National Ambient Air Quality Objectives For Ground-Level Ozone- Summary - Science Assessment Document

11. Conclusions

Ground-level ozone is formed when its precursors, nitrogen oxides and volatile organic compounds, combine in the presence of sunlight. Ozone is one of several pollutants that combine to form a chemical soup that hangs in the warm, still air over many Canadian cities on hot summer days. Scientific evidence clearly links ozone and health impacts. As a mixture, and individually, these air pollutants have the potential to cause adverse health effects.

The population-based epidemiological studies have provided a consistent and coherent evidence of an exposure-response relationship. Non-accidental mortality, hospital admissions,

Emergency Department (ED) visits and reduced activity days increase monotonically as ozone concentration increases. Increased risks for non-accidental mortality, respiratory hospitalization and ED visits respectively are estimated at 0.4%, 1-2% and 6-8.6%, for every 10 ppb increase in ozone. The risks of population health effects increase monotonically, in a ozone concentration-dependent fashion. The controlled human exposure studies have identified a dose-response relationship for lung function changes, symptoms and airway inflammation, the frequency and intensity of response increasing with increases of ozone concentration, exposure duration or ventilation rate. Field (camp and panel) studies and controlled human exposure studies have identified that patients with asthma, COPD and allergic rhinitis are more susceptible to ozone-induced health effects than healthy people. Animal toxicological studies, used qualitatively, have provided evidence of mechanisms for acute and chronic effects of ozone, including mortality. Dosimetric studies of humans and animals have helped to establish a linkage for the use of animal data in predicting ozone effects on humans, and have suggested that ozone doses used to induce various tissue injuries and deaths in animals are relevant to the concentrations encountered by human populations.

Combining the information, there is convincing evidence of a significant association between ambient ozone and adverse health effects. Evidence suggests a biologically plausible mechanistic sequence(s), beginning with an inflammatory response which irritates the respiratory tract, giving rise to cough, pain which inhibits inspiration, and bronchoconstriction which reduces airflow. Ozone-induced impaired endogenous defence system (including injury of immune cells and depletion of antioxidants) would render the individual more vulnerable to viral or bacterial infections. These effects and symptoms, if severe enough, could lead to respiratory dysfunction and a requirement for medical intervention such as doctor or emergency room visits, and hospitalization. Although more data are required to fully explain the mortality associations, it is logical to expect that the biological stress related to these effects could exacerbate underlying conditions (e.g. cardiovascular problems) and lead to acute death. The inflammatory portion of the cascade of effects can be present with or without the accompaniment of pulmonary function changes depending on the sensitivity of the individual.

Results from controlled exposure studies of respiratory patients, along with the epidemiological evidence of hospitalizations, emergency room visits, and reduced activity days, suggest that people who are compromised by pre-existing respiratory diseases are more susceptible to ozone exposure.

Controlled human exposure studies indicate exercise is a potent modifying factor in the response. For a given concentration and duration of ozone exposure, the effect is strongly dependent on the level of exercise, because exercise enhances the ventilation rate and consequently the dose delivered to the lower airways. Results from field studies using lunchtime joggers or competitive bicyclists support this finding.

Estimates of population exposure indicate that large numbers of people are exposed to low levels of ozone in Canada (below the current National Ambient Air Quality Objective). Although ambient ozone levels are higher than personal exposure data, they share the same temporal pattern, suggesting that ambient ozone data used in epidemiological studies can be an effective indicator for population exposure. It is also clear that at concentrations currently experienced in Canada, population health effects (mortality and morbidity) are occurring. The increase in unmeasured morbidity (cough, substernal soreness, increased airway reactivity, increased asthmatic attacks, increased medication use) is substantially greater than the measured outcomes from administrative databases (such as hospitalizations, Emergency Department visits).

Newly published data suggest that there likely exist health effects (pulmonary function decrements and induction of new asthma cases) from chronic exposure to ozone. Future development of air quality policy for ozone may require inclusion of an annual or seasonal objective, since this may be more important in much of Canada where chronic low exposure are more prevalent than short peaks.

The acute effects are by definition related to peaks in ozone levels, with the clinical and population health studies correlating responses with hourly or multi-hour (6 - 8 hour) exposures. Available data did not show a substantial difference in the association of health effects with 1-h maximum ozone or 8-hour average ozone levels. This is expected given the high correlation between 1-hr daily maximum and the maximum 6- or 8-hr concentrations.

An averaging time of 24 hours is not considered a best choice of metric because of the strong diurnal pattern exhibited by ozone and the substantial year to year variation in ozone maxima at different sites across the country. Therefore, the averaging time of any target level needs to be less than 12 hours. In view of the results from most Canadian studies, and from controlled human exposure studies showing effects on respiratory symptoms after an exposure (60 ppb) as short as 16-28 minutes (with continuous vigorous exercise), a slight preference is expressed for retention of an 1-hour averaging concentration.

Recommendations

The data show that there is a significant association between ambient ozone concentrations and health effects. These associations have been demonstrated in epidemiological studies, and a causal relationship has been supported through human clinical studies and animal studies. The risks associated with increases in ambient ozone concentrations and health effects such as mortality and hospitalization have been examined in a number of studies covering cities across the world. The regression analyses performed on 13 Canadian cities represent the ozone effects across Canada. The results are similar to those observed in studies of cities across the world. Because the 13-Canadian city study (Appendix A) is, for the time being, the only study that has appropriately established LOAELs for the health effects of the Canadian population in comparison with other approaches, it is recommended that the results from these analyses be used as the basis for determining Reference Levels for ozone (Chapter 13). For risk and benefit analyses, however, it is recommended that the estimates of risk derived from meta-analyses, using studies worldwide, be used, since these studies have adjusted for possible co-pollutant effects. It must be kept in mind that adjusting for co-pollutants using multi-variate models tends to underestimate the risk attributable to ozone, as ozone and other co-pollutants are often statistically correlated.