variations in PM10 17 Gilmour MI, Daniels M, McCrillis RC, Winsett D, Selgrade MK.
would translate into at least 2100 deaths brought Air pollutant-enhanced respiratory disease in experimental animals.
Environ Health Perspect 2001; 109 (suppl 4): 619–22.
forward by air pollution per year—almost twice the
18 Oberdorster G. Pulmonary effects of inhaled ultrafine particles.
number of deaths due to traffic accidents. Estimates Int Arch Occup Environ Health 2001; 74: 1–8.
derived from the cohort studies are much higher still 19 Peden DB. Air pollution in asthma: effect of pollutants on airway
because these incorporate long-term as well as short- inflammation. Ann Allergy Asthma Immunol 2001; 87: 12–17.
term effects. For Austria, France, and Switzerland 20 Sunyer J. Urban air pollution and chronic obstructive pulmonary
combined (population about 74·5 million), 40 000 disease: a review. Eur Respir J 2001; 17: 1024–33.
21 Sydbom A, Blomberg A, Parnia S, et al. Health effects of diesel
deaths per year are estimated to be attributable to air exhaust emissions. Eur Respir J 2001; 17: 733–46.
pollution, about half to air pollution from traffic 22 Thurston GD, Ito K. Epidemiological studies of acute ozone
specifically.146 Similarly high numbers have been exposures and mortality. J Expo Anal Environ Epidemiol 2001; 11:
estimated for respiratory and cardiovascular hospital 286–94.
admissions, bronchitis episodes, and restricted activity 23 Holland WW, Bennett AE, Cameron IR, et al. Health effects of
particulate pollution: reappraising the evidence. Am J Epidemiol
days. Because of such numbers, health effects from air 1979; 110: 527–659.
pollution have been estimated to be higher than effects 24 Wichmann HE, Mueller W, Allhoff P, et al. Health effects during a
from a long list of other environmental factors. 147 These smog episode in West Germany in 1985. Environ Health Perspect
estimates are based on three major assumptions: 1989; 79: 89–99.
causality of the epidemiological associations, linearity of 25 Dassen W, Brunekreef B, Hoek G, et al. Decline in children’s
pulmonary function during an air pollution episode. J Air Pollut
the exposure-response relations, and that a threshold is Control Assoc 1986; 36: 1223–27.
absent or has a very low value. All of these assumptions 26 Ostro B. A search for a threshold in the relationship of air pollution to
are being tested rigorously; further proof could arise mortality: a reanalysis of data on London winters. Environ Health
from research specifically targeted at assessment of Perspect 1984; 58: 397–99.
changes in air pollution concentrations, such as those 27 Schwartz J, Marcus A. Mortality and air pollution in London: a time
series analysis. Am J Epidemiol 1990; 131: 185–94.
being done in a series of studies in the former German
28 Dockery DW, Pope CA 3rd, Xu X, et al. An association between air
Democratic Republic.138 pollution and mortality in six US cities. N Engl J Med 1993; 329:
Given the high cost of further measures to reduce air 1753–59.
pollution, and the many new findings which suggest that 29 Pope CA 3rd, Thun MJ, Namboodiri MM, et al. Particulate air
health effects can be seen at ever lower concentrations, pollution as a predictor of mortality in a prospective study of US
the health effects of air pollution will need to receive adults. Am J Respir Crit Care Med 1995; 151: 669–74.
30 Abbey DE, Nishino N, McDonnell WF, et al. Long-term inhalable
much scientific and regulatory interest for years to come. particles and other air pollutants related to mortality in nonsmokers.
Am J Respir Crit Care Med 1999; 159: 373–82.
Conflict of interest statement 31 Brunekreef B. Air pollution and life expectancy: is there a relation?
None declared. Occup Environ Med 1997; 54: 781–84.
32 Pope CA 3rd, Burnett RT, Thun MJ, et al. Lung cancer,
cardiopulmonary mortality, and long-term exposure to
References fine particulate air pollution. JAMA 2002; 287: 1132–41.
1 Ministry of Health. Mortality and morbidity during the London 33 Nyberg F, Gustavsson P, Jarup L, et al. Urban air pollution and lung
fog of December 1952. Reports on Public Health and Medical cancer in Stockholm. Epidemiology 2000; 11: 487–95.
Subjects No 95. London: HMSO, 1954. 34 Brunekreef B. All but quiet on the particulate front. Am J Respir
2 Nemery B, Hoet PH, Nemmar A. The Meuse Valley fog of 1930: an Crit Care Med 1999; 159: 354–56.
air pollution disaster. Lancet 2001; 357: 704–08. 35 Hoek G, Brunekreef B, Goldbohm S, Fischer P, Brandt P.
3 Firket J. Fog along the Meuse Valley. Trans Faraday Soc 1936; 32: Association between mortality and indicators of traffic-related air
1192–97. pollution in the Netherlands: a cohort study. Lancet (in press).
4 Bell ML, Davis DL. Reassessment of the lethal London fog of 1952: 36 McMichael AJ, Smith KR. Seeking a global perspective on air
novel indicators of acute and chronic consequences of acute exposure pollution and health. Epidemiology 1999; 10: 1–4.
to air pollution. Environ Health Perspect 2001; 109 (suppl 3): 389–94. 37 Smith KR, Samet JM, Romieu I, Bruce N. Indoor air pollution
5 Anon. An anti-smog bottle. BMJ 1955; 231: 1135. in developing countries and acute lower respiratory infections in
children. Thorax 2000; 55: 518–32.

