April 24, 1997
ISSUE: Risk Assessment Aspects of EPA's Proposed Revisions to the National Ambient Air Standard for Ozone (O3)
SPEAKER: Morton Lippmann, Ph. D., Professor of Environmental Medicine
AFFILIATION: New York University Medical Center


1. Academic Peer-Reviewed Research Incorporated into O3 and PM Criteria Documents, which has been supported by the National Institute of Environmental Health Science, the Electric Power Research Institute, the Environmental Protection Agency, and the Health Effects Institute. This includes:

a) Respiratory tract deposition and clearance of airborne particles

b) Controlled human and animal inhalation studies of physiological responses to acidic particles

c) Field studies of population responses to air pollution exposures

d) Development and evaluation of air sampling and monitoring techniques

2. Academic Air Pollution Research Study Advisement

a) Member and Chair of External Advisory Comm., Harvard 6-Cities Study (1978-1987)

b) Member of External Advisory Comm., Harvard - Health Canada - Multi-city Air Pollution Health Effects Study (1987-1991)

c) Chair of External Advisory Comm., USC - CA Air Resources Board Study of Effects of Air Pollution on Children (1992-present)

d) Chair of External Advisory Comm., Yale Univ. -Pierce Foundation Study of Health Effects of Kerosene Space Heater Effluents (1993-present)

3. Federal Agency Service on Committees Focussing on Inhalation Hazards

a) Chair, Clean Air Scientific Advisory Committee (CASAC) (1983-1987)

b) Member, CASAC Subcommittees on O3 (1988-1997) and PM (1993-1997)

c) Chair, Physical Effects Review Subcommittee of Clean Air Act Advisory Council (1994-1997)

d) Chair, EPA Science Advisory Board (SAB) Review Committee for Risk Assessment for Environmental Tobacco Smoke (1991-1993)

e) Chair, SAB Review Committee for Risk Assessment for Dioxin and Related Compounds (1994-1997)

f) Chairman, Indoor Air and Total Human Exposure Advisory

Committee, U. S. Environmental Protection Agency (EPA) , 1987-1993

g) Co-Chair, 4th Task Force for Research Planning in Environmental Health Sciences, National Institute of Environmental Health Sciences (1992)

h) Chair, Board of Scientific Counselors, National Institute for Occupational Safety and Health (1990-1992)

4. International Service on Air Pollution Issues

a) Chair of Working Group on Acute Health Consequences of Winter-type and Summer-type Smog Episodes. World Health Organization-European Region (1990-1991)

b) Member of Working Groups on Air Quality Guidelines. World Health Organization-European Region (1985-1987 and 1994-1997)

5. National Academy of Science Committees

a) Member, Committee on Measurement and Control of Respirable Dust in Mines, National Materials Advisory Board, National Research Council, 1978-1979

b) Member, Committee on Toxicity Data Elements, Board of Toxicology and Environmental Health Hazards, National Research Council, 1980-1983

c) Member, Committee on Methods for the In Vivo Toxicity Testing of Complex Mixtures from the Environment, Board on Toxicology and Environmental Health Hazards, National Research Council, 1985-1987

d) Member, Committee on Research and Peer Review in EPA, Commission on Geosciences, Environment, and Resources, National Research Council, 1994-1997


Controlled Human Exposure Studies

Field Studies of Responses to O3 in Ambient Air

Epidemiological Studies of Large Populations

EPA's Use of the Available Peer-Reviewed Literature

Role of Ozone Formation in Ambient Air on the Health Effects of Fine Particulate Matter




Controlled Human Exposure Studies

The EPA's own Clinical Studies Laboratory in Chapel Hill, NC, has, over the past fifteen years, conducted an extensive and highly meritorious series of studies involving the controlled inhalation exposures of human volunteers to O3 during mild to heavy exercise over time periods ranging from 2 to 8 hours; some studies involved repetitive daily exposures. Most of the volunteer subjects have been healthy young adults, but a few studies have also examined responses in healthy children and adults with chronic respiratory diseases. The studies initially focused on characterizing the extent and distribution of transient changes in respiratory function and symptoms, but most recent studies have also made measurements of other responses more closely associated with adversity, such as increases in the concentrations of inflammatory cells and mediators in the lung airways and increases in the permeability of the airway walls. These studies, involving lung lavage analyses, have shown that inflammation of the lungs is more persistent than increased airflow resistance and reduced lung volumes. Other controlled human O3 exposure studies, performed by other investigators, have produced consistent findings. Thus, the O3 Criteria Document (CD) and Staff Paper (SP) could draw on a rich data base on human responsiveness to short-term exposures of people engaged in exercise to O3 added to a purified air stream. The proposed 8-hr NAAQS of 0. 08 ppm is based largely on this data base.

