Home » Nitrogen Dioxide » Level 3 » Question 4

Air Pollution Nitrogen Dioxide

4. Should current NO2 guidelines be reconsidered?

  • 4.1 Have positive impacts on public health of reductions of emissions and/or ambient concentrations of NO2 been shown?
  • 4.2 What averaging period (time pattern) is most relevant from the point of view of protecting human health?
  • 4.3 Is there new scientific evidence to justify reconsideration of the current WHO Guidelines for NO2?

4.1 Have positive impacts on public health of reductions of emissions and/or ambient concentrations of NO2 been shown?

The source document for this Digest states:


No recent peer reviewed publication could be found to answer this question.


It has not been possible to study impacts of reduction in NOx emissions or NO2 concentrations in the ambient air because there have been no good examples of such reductions. In Europe, there have been moderate decreases in emissions (416) and ambient concentrations in urban areas (418) in the last decade. In the United States of America, reduced emission rates from individual vehicles and power plants have been offset by increases in vehicle km travelled from road transport, leaving ambient levels relatively constant over the past decades.

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 7 Nitrogen dioxide, Section 7.2 Answers and rationales, Question 11

The same information on
Particulate MatterOzone

4.2 What averaging period (time pattern) is most relevant from the point of view of protecting human health?

The source document for this Digest states:

Would additional protection be provided by setting standards for more than one averaging period for NO2?


With regard to protection against acute health effects, either the peak-hour average or 24hr (daily) average NO2 concentrations can be used as a measure of direct short-term exposure, since they are highly correlated in urban areas. Having a longer-term guideline value is also supported by the evidence on possible direct effects of NO2, and on its indirect consequences through the formation of secondary pollutants.


The limited data on short-term responses to controlled NO2 exposures in chamber studies does not provide any guidance on the relevant averaging time for the physiological responses on the scale of hours to days. However, this is a less important issue in terms of a guideline because of the high correlation of hourly maximum with daily average values.

In terms of the long-term effects of NO2 and/or the secondary pollutants that result from the presence of NO2 in the ambient air, maintaining an annual NO2 WHO guideline value is most appropriate. Effective implementation of an annual NO2 guideline, especially at or near the current level, would necessitate reductions in ambient NO2 concentrations in highly populated regions in Europe. Such reductions in NO2 can be expected to reduce the secondary pollutant concentrations as well.

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 7 Nitrogen dioxide, Section 7.2 Answers and rationales, Question 12

4.3 Is there new scientific evidence to justify reconsideration of the current WHO Guidelines for NO2?

The source document for this Digest states:


The current WHO guideline values for NO2 are a 1-hour level of 200 µg/m3 and an annual average of 40 µg/m3. Since the previous review, only a small number of additional human exposure studies have been carried out. These do not support the need to change the 1-hour guideline value. With regard to the annual average, there have been some new epidemiological studies reporting associations of longer-term exposure with lung function and respiratory symptoms. The former group that proposed the annual guideline value of 40 µg/m3 acknowledged that “although there is no particular set of studies that clearly support the selection of a specific numerical value for an annual average guideline the database nevertheless indicates a need to protect the public from chronic nitrogen dioxide exposures”. Because of a lack of evidence, the former group selected a value from a prior WHO review. The new evidence does not provide sufficient information to justify a change in the guideline value. Given the role of NO2 as a precursor of other pollutants and as a marker of traffic related pollution, there should be public health benefits from meeting the current guidelines. Thus the present working group did not find sufficient evidence to reconsider the current 1-hour and annual WHO guidelines for NO2.


Ambient air NO2 is in large part derived from the oxidation of NO, the major source of which is combustion emissions, mainly from vehicles. NO2 is therefore a clear indicator for road traffic. NO2 is also subject to extensive further atmospheric transformations that lead to the formation of O3 and other strong oxidants that participate in the conversion of NO2 to nitric acid and SO2 to sulphuric acid and subsequent conversions to their ammonium neutralization salts. Thus, through the photochemical reactions sequence initiated by solar-radiation-induced activation of NO2, the newly generated pollutants formed are an important source of nitrate, sulphate and organic aerosols that can contribute significantly to total PM10 or PM2.5 mass. For these reasons, NO2 is a key precursor for a range of secondary pollutants whose effects on human health are well documented.