THE LANCET • Vol 360 • October 19, 2002 • 11

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

38 Ezzati M, Kammen D. Indoor air pollution from biomass

combustion and acute respiratory infections in Kenya: an exposure- 64 Spengler JD, Koutrakis P, Dockery DW, Raizenne M, Speizer FE.
response study. Lancet 2001; 358: 619–24. Health effects of acid aerosols on North American children: air
39 Vichit-Vadakan N, Ostro BD, Chestnut LG, et al. Air pollution and pollution exposures. Environ Health Perspect 1996; 104: 492–99.
respiratory symptoms: results from three panel studies in Bangkok, 65 Gauderman WJ, McConnell R, Gilliland F, et al. Association
Thailand. Environ Health Perspect 2001; 109 (suppl 3): 381–87. between air pollution and lung function growth in southern California
children. Am J Respir Crit Care Med 2000; 162: 1383–90.
40 Ostro B, Chestnut L, Vichit-Vadakan N, Laixuthai A. The impact of
particulate matter on daily mortality in Bangkok, Thailand. J Air 66 Avol EL, Gauderman WJ, Tan SM, London SJ, Peters JM.
Waste Manag Assoc 1999; 49: 100–07. Respiratory effects of relocating to areas of differing air pollution
levels. Am J Respir Crit Care Med 2001; 164: 2067–72.
41 WHO. Air Quality Guidelines for Europe, 2nd edn. WHO Reg Publ
Eur Ser 2000; 91: 1–287. 67 Kaur B, Anderson HR, Austin J, et al. Prevalence of asthma
symptoms, diagnosis, and treatment in 12–14 year old children
42 Katsouyanni K, Zmirou D, Spix C, et al. Short-term effects of air
across Great Britain (international study of asthma and
pollution on health: a European approach using epidemiological time-
allergies in childhood, ISAAC UK). BMJ 1998; 316: 118–24.
series data. The APHEA project: background, objectives, design.
68 The International Study of Asthma and Allergies in Childhood
Eur Respir J 1995; 8: 1030–38.
(ISAAC) Steering Committee. Worldwide variation in prevalence of
43 Katsouyanni K, Touloumi G, Samoli E, et al. Confounding and symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema:
effect modification in the short-term effects of ambient particles on ISAAC. Lancet 1998; 351: 1225–32.
total mortality: results from 29 European cities within the APHEA2
69 McDonnell WF, Abbey DE, Nishino N, Lebowitz MD. Long-
project. Epidemiology 2001; 12: 521–31.
term ambient ozone concentration and the incidence of asthma in
44 Atkinson RW, Anderson HR, Sunyer J, et al. Acute effects of nonsmoking adults: the AHSMOG Study. Environ Res 1999; 80:
particulate air pollution on respiratory admissions: results from 110–21.
APHEA 2 project. Air Pollution and Health: a European Approach. 70 McConnell R, Berhane K, Gilliland F, et al. Asthma in
Am J Respir Crit Care Med 2001; 164: 1860–66. exercising children exposed to ozone: a cohort study. Lancet
45 Le Tertre A, Medina S, Samoli E, et al. Short term effects of 2002; 359: 386–91.
particulate air pollution on cardiovascular diseases in eight European 71 Rahman I, MacNee W. OXidative stress and regulation of glutathione
cities. J Epidemiol Community Health (in press). in lung inflammation. Eur Respir J 2000; 16: 534–54.
