Home » Ozone » Level 3 » Question 4

Air Pollution Ozone

4. Should current O3 guidelines be reconsidered?

  • 4.1 Have positive impacts on public health of Ozone reductions been shown?
  • 4.2 Averaging period most relevant for Ozone standards to protect human health
  • 4.3 Reconsideration of the current WHO Guidelines for Ozone

4.1 Have positive impacts on public health of Ozone reductions been shown?

The source document for this Digest states:


There are very few opportunities to evaluate O3 reduction per se. One study of intra-state migrants showed a beneficial effect on lung function in children who moved to lower PM and O3 areas. A decrease in O3 during the 1996 Olympics was associated with a reduction of asthma admissions. The interpretation of these findings is unclear.


Emission reductions of O3 precursors (NOx and volatile organic compounds) can result in lower concentrations of not only NO2 and O3, but in fine particles (PM2.5) as well. Without the oxidants generated in the photochemical reaction sequences, there would be a reduction in the oxidation of SO22 and NO2, which leads to acidic sulfate and fine particles and nitric acid vapour, as well as less formation of organic fine particles. Therefore an assessment of the beneficial effect of reducing O3 only is difficult.

Children in the Southern California cohorts who moved from communities with relatively high PM and O3 concentrations to communities with lower concentrations had better lung function growth than children who remained in those communities (24), while children who moved from communities with relatively low PM2.5 and O3 concentrations to communities with higher PM2.5 and O3 concentrations had lesser lung function growth than those who remained in the cleaner communities. However, it is not clear whether this results is due to changes in O3 or PM. Friedman et al (291) took advantage of a natural experiment associated with a decrease in O3 exposure in Atlanta during the 1996 Olympics and demonstrated that acute O3 effects to asthma admissions were substantially reduced. In other air pollution situations the beneficial effects mainly of reducing particulate matter (203) and SO22 (292, 293) have been demonstrated. More research is needed in that area but the appropriate settings are few.

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 6 Ozone (O3), Section 6.2 Answer and rationales, Question 11

The same information on
Particulate MatterNitrogen Dioxide

4.2 Averaging period most relevant for Ozone standards to protect human health

The source document for this Digest states:


For short-term exposure, it is clear that the effects increase over multiple hours (e.g., 6–8 hours for respiratory function effects and lung inflammation). Thus, an 8-hour averaging time is preferable to a 1 hour averaging time. The relationship between long term O3 exposure and health effects is not yet sufficiently understood to allow for establishing a long-term guideline.


From controlled human exposure studies it appears that the effects increase over multiple hours (258, 261, 263). The evidence from epidemiological studies is not conclusive, because in practice there is a strong correlation between the different measures. The association of O3 exposure with long-term effects is not yet clear enough to justify recommending a long-term standard (see Rationale to Question 2).

Table 2: Summary of meta-analysis of time-series studies published 1996–2001

Table 3: Summary of studies measuring short-term effect on lung function

Table 4: Short-term effects of ozone on lung function, biological and other responses

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 6 Ozone (O3), Section 6.2 Answer and rationales, Question 12

4.3 Reconsideration of the current WHO Guidelines for Ozone

The source document for this Digest states:


The current WHO Air quality guidelines (AQG) (WHO, 2000) for O3 provide a guideline value of 120µg/m3 (60 ppb), based on controlled human exposure studies, for a maximum 8-hour concentration. The AQG also provide two concentration-response tables, one for health effects estimated from controlled human exposure studies and one from epidemiological studies. No guideline for long-term effects was provided. Since the time these guidelines were agreed, there is sufficient evidence for their reconsideration. Issues to be considered are: the averaging time(s) for the short-term guidelines and their associated levels, the concentration-response functions used in the tables, the outcomes included in the concentration-response tables, whether a long- term guideline and/or complementary guidelines (e.g. restricting personal activity) should be adopted.

Recent epidemiological studies have strengthened the evidence that there are short-term O3 effects on mortality and respiratory morbidity and provided further information on exposure- response relationships and effect modification. There is new epidemiological evidence on long- term O3 effects and experimental evidence on lung damage and inflammatory responses. There is also new information on the relationship between fixed site ambient monitors and personal exposure, which affects the interpretation of epidemiological results.


