Skip to Navigation
Skip to Content

Support Greater Boston PSR!

Join us in building a healthy environment and promoting sensible security policies. Make a donation to GBPSR today.

Donate Now »

Take Action

Tell Congress you won’t support phony chemicals policy reform -- only real, health-protective reform.

Health Effects/New Data: 1996-2000

Author: Jefferson H. Dickey MD

Health Effects/New Data: 1996-2000

New Data / Pulmonary Function / Chamber

A new model has been published which seems to better predict the mean decrease in FEV1 as a function of ozone concentration, minute ventilation, duration of exposure, and age. We currently know that the magnitude of the FEV1 decline increases with ozone concentration, minute ventilation, and duration of exposure, and decreases with age. McDonnell, et al19 combined the results of a series of chamber - spirometry studies performed at the US EPA Clinical Research Facility over a 13 year period on 485 non-smoking healthy subjects. Ozone concentrations ranged from 0 to 400 ppb; ventilation rates varied from about 5 L/min/m2BSA (body surface area) at rest to about 40 L/min/m2BSA. The latter represents a moderately heavy workload. Spirometry was performed at zero, 1 and 2 hours after exposure began.

This model is quite useful within its constraints. Exposures are limited to 2 hours. Exposure to persons working or exercising outdoors in the community are commonly 6 or 8 hours. The subjects were healthy young men. Sensitive subgroups, especially those with asthma, may be more vulnerable. Nonetheless, this model is probably the most useful data set for predicting FEV1 decline as a result of relatively short duration of ozone exposures.

A recent study confirmed previous observations that ozone induced declines in FEV1 do not correlate with development of bronchial hyperreactivity. In chamber exposures to 400 ppb for 3 hours per day on 5 consecutive days at moderate exercise (about 32 L/min), asthmatic subjects demonstrated attenuated FEV1 decline after several days of exposure. However, the ozone caused bronchial hyper responsiveness which did not attenuate.20

Another study of elderly subjects with COPD were much more susceptible to ozone induced declines in FEV1 and increased airway resistance than subjects without COPD. The COPD subgroup decline in FEV1 after 4 hours exposure to 240 ppb ozone with intermittent light exercise was 19% compared with 2% in the healthy controls. About half of this effect was due to ozone and half to the effect of exercise in COPD.21

One study of ozone effects on cardiac parameters found that 300 ppb ozone for 3 hours with intermittent exercise resulted in an increase in cardiac rate and heart rate * blood pressure product, but otherwise no major effects on right heart catheterization cardiac parameters.22

One study exposed 41 subjects aged 9-12 years, both with and without asthma, to a combination of 100 ppb ozone, 0.10 ppm SO2, and 100 mcg/m3 H2SO4 for 4 hours with intermittent exercise. The subjects responded with changes in spirometry, symptoms, and overall discomfort level, but the results were not statistically significant.23

Overall, the chamber spirometry studies continue to find decrements in pulmonary function as a result of ozone exposure. The magnitude of the decrements are relatively small or are observed after prolonged exposure. The clinical implications for pulmonary function declines, hence, continue to be limited to sensitive subpopulations - those with underlying lung disease and those with prolonged outdoor exposures with elevated ventilation rates (children playing outdoors and adults working or exercising outdoors).

New Data: 1996-2000 / BAL / Ozone Chamber Studies

A handful of studies address the effects of air pollution on bronchoalveolar lavage measures of inflammation and its consequences. Some of the more interesting ones are summarized here.

In an attempt to better understand interindividual variability, a chamber study of healthy subjects exposed to 220 ppb of ozone for 4 hours with exercise was performed. The expected declines in FEV1 occurred with interindividual variability - some persons experienced a greater than 15% decline in FEV1 (designated responders) and some experienced a less than 5% decline (non-responders). Increases in BAL polymorphonuclear leukocytes, and interleukins 6 and 8 appeared early, and lymphocytes, mast cells, and eosinophils increased later in all groups, regardless of smoker or FEV1 responder status.24

A similar result occurred in a study which exposed asthmatic and healthy individuals to 125 and 250 ppb ozone over 3 hours of intermittent exercise. The magnitude of the inflammatory response measured on BAL was consistent within individuals, and the decrement in FEV1 was highly reproducible, but the BAL inflammation was not highly correlated with the decrement in FEV1.25

These studies help clarify the poor correlation between lung inflammation and declines in lung function.