46 Touloumi G, Katsouyanni K, Zmirou D, et al. Short-term effects of 72 Bayram H, Sapsford RJ, Abdelaziz MM, Khair OA. Effect of ozone
ambient oXidant exposure on mortality: a combined analysis within and nitrogen dioXide on the release of proinflammatory mediators
the APHEA project. Air Pollution and Health: a European from bronchial epithelial cells of nonatopic nonasthmatic subjects
Approach. Am J Epidemiol 1997; 146: 177–85. and atopic asthmatic patients in vitro. J Allergy Clin Immunol 2001;
47 Samet JM, Dominici F, Zeger SL, Schwartz J. Dockery DW. The 107: 287–94.
National Morbidity, Mortality, and Air Pollution Study. Part I: 73 Mochitate K, Katagiri K, Miura T. Impairment of microbial billing
methods and methodologic issues. Res Rep Health Eff Inst 2000; 75: and superoXide-producing activities of alveolar macrophages by a low
5–14. level of ozone. J Health Sci 2001; 47: 302–09.
48 Samet JM, Dominici F, Curriero FC, Coursac I, Zeger SL. Fine 74 Nichols BG, Woods JS, Luchtel DL, Corral J, Koenig JQ. Effects of
particulate air pollution and mortality in 20 US cities, 1987- ozone exposure on nuclear factor-kappaB activation and tumor
1994. N Engl J Med 2000; 343: 1742–49. necrosis factor-alpha expression in human nasal epithelial cells.
49 Zanobetti A, Schwartz J, Dockery DW. Airborne particles are a risk Toxicol Sci 2001; 60: 356–62.
factor for hospital admissions for heart and lung disease. Environ 75 Bassett D, Elbon-Copp C, Otterbein S, et al. Inflammatory cell
Health Perspect 2000; 108: 1071–77. availability affects ozone-induced lung damage. J Toxicol Environ
50 Pope CA 3rd, Dockery DW, Spengler JD, Raizenne ME. Respiratory Health A 2001; 64: 547–65.
health and PM10 pollution: a daily time series analysis. Am Rev 76 Jorres R, Nowak D, Magnussen H. The effect of ozone exposure on
Respir Dis 1991; 144: 668–74. allergen responsiveness in subjects with asthma or rhinitis.
51 Pope CA. 3rd, Dockery DW. Acute health effects of PM10 pollution Am J Respir Crit Care Med 1996; 153: 56–64.
on symptomatic and asymptomatic children. Am Rev Respir Dis 77 Kleeberger SR, Levitt RC, Zhang LY, et al. Linkage analysis of
1992; 145: 1123–28. susceptibility to ozone-induced lung inflammation in inbred mice.
52 Roemer W, Hoek G, Brunekreef B, et al. Daily variations in air Nat Genet 1997; 17: 475–78.
pollution and respiratory health in a multicentre study: the 78 Cho HY, Zhang LY, Kleeberger SR. Ozone-induced lung
PEACE project. Pollution Effects on Asthmatic Children in inflammation and hyperreactivity are mediated via tumor necrosis
Europe. factor-alpha receptors. Am J Physiol Lung Cell Mol Physiol 2001;
Eur Respir J 1998; 12: 1354–61. 280: L537–46.
53 Roemer W, Hoek G, Brunekreef B. Pollution effects on asthmatic 79 Bergamaschi E, Palma G De, Mozzoni P, et al. Polymorphism of
children in Europe, the PEACE study. Clin Exp Allergy 2000; 30: quinone-metabolizing enzymes and susceptibility to ozone-induced
1067–75. acute effects. Am J Respir Crit Care Med 2001; 163: 1426–31.
54 Peters A, Doring A, Wichmann HE, Koenig W. Increased plasma 80 Mudway IS, Krishna MT, Frew AJ, et al. Compromised
