The Committee has requested that I provide my views on the ozone and particulate matter standards, which EPA has published in the Federal Register and intends to implement under the Clean Air Act. I am pleased to do that and would also like to extend my testimony to include the research effort of EPA because it directly affects the standard-setting process. I understand that the two standards present different problems in terms of the form of the standard, the scientific data supporting each standard and the process by which the standard was promulgated. In my view, however, there are similarities between the two standards that reflect a major deficiency in EPA's efforts. The common deficiency is the lack of solid scientific data. EPA is a grossly underfunded agency given the scope of its responsibilities. EPA has not done well with its resources by not sustaining research to meet the long-term goals of the agency. Thus, I hope that the Committee will allow me to express my concerns about the research planning at EPA.
Air Pollution Is A Major Long Term Public Health Problem
Air pollution is a worldwide problem. In the United States air pollution is of such public health importance that it is critical that a national debate be undertaken on the future directions of air pollution research and regulation. This Committee is providing a very valuable forum to the people so that they may learn more about the scientific controversy surrounding these two air pollutants and the alternative views that exist concerning the future of air pollution remediation efforts. I am at the moment writing a review of the toxicology of ozone \1\. This will be the third review of ozone that I have written for the scientific literature. Almost ten years have elapsed since my last effort, and I was surprised and saddened to note on examining the literature that questions which we raised in the review in 1988 still remain unresolved. Much new human data has become available on ozone supporting a lower standard and shorter averaging time, but the book is far from closed on ozone. I also wrote the first part of the health section of the SOx (sulfur oxides) Particulate Matter Criteria Document for EPA in 1980. Many of the questions raised in that document also remain unanswered. As a consultant to the Clean Air Scientific Advisory Committee I assisted in the review of the current Particulate Matter Criteria Document. Not only were the fundamental questions raised in the original SOx Particulate Matter Criteria Document still existent, but new important questions arose for which we have no answer. All of these experiences suggest to me that a greatly enhanced and invigorated research effort in air pollution is needed if we are to make sound, reasonable and rational decisions on the implementation of clean air standards. If anything, air pollution research is now more important to the national public health than ever before.
\1\Shoaf C.R. and Menzel, D.B. Oxidative damage and toxicity of environmental pollutants. In: Cellular Antioxidant Defense Mechanisms. (ed., C. K. Chow) CRC Press, Inc, Vol. 1:197-213 1988
Both the ozone and particulate matter standards have vast implications for the quality of life and the economy of the United States. It is my opinion that the vast majority of Americans support improving and enhancing the quality of their life by eliminating or decreasing air pollution. Americans are quite willing to shoulder the burden of cleaner air, cleaner water, and cleaner food if they can understand clearly the benefits to be gained by these activities. The confidence of the American people in the decisions being made on environmental issues is critical to the ability of this government to govern and implement these decisions. If ever the public loses confidence in the environmental strategies promulgated by the Federal Government then it will be impossible to carry out large national programs designed to eliminate or at least ameliorate the adverse effects of air pollution. I am very concerned that the Environmental Protection Agency and the Congress maintain the confidence of the U.S. public and demonstrate to the public their vigorous support for a better quality of life and clean air. Scientific truth is the only lasting commodity upon which decisions can be based.
From my view the difficulties that we face with both the ozone and the particulate matter standard stem from generic issues in toxicology which must be addressed in a sound scientific manner. The first of these generic issues is a plausible biological mechanism of action for the particular pollutant. The second is the nature of the dose response relationship. I will address each of these and give examples of how they impinge upon the two standards that we are discussing today.