Previous work has shown that the decline in FEV1 attenuates after several days of exposure to ozone, although the bronchial hyperreactivity tends to persist. A newer study finds that when subjects are repeatedly exposed to ozone (200 ppb over 4 hours over 4 days), significant decreases in the number of PMNs, fibronectin, and IL-6 were found after 4-d exposure versus single-day exposure.26

One study of dust mite allergic asthmatics found ozone exposure (160 ppb) to be associated with significant increases of eosinophils in the BAL fluid.27 Another study compared BAL of asthmatic subjects with that of normal subjects after exposure to ozone (200 ppb over 4 hours). The asthmatic subjects showed significantly greater O3-induced increases in several inflammatory endpoints (percent neutrophils and total protein concentration) in BAL as compared with normal subjects.28

Substance P (a neurotransmitter with many functions, including nocioception) has been shown to be observed in greater amounts in BAL fluid in larger amounts after ozone exposure.29

In sum, although the exposures above are generally higher than commonly encountered in the community, data continue to accrue that ozone induces bronchioalveolar inflammation, that asthmatics may have a heightened response, and that ozone induced inflammation observed on BAL does not correlate well with declines in FEV1.

New Data: 1996-2000 / BAL / Community Air Pollution

One of the more interesting BAL studies examined the effects of community air pollution on recreational joggers in New York City.30 15 subjects were examined during summer (high ozone period) and retested during winter (low ozone period). Release of reactive oxygen species was lower in the summer than the winter. In contrast, LDH, IL-8, and PGE2 levels were all roughly two fold higher in summer. These results suggest a possible ongoing inflammatory response in the lungs of recreational joggers exposed to ozone and associated co-pollutants during the summer months, and that the inflammatory response observed during controlled chamber exposures seems to be occurring during community exposures.

New Data: 1996-2000 / Chamber Studies / Allergen - Pollutant Interaction

One of the most interesting areas of current research concerns the potential for ozone to exacerbate asthmatic allergen induced bronchoconstriction. Previous work had discovered that ozone at higher doses increased asthmatic allergen sensitivity, but it was controversial whether lower ozone doses had the same effect. The current data suggest that 1 hour exposure of mild asthmatics at rest to 120 ppb, or with intermittent moderate exercise to 100 ppb ozone, does not enhance allergen sensitivity, but 3 hours exposure to 200 ppb with intermittent moderate exercise does. Unanswered at this time is whether asthmatics with more severe disease, or exposed to lower levels for longer periods or with higher activity levels (minute ventilation rates), would demonstrate ozone induced allergen hypersensitivity.31-33

Unusually high exposures to NO2 (with our without SO2) appear to increase asthmatic sensitivity to allergen as well.32, 34-36

Although this data suggests that air pollutants at levels uncommonly encountered in most areas of the USA will enhance asthmatic allergen sensitivity, at lower levels which are commonly encountered, the pollutants do not consistently enhance asthmatic allergen sensitivity.

New Data: 1996-2000 / PEFR - Time Series Diary

I identified 24 studies which examined the relationship between air pollutants and asthma exacerbations and declines in peak flow rates. All studies except one demonstrated significant associations between air pollution levels and one of the health effects. Generally, studies which failed to find associations with drops in peak flow rates found increases in asthma medication use.37-60

New Data: 1996-2000 / PEFR - Time Series Diary / Combined Analysis

I found 2 studies which performed combined analysis of time series peak flow studies in children. One61 looked at 5 previously performed studies of peak flow decrements as a function of PM10 particulate air pollution levels. They found a particulate to be associated with lower average peak flow rates and a higher prevalence of significant drops in peak flow rates.

The other study62 re-analyzed six studies which examined the effect of ozone air pollution on children playing outside at summer camps in New Jersey, New York, Ontario, and southern California. All of the studies found ozone to be associated with drops in FEV1. Combining the data, the authors estimate a decline of 0.5 ml of FEV1 per ppb O3. This is a surprisingly large, clinically important result, and suggests that chamber exposure studies are underestimating the effect of exposure in the community.

New Data: 1996-2000 / Pulmonary Function / Cross Sectional

I identified 7 studies which examined the association between air pollution and cross sectional measures of lung function. All found significant associations between air pollution levels and lower lung function. This type of study is generally much less likely to differentiate between the effects of different types of pollutants unless very large numbers of cities are examined.

Abbey et al63 examined a group of 1,391 non-smokers and found that PM10, ozone, and SO2 all had significant associations with lower lung function. Interestingly, persons with a family history of lung disease (asthma, bronchitis, emphysema, or hay fever) may be much more vulnerable to the air pollution effects.