viscosity during an air pollution episode: a link to mortality? Lancet concentrations of ascorbate in fluid lining the respiratory tract in
1997; 349: 1582–87. human subjects after exposure to ozone. Occup Environ Med
55 Peters A, Frohlich M, Doring A, et al. Particulate air pollution 1999; 56: 473–81.
is associated with an acute phase response in men; results from 81 Samet JM, Hatch GE, Horstman D, et al. Effect of antioXidant
the MONICA-Augsburg Study. Eur Heart J 2001; 22: 1198– supplementation on ozone-induced lung injury in human subjects.
204. Am J Respir Crit Care Med 2001; 164: 819–25.
56 Peters A, Perz S, Doring A, et al. Increases in heart rate during an air 82 Holz O, Jorres RA, Timm P, et al. Ozone-induced airway
pollution episode. Am J Epidemiol 1999; 150: 1094–98. inflammatory changes differ between individuals and are
reproducible. Am J Respir Crit Care Med 1999; 159: 776–84.
57 Peters A, Liu E, Verrier RL, et al. Air pollution and incidence of
cardiac arrhythmia. Epidemiology 2000; 11: 11–17. 83 Schelegle ES, Eldridge MW, Cross CE, Walby WF, Adams WC.
Differential effects of airway anesthesia on ozone-induced pulmonary
58 Peters A, Dockery DW, Muller JE, Mittleman MA. Increased
responses in human subjects. Am J Respir Crit Care Med 2001; 163:
particulate air pollution and the triggering of myocardial infarction. 1121–27.
Circulation 2001; 103: 2810–15.
84 Jorres RA, Holtz O, Zachgo P, et al. The effect of repeated ozone
59 Ackermann-Liebrich U, Leuenberger P, Schwartz J, et al. Lung exposures on inflammatory markers in bronchoalveolar lavage
function and long term exposure to air pollutants in Switzerland. fluid and mucosal biopsies. Am J Respir Crit Care Med 2000; 161:
Study on Air Pollution and Lung Diseases in Adults (SAPALDIA) 1855–61.
Team. Am J Respir Crit Care Med 1997; 155: 122–29.
85 Blomberg A, Krishna MT, Bocchino V, et al. The inflammatory
60 Zemp E, Elsasser S, Schindler C, et al. Long-term ambient air effects of 2 ppm NO2 on the airways of healthy subjects. Am J
pollution and respiratory symptoms in adults (SAPALDIA study). Respir Crit Care Med 1997; 156: 418–24.
Am J Respir Crit Care Med 1999; 159: 1257–66. 86 Sandstrom T, Helleday R, Bjermer L, Stjernberg N. Effects of
61 Braun-Fahrlander C, Vuille JC, Sennhauser FH, et al. Respiratory repeated exposure to 4 ppm nitrogen dioXide on bronchoalveolar
health and long-term exposure to air pollutants in Swiss lymphocyte subsets and macrophages in healthy men. Eur Respir
schoolchildren. Am J Respir Crit Care Med 1997; 155: 1042–49. J 1992; 5: 1092–96.
62 Dockery DW, Cunningham J, Damokosh AI, et al. Health effects of 87 Frampton MW, Smeglin AM, Roberts NJ Jr, Finkelstein JN,
acid aerosols on North American children: respiratory symptoms. Morrow PE, Utell MJ. Nitrogen dioXide exposure in vivo and
Environ Health Perspect 1996; 104: 500–05. human alveolar macrophage inactivation of influenza virus in vitro.
63 Raizenne M, Neas LM, Damokosh AI, et al. Health effects of acid Environ Res 1989; 48: 179–92.
aerosols on North American children: pulmonary function. Environ
Health Perspect 1996; 104: 506–14.