Plausible Biological Mechanisms
What is a plausible mechanism? We have learned a great deal about the quantitative nature of toxic reactions in the last 40 years. It is now possible to divide biological reactions to toxicants into several categories under which plausible mechanisms have been elucidated. A plausible mechanism of action for a toxin places the toxin within the context of our knowledge of disease processes. Having a plausible mechanism of action increases our confidence that health effects observed in animals will occur in humans. Understanding a mechanism of action also makes experiments more meaningful and relevant. In this forum it is not possible for me to elaborate in greater technical detail on how a plausible mechanism influences the experimental design and interpretation of the results of experiments. Experimental design and the concept of plausible mechanism of action are dealt with in standard textbooks of toxicology, such as "Casserett and Doull's Fundamentals of Toxicology".
A plausible mechanism of action is critically essential to controlled human exposure studies. The extrapolation from animal experiments to human exposures as they occur in nature, that is with free-living people, depends upon an intermediate link of controlled exposures of human volunteers to the toxin. We must have a clear idea of a plausible mechanism so that human studies can be developed with due care that no harm will ever result to the volunteers who courageously commit themselves to these kinds of experiments. In air pollution many of the human studies have been very limited because of the lack of a clear understanding of a plausible mechanism. Investigators have been very reluctant to engage in high level exposures of human subjects because they fear that some long-term harm will result from their experiments. Clearly, we cannot and will not tolerate human experimental studies that result in harm to the volunteer. This is simply not ethically acceptable.
Plausible Mechanism of Ozone Toxicity
One plausible mechanism of action of ozone is the production of free radicals by the reaction of ozone with cellular constituents. The free radical theory is that which we proposed in 1971\2\. It is now clear that this mechanism of action is too naive and simplistic and clearly does not explain the consequences of chronic exposure to ozone. Studies with experimental animals clearly show that the results of a continuous or intermittent lifetime exposure to ozone are highly complex and are not predictable from the free radical hypothesis alone. Further experiments are needed with life-term exposures of experimental animals using the most modern molecular biology techniques. The complex pattern of lifetime ozone exposure must involve multiple signal transduction pathways. Simply put, the adverse health effects of chronic exposure to ozone are complex and beyond the free radical theory which we now recognize as accounting for the brief initial contact of ozone with the lung.
\2\Roehm, J.N., Hadley, J.O. and Menzel, D.B. Oxidation of Unsaturated Fatty Acids by Ozone and Nitrogen Dioxide: A Common Mechanism of Action. Arch. Environ. Health, 23:142-148 1971.
Chronic exposure is the critical issue in ozone exposure. EPA initiated and was carrying out an excellently conceived and implemented research program on the chronic effects of ozone in support of the current ozone standard. But this research has stopped and support for ozone research by other Federal agencies has stalled. Basic research support for ozone by the National Institutes of Health and particularly the National Institute of Environmental Health Sciences (NIEHS), has fallen away. The scientific community is in error in allowing this to have happened.
Very compelling controlled human exposure experiments suggest that the current ozone standard (0.12 ppm) may be toxic. The short term exposures under which humans can be safely exposed does not allow us to study the chronic effects of ozone exposure. Epidemiologic studies are underway in the South Coast Air Basin, particularly those by Professor John Peters of the University of Southern California but this study is hampered because no quantitative biomarker of ozone health effects has been developed.
We would not be sitting here and engaging in this discussion if EPA's chronic ozone study in experimental animals had been carried out. Nor would we still have doubts about the ozone standard if ozone research had received a high priority in research support by the other Federal research agencies such as NIH and NSF.
In summary, there is a preliminary biologically plausible mechanism of action for ozone. The free radical theory is not comprehensive and does not explain all of the effects of chronic exposure to ozone. Much additional work is needed to understand the chronic effects of ozone.
In contrast to the ozone problem, no plausible biological mechanism of action has so far been proposed for particulate matter. It has been very difficult to demonstrate toxicity for particulate matter in experimental animals. In my laboratory and that of my colleagues at UCI we have not been able to show major toxicity with particulate matter at potencies approaching the levels reported from epidemologic studies \3\ \4\.
