The philosophy of establishing air quality criteria as guides for the setting or air pollution control standards in the United States: inter-regional symposium on criteria for air quality and methods of measurement, Geneva, 6-12 August 1963

WORLD HEALTH ORGANIZATION Geneva, August ORGANISATION MONDIALE DE LA SANTE WHO/AP/17 ~ 30 July 1963 ENGLISH ONLx THE PHILOSOPHY OF ESTABLISHING AIR Q.UALI'IY CRITERIA AS GUIDES FOR THE SETTING OF AIR POLLUTION CONTROL STAJTDARDS IN THE UNITED STATES by V. G. MacKenzie Chief, Division of Air Pollution US Public Health Service Air pollution was recognized as a problem of national significance by the Con- gress of the United States of America in 1955 with the passage of the Federal_Air Pollution Control Act. The law authorized the Public Health Service of the Federal Government to conduct a programme of research, technical assistance to State and local governments, training of technical personnel, and the dissemination of infor- mation; it further specifically reserved to the States and local governments the primary responsibility for the regulatory control of air pollution. In 1960, Congress directed the Public Health Service to determine from the stand- point of human health, the amounts and kinds of substances which may be safely dis- charged to the air from motor vehicles. A progress report, submitted to the Congress two years later, provided limited qualitative evidence that automobile emissions do adversely affect the health of humans. Thus in the United States of America there is division of responsibility between levels of government as to the control of air pollution. On the national governmental level, there is responsibility for research as to causes, ~ffects, and means of con- trolling air pollution; and to assist the States, communities, and private organiza- tions in putting to work on air pollution problems the knowledge that is now.available and will become available in the future. The States and the local governments have responsibility for the regulatory control of tl1e sources of air pollution, including the enactment of air pollution AP/17 ~e2 control legislation, the promulgation of regulations, and for enforcement of the laws and regulations within the State or community. The setting of stan- dards of air quality is considered to be a part of this regulatory control responsibility. Legislation now pending in Congress would specifically direct the Federal Public Health Service to establish criteria of air quality guides for those State and local agencies legally responsible for the setting of standards of ambient air quality. In the United States of America~ air pollution has been regulated prima- rily on the basis of judgments as to what was reasonably practicable in re- ducing the emissions from various pollutant sources. With few exceptions, such as the States of California and Oregon where a beginning has been made in promulgation of ambient air quality standards, this pragmatic approach is still the basis for most American air pollution control practice. Over about the past decade, however, as air pollution problems have received increasing public attention, a broad consensus has developed as to the desirability of a more rational approach to air pollution control. Such an approach has ap- peared to be most feasible through the derivation of allowable air contami- nant concentrationsj the subsequent determination of related limits of the associated emissions, and the consequent fixing of engineering controls that should be applied. This procedure would seem to be logically designed to at- tain the desired purposes and to be absolutely just and fair to all the par- ties concerned. Viewed in this perspective~ nobody can be against the deve- lopment and application of air quality standards. In fact, this is virtually how they are viewed at present in the United States - ever,yone is for them. However, this in itself does not cnsur~ their development. Those who begin in earnest to deal with the development of air quality criteria or standards soon discover that there are underlying questions of fundamental significance which must be recognized and considered. In the United States these matters are now receiving broad attention for the first time - on some there is agreement and on others no broad consensus has yet been reached. Unless scientific con- sensus, and indeed public consensus, can be reached on many of these funda- mental questions~ the practical application of the air quality approach to the control of air pollution is likely to elude us. AP/17 page 3 The first of these questions concerns the need to develop a basic point of view about, and a basic agreement on, the objective~ of the air quality criteria which everyone agrees should be developed. Certainly, to be broadly used these _, criteria nrust be acceptable not only to those who \'lould need to apply them, but also to the many who would benefit from them. Let's look first at what the cri- teria or standards should accomplish insofar as the general public is concerned. It appears certain that the standards would have to be "good" enough that they would not allow contaminants to adversely affect ~~e health status or important body functions of even the most sensitive or susceptible groups in the popula- tion. In the United States, I believe also that the public would not support any standards which would allow air contaminants to exceed the threshold of sen- sory irritation, of taste, or of odour, or which would allow demonstrable damage to animals, to ornamental plants, or to agricultural crops. This much seems clear; but with respect·to another group of air pollution effects related to ruch matters as reduction of visibility, the soiling and corrosion of fabrics, metals, and other materials, there is 'Little evidence as to any public consensus of how "clean" the air should be. With respect to this latter group of ail" pol- lution effects, I think it is a safe ass~ption that the public wants what it considers to be "clca'-111 air, and would perhaps be content to leave the specific deftnitions of .clean