12 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing
 REVIEW

88 Chauhan AJ, Krishna MT, Frew AJ, Holgate ST. EXposure to

nitrogen dioXide (NO2) and respiratory disease risk. Rev 113 Schwartz J. Is there harvesting in the association of airborne particles
Environ Health 1998; 13: 73–90. with daily deaths and hospital admissions? Epidemiology 2001; 12:
89 Tunnicliffe WS. Burge PS, Ayres JG. Effect of domestic 55–61.
concentrations of nitrogen dioXide on airway responses to 114 Zanobetti A, Schwartz J, Samoli E, et al. The temporal pattern of
inhaled allergen in asthmatic patients. Lancet 1994; 344: mortality responses to air pollution: a multicity assessment of
1733–36. mortality displacement. Epidemiology 2002; 13: 87–93.
90 Fujii T, Hayashi S, Hogg JC, Vincent R, Van Eeden SF. Particulate 115 Zeger SL, Dominici F, Samet J. Harvesting-resistant estimates of
matter induces cytokine expression in human bronchial epithelial air pollution effects on mortality. Epidemiology 1999; 10: 171–75.
cells. Am J Respir Cell Mol Biol 2001; 25: 265–71. 116 Schwartz J. What are people dying of on high air pollution days?
91 Lundborg M, Johard U, Lastbom L, Gerde P, Camner P. Human Environ Res 1994; 64: 26–35.
alveolar macrophage phagocytic function is impaired by aggregates of 117 Janssen NA, Hartog JJ de, Hoek G, et al. Personal exposure to fine
ultrafine carbon particles. Environ Res 2001; 86: 244–53. particulate matter in elderly subjects: relation between personal,
92 Stearns RC, Paulauskis JD, Godleski JJ. Endocytosis of ultrafine indoor, and outdoor concentrations. J Air Waste Manag Assoc
particles by A549 cells. Am J Respir Cell Mol Biol 2001; 24: 108– 2000; 50: 1133–43.
15. 118 Sarnat JA, Schwartz J, Catalano PJ, Suh HH. Gaseous pollutants in
93 Renwick LC, Donaldson K, Clouter A. Impairment of alveolar particulate matter epidemiology: confounders or surrogates? Environ
macrophage phagocytosis by ultrafine particles. Toxicol Appl Pharm Health Perspect 2001; 109: 1053–61.
2001; 172: 119–27. 119 Rijnders E, Janssen NA, Vliet PH van, Brunekreef B. Personal and
94 Beck-Speier I, Dayal N, Karg E, et al. Agglomerates of ultrafine outdoor nitrogen dioXide concentrations in relation to degree of
particles of elemental carbon and TiO2 induce generation of lipid urbanization and traffic density. Environ Health Perspect 2001;
mediators in alveolar macrophages. Environ Health Perspect 2001; 109 (suppl 3): 411–17.
109 (suppl 4): 613–18. 120 Oglesby L, Kunzli N, Roosli M, et al. Validity of ambient levels
95 Dellinger B, Pryor WA, Cueto R, et al. Role of free radicals in the of fine particles as surrogate for personal exposure to outdoor air
toXicity of airborne fine particulate matter. Chem Res Toxicol 2001; pollution—results of the European EXPOLIS-EAS Study (Swiss
14: 1371–77. Center Basel). J Air Waste Manag Assoc 2000; 50: 1251–61.
96 Sun G, Crissman K, Norwood J, et al. OXidative interactions of 121 Hoek G, Fischer P, Van Den Brandt P, Goldbohm S, Brunekreef B.
synthetic lung epithelial lining fluid with metal-containing particulate Estimation of long-term average exposure to outdoor air pollution for