\3\ Saldiva, P. H., Pope, C. A., Schwartz, J., Dockery, D. W., Lichtenfels, A. J., Salge, J. M., Barone, I. & Bohm, G. M. (1995) Air pollution and mortality in elderly people: a time-series study in Sao Paulo, Brazil. Arch. Envfron. Health 50:159-163.
\4\Schwartz, J. (1995) Short term fluctuations in air pollution and hospital admissions of the elderly for respiratory disease. Thorax 50: 531-538.
To place this problem in a more global context, urban particulate matter is a universal problem. Particulate matter seems to be a common result of human concentration in urban areas. To eliminate all of the particulate matter in our cities would, in my view, be only possible by the elimination of all human activity. Clearly this Draconian approach is not reasonable.
The studies of Schwartz and his colleagues\3\,\4\ have challenged our conclusions from experimental animal studies. These studies indicate that all particles regardless of their geographic origin have the same toxicity. It is well known that the chemical composition of the urban particles differ widely between geographic areas. For example, in the western US, especially in the South Coast Air Basin of Los Angeles and its environs, the chemical processes responsible for the formation of particulate matter depend on photochemical reactions. Nitric acid is the dominant end product. There are very few oxides of sulfur present because of the nature of the fossil fuels used in California. On the other hand, in the East Coast Corridor the consumption of sulfur-containing fuels is much greater, and the chemistry of the reactions leading to the formation of particulate matter is not as dependent upon photochemistry as it is upon chemical reactions. Sulfuric acid, not nitric acid, is the dominant end product present in particulate matter. The chemical nature of the particles formed in California are quite different those of the East Coast Corridor. Yet the health effects measured by epidemiologic techniques suggests that all particles have the same effect despite the differences in chemical composition. This is a very troublesome problem. One of the basic tenets of toxicology is that the toxicity occurs via chemical reaction. How then can the same effect result from very different kinds of chemistries? We must conclude that there is no plausible mechanism now available for particulate matter which can account for the reported results.
Particle Size and Site of Action of Respirable Urban Particles
The toxicity of particles also depends on the site within the respiratory tract where they are deposited. A major advance has been the recognition of the dependence of toxicity on the site of deposition. The site of deposition in the respiratory tract depends,. in turn, on the physical size of the particle. By measuring the amount of particles within the size range which can be deposited in the human lung, EPA adopted a biologically based criterion for its standard setting. This concept of defining particulate air pollution in terms of the size of particles most likely to be responsible for the adverse health effects is referred to as PM10 where 10 refers to particles of 10 micrometers aerodynamic mass median diameter or less. PM10 is a fairly good surrogate measurement for the amount of material that would actually be inhaled and deposited in the human respiratory tract. Schwartz and his colleagues extrapolated from measured PM10 values. PM10 is a major advance in public health policy pioneered by EPA. The PM10 concept shifts emphasis to particles of that size which are likely to be the most harmful to people. A network of PM10 monitors has been constructed in the US and large amounts of data have been accumulated.
Schwartz and his colleagues went beyond PM10 and extrapolated from a very limited set of measurements of PM2.5 and PM10 to estimate PM2.5 values and to relate mortality and morbidity to particulate matter exposure smaller than PM10 or particles less than 2.5 micrometers mass median aerodynamic diameter. Only a few data exist on the PM2.5 exposure in our major cities. By shifting from PM10 to PM2.5 values, a major difference in the regional deposition within the lung of these particles is suggested as the site of action. The smaller the particle the more deposition occurs in the deeper parts of the lung. By assigning toxicity to particles in the PM2.5 range the site of action is also assigned to the thoracic region of the lung. Because these PM2.5 values are calculated and not measured, it is very difficult to place the heavy weight of evidence on this ultrafine particle range as EPA has done in its criteria document. Even with a shift in attention to particles of this size range, there is still is no plausible mechanism for toxicity. Further, some of the CASAC members questioned the potency of the particles calculated from the mortality and mobility data. All of this underscores the importance of the research program reviewed by CASAC as part of the particulate matter standard setting process.