Since 1996 several epidemiological studies assessed the short-term effects of O3 on various health outcomes. Based on a meta-analysis of studies published during the period between 1996 and 2001 on short-term effects of O3 on all non-accidental causes of death in all ages (or older than 65 years), significant increase of the risk of dying (between 0.2 % and 0.6 % per each increase in 10 µg/m3 or 5 ppb) was shown (210) whatever the lag period, the season of study or the timing of the ozone measurement (Table 2). In some instances, the effects coefficients observed were higher in places with low O3 concentrations. This may be a reflection of a curvilinear concentration – response, or of other specific characteristics of populations where influential studies were done (see also rationale to question 3). Studies limited to the summer season tend to reveal a larger effect, while the strength of the effect increases with longer average times (> 1 hour) of O3 measurement. Estimates remained very similar if studies using Generalized Additive Models (GAM) were excluded to avoid a possible bias which has recently been reported (31). In addition, a large multi-centre study from the United States of America, the NMMAPS study, reported a significant effect of O3 during the summer season, of 0.41 % increase in mortality associated with an increase of 10 ppb (20 µg/m3) in daily O3 concentrations at lag 0 (i.e. the same day). A larger effect was found at lag 2 (levels two days earlier), independently of other pollutants (27, 211). Ozone daily levels were associated with hospital respiratory admissions at all ages in most of the studies using 8-hour measures (Table 2) and also in many of the studies using other averaging periods. The magnitude of the association was slightly larger than that obtained for mortality (0.5 to 0.7 % increase in admissions per increase of 10 µg/m3 or 5 ppb in O3; Table 2). There are very few studies reporting data on lag 0. Studies on admissions for asthma in children did not find conclusive associations with any O3 measurement. However, there is evidence that during days when ozone levels are high, asthmatic subjects increase their use of medication (212) that may mask any adverse O3 effect (213).

It should be noted that O3 usually displays a strong seasonality (with a summer peak), which is different from the seasonal patterns of other pollutants and of the above health outcomes. Therefore, if careful control of seasonal patterns is not applied, the effect of O3 is underestimated (and may appear protective). All the above studies have allowed for seasonal adjustment in various ways.

In addition, all studies reported from 1996 and 2001, which give estimates of O3 effects on lung function measures were considered (210). The estimates were grouped by the subjects’ characteristics but there was a mixture of lags and averaging times. Therefore, summary estimates are not provided (Table 3). Overall, the majority of studies showed a negative impact of acute effects of O3 on lung function.

Some epidemiological studies on long-term effects of O3 have been published during the period from 1996 to 2002, giving some evidence of long-term effects on various health endpoints. These studies are discussed in more detail in the rationale to question 2.

New data from experimental studies have not contributed much additional evidence for O3 effects at current ambient levels. Results from experimental studies show the potential of O3 exposure to cause effects and have provided some insights to underlying mechanisms. Some of the most relevant findings of these studies are presented below. In interpreting these, it is important to note that only healthy or mildly asthmatic subjects were included in the study populations.

From a controlled exposure study (214) in healthy and allergic asthmatic nonsmokers there is evidence of lung function decrements after 3 hrs of exposure to either 100 µg/m3 of H2SO4 or NaCl (control) aerosol followed by 360 µg/m3 (180 ppb) of O3, with greater decrements for those exposed to H2SO4. Repeated daily, short term exposures of healthy and mildly asthmatic subjects to O3 attenuates the acute lung function and, to a less extent the inflammatory response, reaching a maximum over 3 to 5 days and with a recovery over four to seven days after the end of the exposure (215, 216, 217, 218, 219, 220). Bronchoalveolar lavage demonstrates that mucosal damage and inflammation continue despite adaptation documented by lung function and clinical assessment (220, 221).

Since the last WHO evaluation new non-invasive tests have become available in both humans and animals allowing non-invasive exploration of lung damage and inflammation not only under controlled exposures studies but also under field conditions on subjects exposed to ambient O3 (222). These tests include the assay in serum of lung-specific proteins to detect lung epithelium permeability changes or the analysis of inflammatory markers in exhaled air or in the condensates of exhaled breath condensate. Compared to lung lavage techniques and other tests of lung damage, these non-invasive tests present several advantages, such as sensitivity, repeatability, non-invasiveness and applicability in field studies. In particular, they allow for monitoring of lung inflammation or damage induced by ambient O3, especially in sensitive groups such as children.

There is evidence for a significant association between short-term peaks in ambient air concentrations of O3 and lung epithelial damage (222) as measured by the intravascular leakage Clara cell protein (CC16). Other studies in humans have shown that spirometric variables show adaptation in young adults; and persistent small airway dysfunction/resistance (0.25 ppm for 2 hours over 4 days) (223) and that repeated exposure (0.125 ppm for 2 hours over 4 days) of allergic asthmatics enhances progressively both functional and inflammatory, bronchial responses to inhaled allergen challenge (224), see also Table 4.

Studies in animals undergoing controlled exposures to O3 have also shown various biological responses at different schemes and levels of exposure (225, 226, 227).

A discussion of the relationship between fixed site ambient monitors and personal exposure can be found in the rationale to Question 9.

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 6 Ozone (O3), Section 6.1 Answer and rationales, Question 1

Themes covered
Publications A-Z