Short-term chamber studies with humans (384, 385) as well as in animal exposure studies continue to support the view that at the levels encountered in the ambient outdoor air, direct effects of NO2 alone on the lungs (or any other system) are minimal or undetectable. At NO2 concentrations in excess of those achieved in outdoor air (with the possible exception of road tunnels), mild airway inflammation occurs (386). Human chamber studies have shown that in allergic subjects NO2 can enhance the effect of allergens (387, 388, 389). Bronchial reactivity also was increased in the presence of NO2 (390). These studies show effects at levels twice or more of the current guideline value and therefore do not provide evidence to justify a change.

The number of time-series studies using peak hourly concentrations and/or daily mean concentrations of NO2 as air pollution indicators has grown substantially over the past five years. Overall, these studies have supported associations between ambient NO2 concentrations and a range of adverse health effects. A meta-analysis on mortality showed consistent associations with NO2 (391). Hospital admissions for respiratory disease were shown to increase with increasing levels of NO2 in some areas (357). While such studies have supported an association between ambient NO2 concentrations in relation to multiple respiratory and cardiovascular health outcome measures, these cannot establish causality for an NO2 effect. It is likely that many health effects observed occur as a result of exposure to confounding traffic related pollutants and/or secondary pollutants that include O3, acid aerosol and particles. In a significant proportion of the epidemiological studies increased risks of adverse health outcomes diminish considerably or become non-significant when other pollutants are considered in the statistical models. In others, however, NO2 enhances the magnitude of effects observed with other pollutants (29, 349).

Recent epidemiological studies have shown consistent associations between long term exposure to NO2 and lung function in children (22, 23, 236) as well as with lung function and respiratory symptoms in adults. These effects cannot be attributed to NO2 exposure per se. Few studies have attempted (or had the data) to evaluate the relative contribution of particulate matter and NO2 on the health outcomes described. Those who were able to do this show a separate but smaller contribution of NO2 (392).

The importance of NO2 as a key component for the rise of secondary toxic pollutant concentrations in ambient air and its potential in mixtures to enhance the effects of other environmental pollutants including allergens, warrants a guideline that limits the resulting health effects. Indeed, some additional emphasis might be given to NO2 for the very reason that it is a good indicator of traffic-related air pollution and an important source of a range of more toxic pollutants that probably act in combination to produce adverse health effects.

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 7 Nitrogen dioxide, Section 7.2 Answers and rationales, Question 1

What is the basis for maintaining the WHO NO2 annual specific guideline value of 40 µg/m3


There is evidence from toxicological studies that long-term exposure to NO2 at concentrations higher than current ambient concentrations has adverse effects.

Uncertainty remains over the significance of NO2 as a pollutant with a direct impact on human health at current ambient air concentrations in the European Union, and there is still no firm basis for selecting a particular concentration as a long-term guideline for NO2. NO2 is an important constituent of combustion generated air pollution. In recent epidemiological studies of the effects of combustion-related (mainly traffic generated) air pollution, NO2 has been associated with adverse health effects even when the annual average NO2 concentration is within a range that includes 40 µg/m3, the current guideline value. However, we are unable to establish an alternative NO2 guideline from these studies. We therefore recommend that the WHO annual specific guideline value of 40 µg/m3 should be retained or lowered.


The WHO Air Quality Guideline value of 40 µg/m3 as annual mean was based on limited but suggestive evidence from indoor studies that long-term exposure to combustion products from indoor gas appliances including NO2 had a deleterious effect on health. The figure of 40 µg/m3 was adopted from the International Programme on Chemical Safety (IPCS; Environmental Health Criteria 188) and reflected the association between indoor exposure to combustion products from indoor gas appliances including NO2 and lower respiratory tract illnesses in children. There was some uncertainty over the appropriate numerical value for the guideline, as NO2 was not directly measured in all studies. It was also noted that outdoor epidemiological studies had shown associations between outdoor concentrations of NO2 and small reductions in lung function and a slightly increased prevalence of asthma.