Field Studies of Responses to O3 in Ambient Air

The O3 CD and SP also cited and discussed a series of field studies in which children in summer camps and adults engaged in outdoor recreational activities had sequential daily measurements of lung function. They showed that lung function was decreased in proportion to the O3 concentration in the ambient air. Most of these studies were performed under my direction with contractual or cooperative agreement support from EPA.

The studies of children and others engaged in outdoor recreational programs, and studies of workers exposed outdoors were not, in my view, given sufficient attention by EPA, as I have previously noted in my review comments to EPA. The EPA chose not to rely on them as much as I believe they should have because of acknowledged uncertainties about the influence of confounding factors always present in such studies. Some of the uncertainties raised about these studies concern whether the camp studies are able to eliminate other environmental factors and air irritants such as weather, dust, pollen that may mask ozone's effect. In our analyses, temperature and humidity were taken into account analytically, and shown not to be important confounders. The studies were performed during periods when allergen exposures were minimal. Exercise was taken into account in general terms, but precise measures were not a practical option. In our own (NYU) studies, we chose camps with relatively vigorous activity programs in order to increase the probability of detecting effects that were occurring.

Another important consideration is that the chamber studies were done under conditions that do not reflect those for the "sensitive populations" or even those that we experience daily. While the recent chamber studies used exercise levels that are generally considered to be well above normal (ventilation rates of 50-68 liters/minute) , they were not necessarily extreme. In a study that my group did on adults engaged in regular outdoor exercise (brisk lunch hour walks or jogging in Tuxedo, NY) the volunteer subjects self-selected their exercise level and duration (average 30 minutes). Ten of thirty subjects chose minute ventilations above 100 liters, and ten others were in the range of 60-100 liters. For those with rates between 60 and 100 liters/min, their response to O3 in ambient air after 1/2 hour were about twice those of the subjects exercising at comparable rates in chambers for alternate 15 minute periods over a span of two hours. Thus, the results of the chamber studies underpredict responses in real life (See: Spektor, D. M., et al., Am. Rev. Respir. Dis. 138: 821-828, 1988). One possible reason for this greater response is that ozone in ambient air is always part of a mixture. Since the other parts of the mixture are not correlatable with the responses characteristic of ozone, then the ozone NAAQS should recognize that the mixture is generally more toxic than ozone alone and that the NAAQS needs to be more restrictive than one that the results of the controlled exposure studies alone would support. In summary, the weaknesses of the chamber studies lies in: 1) small populations; 2) single exposures, which does not correspond to the ambient exposures which people have on a daily basis; 3) use of O3 alone in purified air, which eliminates synergism with other pollutants normally present with O3 in ambient air.

Epidemiological Studies of Large Populations

The O3 Criteria Document (CD) thoroughly summarizes the myriad well-documented health effects that occur in both healthy people and asthmatics as a result of exposure to O3 in ambient air, including pulmonary function deficits, lung inflammation, increased lung permeability and responses to allergic stimuli, altered lung clearance of inhaled particles and increased infectivity of disease agents, increased rates of usage of clinics, emergency rooms and hospital beds for respiratory diseases, and lost-time from work and school. It also discussed equivocal evidence for excess daily mortality on peak O3 days. More recent positive findings in peer-reviewed papers on studies in London, England; Rotterdam and Amsterdam in the Netherlands; and Brishane, Australia increase the likelihood that O3 exposure does indeed cause excess mortality.

The O3 Staff Paper evaluated data from a NYU study of excess daily hospital admissions for asthma in New York City (of which I was a co-author) as a key example of an adverse health effect of O3 exposure. The tabular summary of this analysis in the Staff Paper indicated that the number of asthma admissions attributable to O3 was a relatively small fraction of the total number of year-round asthma admissions. This is certainly true, but ozone is a summertime phenomenon, when other causes of asthma exacerbation are at their lowest levels, and the year-round denominator is therefore not the appropriate divisor. Asthma is a serious and growing problem to millions of people and the health-care community. There is no good evidence that O3 causes new cases of asthma, but clear evidence that it exacerbates the condition in the numerous people who suffer from asthma.