Peters et al64 studied 3,293 school children from 12 southern California communities with different air pollution levels. Significant associations were found between measures of lung function (FEV1 and FVC) and PM10, PM2.5, NO2, and O3 in girls, and O3 in boys. Both effects were stronger when stratified according to the amount of time the child spent outdoors. This study is forming the basis of a prospective cohort, and follow up data will be available in a few years.

In the largest study of its kind, Raizenne et al65 examined 10,251 children from 24 US and Canadian cities and found that acid aerosols were associated with about a 3% lower lung function (FEV1 and FVC).

Two studies examined the association between bronchial hyperreactivity and air pollution. Both found bronchial hyperreactivity to be associated with air pollution.66, 67

Other studies found significant associations between air pollution and cross sectional lung fucntion decrements.68, 69

New Data: 1996-2000 / Pulmonary Function / Prospective Cohort

I found one study70 which examined the relationship between chronic air pollution and lung growth in children. This study followed 1,150 children prospectively for 3 years, performing spirometry at the beginning and end of each summer. Air pollution in the various communities was evaluated for PM10, SO2, NO2, and O3. Adjusting for a child's sex, atopy, passive smoking, baseline lung function, and increase in height, the researchers found that summertime ozone was associated with a lesser summertime children’s lung growth as reflected by a lower than expected increase in FEV(1) and FVC. PM10, SO2, and NO2 did not show this association. One study should never be considered to have proven anything, and this is no exception. Alternative explanations are not excluded; however, this study suggests that ozone air pollution may impair normal lung growth in children.

New Data: 1996-2000 / Morbidity / Prospective Cohort

Data continue to be generated by one of the most important ongoing prospective cohort projects - the Seventh Day Adventists study. This study is following a large cohort of a non-smoking population and associating local air pollution levels with health outcomes. In this analysis, 3,091 non-smoking subjects were followed for 15 years and about 3.5% of them reported a new diagnosis of asthma.71 The authors report that, regardless of adjusting for other air pollutants, the asthma rates were roughly doubled in the men exposed to the higher mean ozone levels. This effect was not observed in women. The analysis was not adjusted for the percentage of time working outdoors. While this analysis cannot be considered to have proven that ozone causes asthma, because the results are consistent with expectations generated from pathophysiologic studies, it raises important questions about whether ozone causes asthma. Unfortunately, this study is going to be difficult to replicate because of the unique population, large group size, and long duration required.

New Data: 1996-2000 / Cardiac Monitor / Time Series / Particulate

Many of the epidemiologic studies examining the cause of death in persons dying on high particulate air pollution days

I found 2 studies72, 73 which examined cardiac rate (24 hour holter monitor) in association with particulate air pollution levels. Both studies found elevated cardiac rates on days with elevated particulate levels. One of these studies additionally examined heart rate variability and found particulate to be associated with decreased overall heart rate variability.


I identified 14 studies in the recent medical literature which examined the association between air pollution and acute physician consultations or emergency room visits. Eight of these specifically examined asthma or acute wheezy episodes; 3 additional studies examined respiratory complaints generally; 1 each reviewed doctors house calls in Paris, France and childrens’ ER visits in Santiago, Chile. One study looked at ER visits during a series of bush fires in Sidney, Australia. I could not obtain the results of one study.74

Only the study of Australian bush fires75, and another in Switzerland76 (which admitted to potentially poor exposure assessment) failed to find significant associations.

One important study examined ER visits in one summer for respiratory disease in Montreal. Ozone, PM10, PM2.5 and sulfate were all associated with emergency room use for persons over the age of 65. A 36% increase in ozone levels was associated with a 21% increase in ER visits for respiratory complaints. In examining the effects of particulate and acid aerosols, the relative mass effects were PM2.5 > PM10 > SO4.77

The association between asthma ER visits and particulate was observed even when air pollution levels were below the new NAAQS for PM2.5. In this study, a moderate increase in air pollution (11 microg/m3 in fine PM) was associated with a 15% increase in the rate of ER visits.78

All the other studies also found consistent associations between air pollution levels and emergency room visits and respiratory disease.79-87

In all, 11 of the 13 studies for which results were available found significant associations between air pollution and emergency room use or acute physician consultations.