matter. Am J Physiol Lung Cell Mol Physiol 2001; 281: L807–15. a cohort study on mortality. J Expo Anal Environ Epidemiol 2001; 11:
97 Van Eeden SF, Tan WC, Suwa T, et al. Cytokines involved in the 459–69.
systemic inflammatory response induced by exposure to particulate 122 Seaton A, MacNee W, Donaldson K, Godden D. Particulate air
matter air pollutants (PM(10)). Am J Respir Crit Care Med 2001; 164: pollution and acute health effects. Lancet 1995; 345: 176–78.
826–30. 123 Peters A, Wichmann HE, Tuch T, Heinrich J, Heyder J. Respiratory
98 Salvi SS, Nordenhall C, Blomberg A, et al. Acute exposure to diesel effects are associated with the number of ultrafine particles.
exhaust increases IL-8 and GRO-alpha production in healthy human Am J Respir Crit Care Med 1997; 155: 1376–83.
airways. Am J Respir Crit Care Med 2000; 161: 550–57. 124 Penttinen P, Timonen KL, Tiittanen P, et al. Number concentration
99 Salvi S, Blomberg A, Rudell B, et al. Acute inflammatory responses in and size of particles in urban air: effects on spirometric lung function
the airways and peripheral blood after short-term exposure to diesel in adult asthmatic subjects. Environ Health Perspect 2001; 109: 319–
exhaust in healthy human volunteers. Am J Respir Crit Care Med 23.
1999; 159: 702–09. 125 van Vliet P, Knape M, de Hartog J, et al. Motor vehicle exhaust
100 Mukae H, Vincent R, Quinlan K, et al. The effect of repeated and chronic respiratory symptoms in children living near freeways.
exposure to particulate air pollution (PM10) on the bone marrow. Environ Res 1997; 74: 122–32.
Am J Respir Crit Care Med 2001; 163: 201–09. 126 Janssen NA, Schwartz J, Zanobetti A, Suh HH. Air conditioning and
101 Wu W, Samet JM, Ghio AJ, Devlin RB. Activation of the EGF source-specific particles as modifiers of the effect of PM(10) on
receptor signaling pathway in airway epithelial cells exposed to hospital admissions for heart and lung disease. Environ Health Perspect
Utah Valley PM. Am J Physiol Lung Cell Mol Physiol 2001; 2002; 110: 43–49.
281: L483–89. 127 Roemer WH, van Wijnen JH. Daily mortality and air pollution along
102 Puddicombe SM, Davies DE. The role of MAP kinases in busy streets in Amsterdam, 1987–1998. Epidemiology 2001; 12:
intracellular signal transduction in bronchial epithelium. Clin Exp 649–53.
Allergy 2000; 30: 7–11. 128 Laden F, Neas LM, Dockery DW, Schwartz J. Association of fine
103 Seaton A, Soutar A, Crawford V, et al. Particulate air pollution particulate matter from different sources with daily mortality in six
and the blood. Thorax 1999; 54: 1027–32. US cities. Environ Health Perspect 2000; 108: 941–47.
104 Hashimoto K, Ishii Y, Uchida Y, et al. EXposure to diesel 129 Ghio AJ, Devlin RB. Inflammatory lung injury after bronchial
exhaust exacerbates allergen-induced airway responses in guinea instillation of air pollution particles. Am J Respir Crit Care Med 2001;
pigs. 164: 704–08.
Am J Respir Crit Care Med 2001; 164: 1957–63. 130 Schwartz J, Dockery DW, Neas LM. Is daily mortality associated