DOSE RESPONSE RELATIONSHIP
The dose response relationship is a curve that relates the number of individuals responding with an adverse reaction (mortality, morbidity or the like) to a certain exposure concentration of the chemical. The shape of the dose response curve is important when setting standards. All theories of the dose response relationship so far indicate that these curves will be non-linear; that is, there will be a point at which the probability that a response would occur is very unlikely. To put it another way, all theories suggest that there is a concentration at which nothing will occur while above that concentration adverse effects will occur. The point at which there is nothing detectable is the threshold. The dose-response relationship is at the heart of the risk assessment. In both the particulate matter and ozone standard the dose-response relationship is only poorly understood. Consequently, estimates of risk are also uncertain. Examples for ozone and particulate matter follow.
The Particulate Matter Dose Response Curve Is Linear Not Curved
The current assumption of epidemiologic studies is that the mortality or morbidity is a linear function passing through zero at zero concentration of particles. The dose-response function has no point at which no adverse effects occur. The linear dose-response curve is in opposition to all of the theories and experimental data derived for a host of chemicals acting by a variety of different mechanisms of action.
The epidemiologic basis for a linear relationship between effect and dose is very poor. The data are not supported by any kind of a generalized theory and are in many cases a default assumption coming about because the epidemiologic data are weak. It is very difficult for epidemiologists to relate exposure to effect. The methodologies of epidemiology at present are insensitive to the concentration or exposure effect. This is especially true in ecological studies where indirect evidence is used for adverse health effect.
For example, the epidemiologic studies of particulate matter health effects depend upon death certificates and the coincidence of an increase in death with an increase in particulate matter exposure. These studies again provide no indication of how a person might have died from the exposure to particulate matter. The studies only associate the death with the exposure to particulate matter. Nonetheless, the increases in mortality associated with particulate matter are troublesome. If the magnitude of mortality suggested by these studies is correct, then we are faced with a major public health problem that demands immediate attention.
Time and Intensity Relationships in Ozone Health Effects
EPA initiated a time and intensity study in cooperation with the USSR. This program was well thought out and attacked the question of which variable is most important in determining the health effects of ozone. From the data that were generated by this study it appears that the intensity is the most critical factor rather than the duration of exposure for ozone toxicity. These studies of the time and concentration effects on ozone toxicity led to the current hypothesis upon which the proposed ozone standard is based. If it is correct that the magnitude of the exposure is more important, then extremes of exposure should be reduced. One strategy to reduce exposure to extreme concentrations of ozone is to change the averaging time for the standard, making implementation plans stricter for short-term excursions. The US-USSR research program to study the time and concentration dependency of ozone adverse health effects was very productive and was progressing along a track which would, if continued, have provided us a great deal of information at this time. Unfortunately, EPA chose to reduce and essentially eliminate this line of study. Extramural support for the program lagged and ozone in general has become an unpopular topic for support by other government agencies such as NIEHS.
Based on the fragmentary information that we have available, I feel that it is appropriate to support the EPA proposal of changing the averaging time for the ozone standard so that large excursions over short time periods will be eliminated or reduced. However, one should recognize that changing the averaging time will have a major impact on state implementation plans and will have major economic consequences. Clearly, understanding the nature of the dose-response relationship is very important and affects which alternatives we choose to reduce ozone health effects.
Time and Intensity Relationship for Particulate Matter Health Effects Are Unknown
As stated above, most time and intensity (dose and dose-rate) relationships for chemicals follow a simple relationship that the product of the dose rate and the time of exposure form a constant. This constant is arbitrary and unique for each chemical Epidemiologic studies of the increases in mortality associated with increases in particulate matter are strictly linear with the amount of particulate matter. One reason way this assumption occurs is that a lag period has been assumed. The lag period means that the increase in mortality occurring 2 to 3 days after an exposure are related to the exposure to particulate matter, not earlier or later. The underlying hypothesis is that particulate matter toxicity is not immediately evident but occurs after this lag period. This very short acting time raises the question as to what happens when people are exposed to concentrations of particulate matter over the long term. We really have no data on the chronic effects in humans of exposure to particulate matter. Chronic exposure studies are very difficult to achieve using epidemiologic data.