What is not evident from the analyses in the Staff Paper is that the hospital admissions for asthma is not the only, or even the most serious of the adverse impacts of O3 on human health. Rather, it serves as the "lamppost" under which the evidence was most readily visible. My colleague at NYU, Dr. George D. Thurston, has prepared a visual aid, based on his research and research by others, to more fully illustrate the range and magnitude of the health effects attributable to O3 in New York City in each year that could be avoided by implementation of the proposed revision of the O3 NAAQS. It can be seen that the estimate of 265 hospital admissions for asthma is near the tip of the "iceberg", along with 240 other hospital admissions (for other pulmonary diseases) , 75 cardiopulmonary deaths, and 3,500 emergency room visits. It also can be seen that the total impact extends to millions of excess symptoms and disease incidences.

Most of the epidemiological scientific studies that are relevant to the setting of NAAQS were not designed or performed with that specific application in mind. EPA-supported epidemiological research has been far too limited in scope, nature and extent to provide a data base for standard setting. The fact is that, because of limited research resources, constantly shifting research priorities, and a long-term policy choice to have only a minimal in-house capability for epidemiologic research, the quite limited amount of research most relevant to the chronic health effects of O3 has been performed by academic investigators with resources provided by others such as, for example, the National Institutes of Health, the Health Effects Institute, the Electric Power Research Institute, Health-Canada, and the California Air Resources Board. One result of this welter of diverse sponsorship, and therefore of research goals, is a wealth of information that is, unfortunately, composed of bits and pieces of the overall puzzle. It requires careful sifting to separate those elements of sufficient quality to inform the issues, as well as mature judgment to fit the pieces into an informative framework sturdy enough for summary judgments.

Role of Ozone Formation in Ambient Air on the Health Effects of Fine Particulate Matter

The main reason for linkage between the O3 and PM NAAQS lies in their future implementation rather than in setting their NAAQS. A large part of PM-2.5 is created in the same photochemical reaction sequence that leads to O3 formation. Furthermore, a major part of the rest of the PM-2.5 is acidic aerosol whose formation depends in major part on the photochemical oxidants role in oxidizing SO2 and NOx vapors to form acidic fine particles. Thus, PM-2.5 cannot be effectively controlled without controlling the formation of ozone, and tighter restrictions on ambient O3 concentrations will result in lower exposures to PM-2.5 and the health effects caused by PM-2.5 exposures.

EPA's Use of the Available Peer-Reviewed Literature

Despite some critical remarks that I have made above, I want to acknowledge the incredibly careful sifting of the evidence performed by EPA's National Center for Environmental Assessment (NCEA). The oversight and prodding of CASAC has helped to ensure that essentially all of the relevant peer-reviewed science was examined in detail and appropriately summarized and interpreted in the final draft of the Criteria Document (CD). The corresponding public review sessions by CASAC of the Staff Paper (SP) drafts prepared by EPA's Office of Air Quality Planning and Standards (OAQPS) also ensured that the final draft of that document provided appropriate summary judgements on the scientific aspects of those items in the CD most relevant to the setting of the NAAQS. These items include the effects of concern, populations at special risk, optimal form and averaging times for the NAAQS, and the most likely residual effects associated with exposures to be expected across a possible range of concentration limits. This process thus provided the Administrator with the best possible basis for the difficult NAAQS decisions that are required to be made periodically under mandate of the CAA amendments of 1977.

There has never been a decision point where the Staff, the CASAC, or the Administrator has been satisfied with the available scientific data base, despite the ever increasing size and sophistication of the available data in successive review rounds. Our current knowledge always leads to new questions and concerns. More so than in the past, we are debating whether the measurable effects are sufficiently adverse to warrant public health protection rather than identifying whether measurable effects are occurring. However, the Administrator will still have, and will, I suspect, always have to make a judgment as to the margin of safety to apply in the absence of definitive knowledge.

Despite all of their thoroughly discussed and acknowledged limitations, both the PM and O3 literature reviews and analyses in the CDs and SPs are the best prepared and most comprehensive ever available to an Administrator as a basis for NAAQS decisions. In fact, the favorable contrast of these CDs and SPs with those from prior rounds of NAAQS is really remarkable.