105 Fujieda S, Diaz-Sanchez D, Saxon A. Combined nasal challenge with specifically with fine particles? J Air Waste Manag Assoc 1996; 46:
diesel exhaust particles and allergen induces in vivo IgE isotype 927–39.
switching. Am J Respir Cell Mol Biol 1998; 19: 507–12. 131 Schwartz J, Neas LM. Fine particles are more strongly associated
106 Nel AE, Diaz-Sanchez D, Ng D, Hiura T, Saxon A. Enhancement of than coarse particles with acute respiratory health effects in
allergic inflammation by the interaction between diesel exhaust schoolchildren. Epidemiology 2000; 11: 6–10.
particles and the immune system. J Allergy Clin Immunol 1998; 102: 132 McDonnell WF, Nishino-Ishikawa N, Petersen FF, Chen LH,
539–54. Abbey DE. Relationships of mortality with the fine and coarse
107 Brown DM, Wilson MR, MacNee W, Stone V, Donaldson K. Size- fractions of long-term ambient PM10 concentrations in nonsmokers.
dependent proinflammatory effects of ultrafine polystyrene particles: J Expo Anal Environ Epidemiol 2000; 10: 427–36.
a role for surface area and oXidative stress in the enhanced activity 133 Smith RL, Spitzner D, Kim Y, Fuentes M. Threshold dependence of
of ultrafines. Toxicol Appl Pharm 2001; 175: 191–99. mortality effects for fine and coarse particles in Phoenix, Arizona.
108 Cakmak S, Burnett RT, Krewski D. Methods for detecting and J Air Waste Manag Assoc 2000; 50: 1367–79.
estimating population threshold concentrations for air pollution- 134 Anderson HR, Bremner SA, Atkinson RW, Harrison RM, Walters S.
related mortality with exposure measurement error. Risk Anal 1999; Particulate matter and daily mortality and hospital admissions in the
19: 487–96. west Midlands conurbation of the United Kingdom: associations with
109 Daniels MJ, Dominici F, Samet JM, Zeger SL. Estimating fine and coarse particles, black smoke and sulphate. Occup Environ
particulate matter-mortality dose-response curves and threshold Med 2001; 58: 504–10.
levels: an analysis of daily time-series for the 20 largest US cities. 135 Pope CA 3rd. Respiratory disease associated with community air
Am J Epidemiol 2000; 152: 397–406. pollution and a steel mill. Utah Val Am J Public Health 1989; 79:
110 Brunekreef B, Dockery DW, Krzyzanowski M. Epidemiologic studies 623–28.
on short-term effects of low levels of major ambient air pollution 136 Pope CA 3rd. Particulate pollution and health: a review of the Utah
components. Environ Health Perspect 1995; 103 (suppl 2): 3–13. valley experience. J Expo Anal Environ Epidemiol 1996; 6: 23–34.
111 McMichael AJ, Anderson HR, Brunekreef B, Cohen AJ. 137 Friedman MS, Powell KE, Hutwagner L, Graham LM, Teague WG.
Inappropriate use of daily mortality analyses to estimate longer-term Impact of changes in transportation and commuting behaviors
mortality effects of air pollution. Int J Epidemiol 1998; 27: 450–53. during the 1996 Summer Olympic Games in Atlanta on air quality
112 Schwartz J. Harvesting and long term exposure effects in the relation and childhood asthma. JAMA 2001; 285: 897–905.
between air pollution and mortality. Am J Epidemiol 2000; 151: 138 Heinrich J, Hoelscher B, Wichmann HE. Decline of ambient air
440–48.