To my knowledge there are no experimental animal data or controlled human studies which relate this kind of lag time to exposure to the toxicity of particulate matter. In my laboratory and that of my colleagues at UCI we have found that experimental animals such as the rat are very insensitive to particulate matter exposures. We have never observed potencies equivalent to that proposed for humans based on the epidemiologic data. This again raises the question of a plausible biological mechanism of action.
THE RESEARCH AGENDA
It is difficult for scientists such as myself whose livelihood depends on experimental research to stand before you and justify additional research without seeming to be self-serving. A careful study, however, of EPA's support of research in the past is related to the missing data in the standard setting process. Sadly, we would not be sitting before you if there had been a steady progression of air pollution health effects research. EPA's research strategy has been to ignore problems until the standard setting cycle is near. Then a massive effort is mounted which is expected over two or three years to result in sufficient data to solve the research needs. Regrettably we have seen that this strategy does not work. The same questions recur from criteria document to criteria document. There are just not enough resources put into air pollution health effects research so that we are really certain what we're about.
It is also my opinion that this problem also appears in the low esteem with which Congress holds EPA research. I am acutely aware that one Congress cannot obligate another Congress and that this independence of one Congress from another is fundamental to the development of our country. But I think it is time that the Congress in its wisdom faced up to the need to make its desire known to its successors that support of research for long-term problems in all areas of health is essential.
Air pollution is a long-term problem. From my observations and the data in the literature there is no urban area which does not have air pollution. We are still dependent on the consumption of fossil fuels for energy and the prospects of independence from fossil fuels are far into the future. My colleague, F. Sherwood Rowland, received the Nobel Prize in Chemistry for his contributions to the global problem of depletion of the ozone layer. Dr. Rowland's contributions clearly show that this air pollution problem is global. He also was able with his colleagues to demonstrate that this was a long-term process. I see no way that this is not also true for other kinds of air pollution problems.
As I mentioned above in my current review of the literature of ozone I found very little progress had been attained in ozone research over the last ten years. It is essential then that the Congress mandate to EPA a sustained basic research effort, Only if EPA clearly is committed to a long-term research effort will we solve the problems that still exist today as they existed ten years ago.
In addition, the Congress should resist any rush to judgment. I am deeply concerned over the effects of particulate matter exposure as currently revealed by epidemiologic studies. Similarly, I am concerned that we have not demonstrated an important increase in a health benefit from a small decrease in ozone concentration. Both of these alternatives however, are significant commitments on the part of society to change the underlying causes of both ozone and particulate matter generation in our cities. It is my firm opinion that the U.S. public would be willing to engage in whatever is necessary, but they will not support any arbitrary change that results in a significant economic and personal commitment.
Our experience in science policy clearly shows that the U.S. government is capable of mounting major efforts to solve major problems. No one could have predicted just five years ago the remarkable success which is being achieved in AIDS treatment. There is a similar likelihood that a large-scale problem such as air pollution could be better defined and directions for engineering applications clearly delineated if we understood more about the biologic aspects of this problem. Inventorying pollutants in the atmosphere is undoubtedly an important issue, but it does us no good to inventory these pollutants in the atmosphere and yet not have a clue as to what their biologic activity is. I may sound arbitrary in my opinion that we are in a state of great ignorance, but I think that once you listen to the testimony of my colleagues here you will come to the regrettable conclusion that I am an optimist. I therefore urge the Senate and this Committee to undertake a new direction in the support of research by EPA and by EPA's sister federal agencies such as the National Institutes of Health and in particular the National Institute of Environmental Health Sciences. These agencies need to be enabled, directed and empowered, indeed mandated, to carry out the long-term large-scale research that is necessary to understand much more fully the effects of air pollution on the U.S. population.