While the present research base provides sound support for the current EPA NAAQS proposals, more can be learned that can aid in the most efficacious implementation of the new standards during the next decade, as well as to provide a basis for still better focussed NAAQS in the next round.

Ozone Health Effects

In terms of research needs for O3, the following are my personal recommendations for health effects research based upon my own research experience and service on CASAC panels.

The most pressing need is for research on the cumulative effects of O3 on lung development in children and on accelerated aging of lung structures that may shorten life-span in adults. We have a lot of data on transient functional effects of O3 from controlled human exposure studies. Such studies can provide information on chronic pollutant effects only to the extent that prior exposures affect the transient response to single-exposure challenges. Most of the limited data we have on the effects of chronic O3 exposures on humans come from epidemiological studies. Epidemiological studies can establish chronic health effects of long-term O3 exposure in relevant populations, and offer the possibility that the analyses can show the influence of other environmental factors on responses to O3 exposure. We know, for O3, for chronic exposures of rats and monkeys, that cumulative changes in lung structure occur, and can these can be described as excess stiffening and/or premature aging. Some lung autopsy research of sudden accident victims in Los Angeles County suggests that similar effects are occurring in humans.

We also need more controlled exposure studies focussed on the mechanisms and patterns of response to inhaled O3 and of the influence of other pollutants and stresses on these responses. Studies of the transient responses to acute exposures can establish the interspecies differences in response among animal species, and between them and humans similarly exposed. Animals are needed for studies of responses that require highly invasive procedures or serial sacrifice to gain information that cannot be obtained from studies on human volunteers. Finally, we should use long-term exposure protocols in animals to study cumulative responses and the pathogenesis of chronic disease in animals. Studies on animals can examine the presence and basis for variations in response that are related to age, sex, species, strain, genetic markers, nutrition, the presence of other pollutants, etc.

Research is also needed to establish the interrelationships between small transient functional decrements, which may not in themselves be adverse effects, and changes in symptoms, performance, reactivity, permeability, and counts of inflammatory cells. The latter may be closely associated with adversity in themselves, or in the accumulation or progression of chronic lung damage.

Chronic human exposures to ambient air appear to produce a functional adaptation that persists for at least a few months after the end of the O3 season but dissipates by the following spring. Several population-based studies of lung function have indicated an accelerated aging of the lung associated with living in communities with persistently elevated ambient O3. The plausibility of accelerated aging of the human lung from chronic O3 exposure is greatly enhanced by the results of chronic animal exposure studies in rats and monkeys. There is little reason to expect humans to be less sensitive. Humans have a greater dosage delivered to the respiratory acinus than do rats for the same exposures. Also, the rat and monkey exposures were to confined animals with little opportunity for heavy exercise. Thus, humans who are active outdoors during the warmer months may have greater effective O3 exposures than the test animals. Finally, humans are exposed to O3 in ambient mixtures. The potentiation of the characteristic O3 responses by other ambient air constituents seen in short-term exposure studies in humans and animals may also contribute toward the accumulation of chronic lung damage from long-term exposures to ambient air containing O3.

In summary, the lack of a more definitive data base on the chronic effects of ambient O3 exposures on humans is a serious failing that must be addressed with a long-term research program.


1. Recognize that EPA Administrator has made a prudent public health judgment in her O3 NAAQS selection. The current NAAQS of a 1-hr max of 120 ppb, not be exceeded more than 4 times in 3 yrs, is equivalent to an 8-hr max of 90 ppb based on the 3rd highest 8-hr value in a year. Thus, the proposed 8-hr max of 80 ppb is only a modest O3 NAAQS reduction. By contrast, the Air Quality Guideline for O3 of the World Health Organization-European Region (WHO-EURO) , adopted late in 1996 is an 8-hr maximum of 60 ppb. In my view, the 8 hr-80 ppb proposal is a prudent step in the right direction at this time and recognizes that any lower limit is probably not achievable without draconian controls. The major advance is the shift to an 8-hr averaging time, providing a much sounder basis for evaluating the public health risk from community exposures.

2. Recognize that while the costs of the research recommended above are substantial, they are quite small in relation to the control costs that can be more effectively targeted an

d reduced through the knowledge gained, and also small in comparison to the health benefits resulting from exposure reductions to O3 and PM-2.5 resulting from the implementation of the revised O3 NAAQS.