THE LANCET • Vol 360 • October 19, 2002 • 13

For personal use. Only reproduce with permission from The Lancet Publishing
 REVI

pollution and respiratory symptoms in children. Am J Respir Crit Care

Med 2000; 161: 1930–36. estimates of the exposure-response relationship between daily
139 Schwartz J, Dockery DW. Increased mortality in Philadelphia mortality and particulate air pollution. Environ Health Perspect 1996;
associated with daily air pollution concentrations. Am Rev Respir 104: 414–20.
Dis 1992; 145: 600–04. 144 Burnett R, Ma R, Jerrett M, et al. The spatial association between
140 Kelsall JE, Samet JM, Zeger SL, Xu J. Air pollution and mortality in community air pollution and mortality: a new method of analyzing
Philadelphia, 1974-1988. Am J Epidemiol 1997; 146: 750–62. correlated geographic cohort data. Environ Health Perspect 2001;
141 Krewski D, Burnett R, Goldberg MS, et al. Reanalysis of the 109 (suppl 3): 375–80.
Harvard six cities study and the American Cancer Society study 145 Greenbaum DS, Bachmann JD, Krewski D, et al. Particulate
of particulate air pollution and mortality: Health Effects air pollution standards and morbidity and mortality: case study.
Institute special report. Boston: Health Effects Institute, 2000. Am J Epidemiol 2001; 154: S78–90.
142 Samet J, Zeger S, Kelsall J, Xu J, Kalkstein L. Does weather 146 Kunzli N, Kaiser R, Medina S, et al. Public-health impact of outdoor
confound or modify the association of particulate air pollution and traffic-related air pollution: a European assessment. Lancet 2000;
with mortality? An analysis of the Philadelphia data, 1973–1980. 356: 795–801.
Environ Res 1998; 77: 9–19. 147 Hollander AE de, Melse JM, Lebret E, Kramers PG. An aggregate
143 Pope CA 3rd, Kalkstein LS. Synoptic weather modeling and public health indicator to represent the impact of multiple
environmental exposures. Epidemiology 1999; 10: 606–17.

Uses of error
Uncertainty
David A Grimes

Everything pointed to the same diagnosis: an ectopic

Alone with this diagnostic dilemma, I excused myself
pregnancy. Nonetheless, I was nervous making the
from the patient and her family and slipped off discreetly
diagnosis by myself. I was a very junior house officer in
to the hospital library. My memory had been correct: every
obstetrics and gynaecology, alone on call at a municipal
gynaecology textbook on the shelf concurred that non-
hospital. My patient had an abnormal last menses about
clotting blood on culdocentesis indicated an ectopic
8 weeks ago, no recent use of contraception, and vague
pregnancy. None, however, mentioned my current
pelvic pain. Her pulse and blood pressure were normal
quandary: intraperitoneal blood that clotted. Given this
without orthostatic changes. She looked uncomfortable
uncertainty, I did not feel comfortable mobilising an
but not in acute distress. Her abdomen was tender, and
operating theatre (which required calling in staff from
the pelvic examination caused her discomfort, more on
home) to perform a laparotomy. Nor did I feel
one side than the other. I could feel no adnexal mass, but
comfortable sending my patient home. I admitted her to
my clinical experience was very limited. Her haemoglobin
the hospital, where I watched and fretted over her all
was around 110 g/L; not unusual for our indigent
night.
patients.
By morning, her pain was unchanged. Her repeat
From my readings and observations as a medical
haemoglobin, however, had plummeted, and my error
student, I knew just what to do to confirm my diagnosis:
was apparent. We rushed her to the operating theatre. At
a culdocentesis. At this time, sensitive pregnancy tests
laparotomy, she had litres of blood in her abdomen and
were not available, diagnostic ultrasound was in its
required transfusion of several units of blood as a
infancy, and laparoscopy had just reached the USA and
consequence of my timidity. I had erred in trusting the
was not available by night at my hospital. After
textbooks instead of my inexperienced clinical hunch. I
obtaining informed consent, I introduced a speculum and
later learned from published articles that a few such
inserted a 20-gauge needle through her vaginal vault into
patients have blood that will not clot. Some authors call
her rectouterine pouch. To my delight, when I withdrew
this “non-conclusive” evidence of an ectopic pregnancy.
the plunger, the syringe promptly filled with dark blood.
I learned an important lesson that night about the
To my considerable dismay, however, the blood
limitations of textbooks and authorities. As Antman and
promptly clotted in the syringe. This was not supposed
colleagues showed several decades later, textbooks and
to happen: the hallmark of an ectopic pregnancy was
their authors are dangerously obsolete. But Mark Twain
non-clotting blood on culdocentesis. I had never read,
may have put it best a century ago: “Be careful about
seen, or heard of blood clotting in this setting.
reading health books. You may die of a misprint.” The
same holds true for omissions.

Family Health International, Research Triangle Park, NC 27709, USA (D A Grimes MD)

14 THE LANCET • Vol 360 • October 19, 2002 •

For personal use. Only reproduce with permission from The Lancet Publishing