Two recent prospective cohort studies of asthmatics examined the relationship between air pollution levels and emergency room visits (a relatively new epidemiologic technique). One88 found a significant association between acid aerosol fog (a complex mixture of pollutants) and ER visits for asthma; the other89 found significant air pollution associations, with air pollution (NOx, SO2, and ozone) plus weather accounting for 69% of the variance.


I identified 19 studies which examined the association between air pollution and daily hospital admissions. Ozone and particulate air pollution were robustly associated with hospital admissions. One additional study examined the relationship between characteristic of air masses and asthma hospital admissions. It found that during the spring and summer, air masses with high air pollution levels were more likely to be associated with increased asthma admissions.

Nineteen90-108 new studies examined the association between particulates and hospitalization rates. Only one98 of these studies failed to find an association between combustion derived air pollution and increased hospitalization rates. Two97, 104 of the studies found the association with SO2 instead, and one found the association with particulate only when NOx was also in the model. Hence, in 18 of 19 studies, the association with particulate air pollution or its chemical antecedents was observed.

Fifteen new studies examined the association between ozone air pollution and hospitalization rates. Twelve90-94, 96, 97, 99, 102, 103, 107, 108 studies found a significant association, and one104 study found the association with NO2 instead. Two95, 98 studies failed to find the association. In all, 13 of 15 studies found the association with ozone or its chemical antecedents.

Outcomes which were commonly associated with air pollution included asthma, COPD (chronic obstructive pulmonary disease = chronic bronchitis and emphysema), and heart disease.

In summary, ozone and particulate air pollution were robustly associated with hospital admission rates. Common outcomes were asthma, COPD, and heart disease.


In late 1995 and early 1996 most of the major review articles emerged examining the relationship between air pollution and adverse health effects. These reviews found consistent and coherent associations between particulate air pollution and daily mortality levels. In addition, data was beginning to emerge suggesting that ozone air pollution was associated with daily mortality. While some authors considered the association between particulate air pollution and daily mortality to be causal, others were more circumspect. Hence, I examined the subsequent data in search of validation or refutation of these concerns.

I was able to identify 21 time series studies published since 1996 examining the association between daily air pollutant levels and daily mortality. These studies generally are quite well designed, adjusting for weather variables and other considerations.

Eighteen14, 105, 106, 109-123 new studies examined the association between particulate air pollution and excess mortality. One120 of these, a study of asthma mortality, failed to find the association.

Twelve new studies examined the association between ozone air pollution and excess mortality. Ten92, 110, 111, 113, 115, 118, 120-122, 124 found the association and one114 found the association with NO2 instead. One study109 found ozone to be marginally associated. No study completely failed to find an association with ozone or its chemical antecedent, NO2.

One120 study specifically examined the relationship between asthma mortality and air pollution. Only NO2 and ozone air pollution were found to be associated with asthma mortality.

I identified 1 study which attempted to differentiate the fine fraction (PM2.5) of particulate air pollution from the coarse fraction (PM2.5 - PM10).14 This study found the fine fraction to be implicated in elevated mortality rates. One106 studies examined acid sulfate aerosol, a common constituent of particulate fine fraction. Those studies found sulfate aerosol to be strongly associated with mortality. (Note is made that previous studies have shown that sulfate aerosol is sufficient, but not necessary, to be associated with mortality).12

Time series studies have continued to flood the medical journal market in the last few years, and continue to overwhelmingly find that particulate air pollution is associated with mortality. Data increasingly suggest that the fine fraction, which generally arises from combustion sources, is consistently implicated. In addition, a remarkably robust data set is emerging associating high ozone exposure with daily mortality. Although some studies find associations between SO2 and NO2 and daily mortality, these association are less consistent.


Several meta-analyses of the association between particulate air pollution and daily mortality have appeared previously in the literature. All have found significant associations between particulate air pollution and daily mortality. I have found one meta-analysis published in the recent literature.

This meta-analysis125 examined the effect of between study variability on the effect estimate of the association between particulate and daily mortality. More precisely, they examined the possible effects of air pollution patterns and characteristics of the exposed population. There was some evidence that PM effects were influenced by climate, housing characteristics, demographics, and the presence of sulfur dioxide and ozone. However, the effect of particulate on mortality was robust, not changing with inclusion of potential confounders and effect modifiers. The increase in daily mortality rate of 0.7% per 10 mcg/m3 increase in PM10 is similar to previous meta-analysis estimates (1% per 10 mcg/m3 increase in PM10), and was found to be higher in locations which had a higher proportion of PM10 attributed to the fine fraction (PM10). In other words, the effect estimate of this meta-analysis was quite consistent with previous meta-analyses, was able to adjust for the presence of many potential confounders and effect modifiers, and supported evidence from other studies which implicates the fine fraction of particulate in excess mortality.