Since the current WHO Air Quality Guidelines were established (WHO, 2000) the evidence regarding the long-term effects of NO2 as a single pollutant and adverse human health has increased a little. There is new evidence from human and animal studies in vivo (Pamanathan et al., 2003; Blombert et al., 1997; van Bree et al., 2000) and from limited in vitro (Devalia et al., 1993; Bayram et al., 1999) studies that short-term exposure to NO2 can be toxic to airway and alveolar epithelial cells. This is a result of its oxidant capacity to cause tissue damage (Persinger et al., 2002). This lends plausibility to the possibility that there could be long-term effects. There are no human studies of long-term exposure to pure NO2; studies have only been performed with short-term exposure to concentrations higher than common ambient concentrations and not at the low levels that might be causing long-term effects (e.g. Chitano et al., 1995; Hyde et al., 1978; Wegmann et al., 2003). There are also studies showing lung damage following long-term exposure to nitrogen dioxide in animals. Again the concentrations used are above common ambient concentrations.

Epidemiological studies of short-term changes in NO2 concentrations suggest that these may be associated with ill health at common ambient concentrations. This may have implications for long-term effects. None of these studies, however, provides a significantly improved basis for the long-term guideline of 40 µg/m3 as annual average for NO2 as a single toxic pollutant gas.

We have been asked to comment on our confidence in this guideline. Our reply is that it remains difficult to provide solid scientific support for the numerical value of the guideline. There still is no robust basis for setting an annual average guideline value for NO2 through any direct toxic effect. However, new epidemiological evidence has emerged that increases our concern over health effects associated with outdoor air pollution mixtures that are apparently well characterized by NO2. We refer to evidence supporting effects of a mixture of NO2 and derived pollutants including nitrate rich particles and nitric acid vapour at mean NO2 concentrations in a range that includes 40 µg/m3 (range 8–75 µg/m3) (McConnell et al., 2003). This study demonstrated an association between bronchitis symptoms among children with asthma and the yearly variability of NO2 concentrations in southern Californian communities. We note that the highest 4 year average concentration of NO2 in the communities studied was 75 µg/m3. Of interest is the high correlation between NO2 and Organic Carbon (OC) in this study (0.69). In two-pollutant models, adjusting for O3, PM10, PM2.5, coarse particles, inorganic acid and elemental carbon, OC was the pollutant that retained most of its significance. PM2.5 and NO2 also had fairly robust associations with symptoms; PM2.5, NO2 and OC all lost their significance in all two-pollutant models including combinations of these, showing that either was representing a gas-particle mixture most likely dominated by traffic emissions. In the same study, Gauderman et al. (2000) reported an association between annual average PM10, PM2.5, inorganic acid and NO2 concentrations and a reduction in lung growth in fourth grade children. In this study, the highest concentration of NO2 recorded was about 45 ppb (80 µg/m3) while in 6 of the 12 communities studied the annual average concentrations was less than 40 µg/m3. Again, there were significant correlations between these pollutants (up to 0.87 for NO2 and inorganic acid), and all of the associations lost significance in two-pollutant models including any combination between the four. A further study from this cohort (Gauderman et al., 2002) included more detailed measurements of acid vapours and of elemental and organic carbon. In this study acid vapour (sum of nitric, formic and acetic acid) was the clearest determinant of reduced lung function growth, and again, acid, NO2 and particle metrics mostly lost significance in two- pollutant models. In this second study, however, NO2 coefficients reduced less after adjustment for PM2.5 or PM10 than vice versa. What these studies point towards is that NO2 is serving as an indicator of a complex mixture, and that its indicator value is reduced, but not removed by adjustment for PM2.5 and other particle metrics.