The Proposed Ozone Standard
It is my opinion that we will have achieved only marginal effects by decreasing the current ambient air quality standards for ozone from 120 parts per billion to 90 parts per billion. The nature of the dose response relationship is such that it may still be at a linear range and thus reduction to much lower levels may be necessary to result in the abolition of detectable health effects from ozone. My colleague, Robert Wolpert, and I published a simple analysis of different kinds of dose response relationships for ozone looking towards this very issue. How much would one have to reduce the ozone concentration in the air in order to be able to find a detectable advance in public health? Because the data are so sparse, a multitude of different kinds of theoretical treatments are possible. None of them, however, are sufficiently sensitive that one could lead to a clear prediction of a health benefit. On the other hand, as I mentioned above, a change in the time constant alone is going to have a great benefit. I endorse EPA's analysis of the time constant and think that EPA's proposal to a change in the averaging time for ozone is likely to be of benefit to the public health.
Still, I think that translating these changes into new state implementation plans may be very difficult. To translate both a change in the concentration, that is the amount of ozone that is permissible in the air and the duration over which it is permissible, will be a very difficult task indeed to implement.
Continued research into the health effects of ozone are urgently needed. Further reductions in the ozone standard may be indicated in the near future. Because of the economic impact of ozone standards and strategies, the highest quality research is needed.
Particulate Matter Standard
As I have said previously, I do not doubt that the particulate matter problem is a very serious problem indeed. We need to place a very strong active and progressive research program into place in order for us to cope with this problem. It is my view that too little is known, In the report of the Clean Air Scientific Advisory Committee to Administrator Carol Browner, the Committee pointed out that one of the areas in which additional research should be undertaken is chronic exposure.
I am not in favor of the use of a PM2.5 standard. A viable network of monitoring instruments and sound research supports the PM10 standard. The PM2.5 standard has no background. There is no existing research quality PM2.5 network. Without a research quality PM2.5 network it is not likely that we will make much progress towards the goal of a new particulate matter standard. We lack information on the actual PM2.5 in the atmosphere of our cities. We do not know the duration of exposure of people to PM2.5. The chemical nature of the PM2.5 fraction is poorly known. We lack a plausible biological mechanism for particulate matter. We do not know if regulation of PM2.5 will be of benefit. A strong aggressive long-term research program is essential to address the current data deficiencies if we are to convince people that this is a major problem.
Avoid Mistakes Of The Past
In my comments above I pointed out that the critical data deficiencies for ozone and particulate matter are generic and extend to the other criteria air pollutants. My criticisms of EPA and of the Congress I am sure have not endeared me to either party. My criticisms are also directed to me and my scientific colleagues. It is time that we faced up to the realities of life. Air pollution is here. Air pollution will be with us. Air pollution is a major problem that cannot be solved in five years.
EPA needs more resources. All of the health research establishment needs more resources to deal with this particular problem. The strategy adopted by NIH to deal with major health problems such as AIDS and cancer is dependent on ideas generated outside of the government. This is not to say that government researchers are not knowledgeable. Rather it is simply the recognition that there is great diversity in the United States. We have a lot of people working on the same problem, and from this diversity we achieve greatness. The Congress should instruct the National Institutes of Environmental Health Sciences, NIEHS, and EPA to place greater emphasis on air pollution, to seek actively support for extramural programs dealing with air pollution, to look for the unique idea to encourage the primary review groups that this is a programmatic area of importance. Lastly I would respectfully ask that the Congress use the legislative hammer in another way. The Congress can have a major impact on the sustainability of research in this area. Clearly the Senate recognizes that regulation of air pollutants is a major national problem. The Congress should, in my judgment, place a burden on the government agencies to carry out the needed long-term research. In doing so, the Congress has to realize that it has to reallocate resources and that air pollution is a national problem of long-term importance requiring additional support.
Thank you for the opportunity to have addressed you.