Three large prospective cohort studies examining the relationship between particulate air pollution and premature mortality were conducted in the last decade.13, 16, 126 Prospective cohort studies are considered the most reliable study possible in air pollution epidemiology because individual subjects are identified and individual risk factors for mortality (such as smoking) are considered and adjusted for. All three prospective cohort studies found significant associations between particulate air pollution levels and premature mortality. Interestingly, all three studies also found associations between air pollution and lung cancer.

The most recent prospective cohort was published in 1999.126 This is a cohort of 6,338 nonsmoking Seventh-day Adventists. PM10 was strongly associated with non-malignant respiratory mortality adjusting for a wide range of potentially confounding factors, including occupational and indoor sources of air pollutants. The mortality rate was 18% higher for persons exposed to 43 days per year with PM10 levels higher than 100 mcg/m3. Both ozone and PM10 were associated with lung cancer in males, and sulfur dioxide showed strong associations with lung cancer in both sexes. Other pollutants showed weak or no associations with mortality.


The new data on health effects or air pollution has been reviewed above. One substantial new review of particulate air pollution has recently been published by the National Institutes of Health127 in their academic journal, Environmental Health Perspectives. This new review concludes that the case for adverse health effects from particulate air pollution has been made, and that the effect should be considered causal.

"The question of when it would be appropriate to conclude that the associations between particulate pollution and various outcomes (including mortality) should be judged as causal in nature has been difficult and controversial. Although such a judgment must be subject to revision, the volume of new information and new experimental findings has been so great that such a reevaluation is required at frequent intervals. The useful summary by Gamble [PM2.5 and Mortality in Long-Term Prospective Cohort Studies: Cause-Effect or Statistical Associations? Environ Health Perspect 106:535-554 (1998)] of the reasons why a causal inference was, in his opinion, not justified provides a basis for reevaluation in the light of new data. Such a reexamination indicates that the associative evidence is now stronger and that the biologic basis for a number of adverse effects has now been demonstrated. All of the useful guideline criteria customarily applied to such questions seem to have been met, although there is still much to be learned about interactive effects and the possibility of statistical thresholds."

NEW DATA: 1996-2000 / COSTS

A recent review has examined the health based financial benefit of reducing particulate air pollution in the USA.128

"Most Americans are exposed daily to airborne particulate matter (PM), a pollutant regulated by the U.S. Environmental Protection Agency. Current national standards are set for PM10 (particles less than 10 microns in diameter) and new standards have been promulgated for PM2.5 (particles less than 2.5 microns in diameter). Both particle sizes have been associated with mortality and morbidity in studies in the United States and elsewhere and an unambiguously safe level of ambient PM has been difficult to identify. PM10 concentrations have been reduced significantly in U.S. cities over the past two decades and relatively few locations continue to exceed national PM10 standards. However, the new PM2.5 standards will require further reductions in PM concentrations and additional expenditures for emission controls. Information about the health and economic benefits of achieving lower PM concentrations is important because: (1) expected costs of further PM reductions rise after the least-cost options are exhausted, and (2) there is uncertainty about the existence of a threshold safe level for PM. This paper develops and applies a methodology for quantifying the health benefits of potential reductions in ambient PM. Although uncertainties exist about several components of the methodology, the results indicate that the annual nationwide health benefits of achieving the new standards for PM2.5 relative to 1994-1996 ambient concentrations are likely to be between $14 billion and $55 billion annually, with a mean estimate of $32 billion."


Finally, lest one doubt that the public health community is behind curtailing air pollution as a needed public health measure, a journal of the American Public Health Association has published a recent review.129

"The connection between energy policy and increased levels of respiratory and cardiopulmonary disease has become clearer in the past few years. People living in cities with high levels of pollution have a higher risk of mortality than those living in less polluted cities. The pollutants most directly linked to increased morbidity and mortality include ozone, particulates, carbon monoxide, sulfur dioxide, volatile organic compounds, and oxides of nitrogen. Energy-related emissions generate the vast majority of these polluting chemicals. Technologies to prevent pollution in the transportation, manufacturing, building, and utility sectors can significantly reduce these emissions while reducing the energy bills of consumers and businesses. In short, clean energy technologies represent a very cost-effective investment in public health."

More . . . Summary and Conclusions