In Europe, a negative association between the development of lung function and ambient NO2 concentrations (Schindler et al., 1998) has been reported, where the highest ambient annual mean concentration of NO2 in the communities studied was 57.5 µg/m3. In this study, the role of co- pollutants was not examined

A recent series of European studies has used Geographic Information Systems (GIS) to generate exposure distributions for traffic-related pollutants, primarily NO2, “soot” (measured as reflectance of particle filters) and PM2.5. These studies were conducted over annual average NO2 ranges of 25–84 µg/m3 (Carr et al., 2002; Nicolai et al., 2003), 15–67 µg/m3 (Hoek et al., 2001; Hoek et al., 2002), 27–44 µg/m3 (Janssen et al., 2001; Janssen et al., 2003), 20–67 µg/m3 (Gehring et al., 2002) and 13–58 µg/m3 (Brauer et al., 2002). In all of these, the correlation between the two or three metrics of exposure was high, so that they could not be separated. In all studies, there were positive associations between NO2 and health endpoints such as respiratory symptoms and mortality indicating that over these low ranges of exposure, NO2 as marker of traffic-related pollution is clearly associated with adverse effects on health. In a European study of lung cancer, NO2 (calculated from NOx) was used explicitly as a marker of road traffic exhaust levels, since historical road maps and traffic flow estimates, and dispersion models, were used to map the road traffic contribution to ambient NO2. These geographical estimates produced an individual average in the range 4–51 µg/m3 for the first decade of the study (Bellander et al., 2001). In a similar recent European study total residential NOx (mainly from vehicles) was historically assessed by dispersion modelling, showing a five year individual average in the range 1–170 µg/m3 (Nafstad et al., 2003). Residential road traffic exhaust levels corresponding to over 30 µg/m3 NO2 or NOx were found to increase the lung cancer risk by about 40% 10–30 years later in the two studies (Nyberg et al., 2000, Nafstad et al., 2003). A point worthy of note is that associations of ambient NO2 concentrations with health effects is manifest in a much smaller geographical scale than has previously reported (Cohen, 2003).

The current annual average NO2 guideline value of 40 µg/m3 is within the exposure ranges reported in these investigations, and one being conducted almost entirely over a range below the current annual average guideline concentration value. These recently published studies document that NO2, as marker of a complex mixture or traffic-related pollution, is consistently associated with adverse effects on health at relatively low levels of long-term average exposure. They also show that these associations cannot be completely explained by co-exposure to PM2.5, but that other components in the mixture (such as organic carbon and acid vapour) might explain part of the association. As such components have not been routinely measured, and as there is much information on NO2 concentrations in ambient air, it seems reasonable to propose to CAFE that a prudent annual average limit value for NO2 be set, acknowledging that this takes account of any direct toxic effect that NO2 might exert and to control complex mixtures of combustion-related pollution (mainly from road traffic).

There are some limitations to using NO2 purely as an indicator for combustion-related air pollution, since the mixture will vary in different places and change with time. However, this limitation would also apply for, e.g. PM2.5 since it is not known what aspects or components of particulate air pollution are responsible for the adverse health effects observed. When more information on the relationship between different aspects of combustion-related air pollution and health is available, it is possible that more efficient protection against the health effects of these complex gas-particle mixtures will be obtained by regulating another metric than NO2 alone or in combination (Seaton & Dennekamp, 2003). Such candidates include black smoke, elemental and organic carbon, measures of acidity, NOx and particulate number concentration.

Research should be undertaken to determine whether NO2 at concentrations achieved in the outdoor environment, has any detectable toxicity on the human lung using a range of outcome measures. Research is also urgently needed to determine which aspects or components of combustion mixtures are responsible for the adverse health effects observed in epidemiological studies.

Source & ©: WHO Regional Office for Europe  Health Aspects of Air Pollution - answers to follow-up questions from CAFE (2004), Question 7

Themes covered
Publications A-Z