air to the technicians. All in all, we believe that we have a reo.sonable notion as to 'lo'1hat the pub- lic would expect of air quality criteria. There is, however, considerably less certainty of agreement as we move into the area of the scientific and technical attitudes. There is littl-e evidence, for example, of a scientific and technical and standards consensus on the desirabiljty of developing air quality crit~ri~n the basis of technical knowledge which is less than perfect. All of us concerned with ·this problem will concede that there is still much to be done in cataloging and cor- relating the specific effects of air pollutants with the contaminant concentra~ tions which may be responsible. It seems to me that we must agree to get on with the job of developing air quality criteria and standards against·a back- ground of technical and scientific knowledge which is not now and probably never will be perfect. To accept this viewpoint does not mean that the standards should be developed arbitrarily or capriciously, but rather that we must exer- cise reasoned judgments on the considerable but necessarily incomplete AP/17 page 4 information which is already available. There are some, however, who would in- sist on rigorous 11 proo!'11 of quantitative cause and effect relationship for indi- vidual pollutants before adopting any air quality criteria. What, indeed, con- stitutes 11 proof11 , or at least adeque.t8 evidence in these instances? Unfortunate- ly, we lack full ~~derste.nding of the relations between many non-biological environmental factors and health and of the basic mechanisms by which such fac- tors exert their effects. We do not have a parallel to Koch's postulates which, if met, have long been accepted as proof of c~use and effect in the infectious disease field. Dr Richard A. PrindJ.e has recently proposed the followirl?; series of tests of evidence of ca~se and cf~ect relationships between non-biological environmental agents and disease~ l. A positive assoc~".ti::' o·:- :;,~:::::..::tio::L::>hip must be demonstrated between a factor or factors and a disease condition in a population. 2. This same factor, or factors, must produce disease states or conditions in biological test subjects which are similar to those observed in the population and which can be shown to be logically reproducible in human beings. 3· Measures designed to remove the factor, or factors, mitigate or protect against its action, or otherwise protect the popu- lation, must prove efficacious. We need agreement on some such simvle series of tests as proposed by Dr Prindle as a working approach to the assessment of evidence of cause and effect relation- ships in the environ~ental health field. In the status of our current knowledge, it seems unlikely that simple single factors n.re n:;sponsible fur many of the effects with which we arc concerned. Rene Dubos may indeed be right: 11 Search for the cause may bo a hopeless pursuit because most disease states are the in- direct outcome of a constellation of cj_rcumstances rather than the direct result of single determinc:.nt fac·cor-.:;. In ;;urn, I submit that ~"'c must be content with reasonable evidence of caus~ and effect relationship and not insist on final "proof" thereof; the continual f3c:arch for better 11 proof11 should not be used as justification for procrastination where human health is involved. There is also li t-'~le evidenC'e of scientific consensus on the question of single-valued stanc.an-~s for individual contaminants versus more complex arrange- ments employing ~ultiplc numbers related to various effects or to types of areas involving various land uses. There is, of course 1 considerable logic to the desire to relate scientifically the concentrations of air contaminants to AP/17 page 5 increasing severity of physiological effect on hunans or to separate frcm these .. the effects on animals, plants, other property, or visibility or other general characteristics of the atmosphere. Nevcrtl1cless~ it does not appear desirable that multiple-numbered criteria~ if used, should provide justification for sub- jecting people in one area to more severe health effects than in another nor sanction the continuance of sensory irritations in the absence of more serious effects. In somewhat parallel fields in the United States involving environmen- tal contaminants, single-valued standards for various substances is the app~oach generally used in regulating the purity of food and drinking water. Such stan- dards, moreover, are not limited to h8alth considerations alone, but include aesthetic considerations as well. Air quality standards arc often referred to as if they could be established solely on the basis of technical and scientific considerations; however~ social considerations must also play a part. The complexities of cur society require the making of many broad decisions involving the balancing of social "benefits" versus "damages" or "risks". A good current example is the problem of "benefits" versus "risk" in the nuclear energy field. Gradually, there has been increasing acceptance of the currently prevailing concept that, for exposure of the general population to radioactive materials, there is no "threshold" below which adverse biological effects do not cccur. In the absencu of such a"thrcshold11 there is no completely 11safe11 level of exposure other than zero; consequently, under these circumstances, the problem of balancing "benefits" versus "risk" becomes very real if any extensive use of nuclear energy is to be made. The containment of radioactive waste materials has therefore become an important ,clement in consi- derations involving the utilization of nuclear energy. There is evidence tha~ in the pollution field, we are c0nfrontcd with some problems in developing standards or criteria for air quality similar to those of radiation exposure. I would like to discuss this matter as it concerns one class of air pollutants the polycyclic hydrocarbons. Over the past several years, the Public Health Service and a few other organizations have devoted con- siderable attention to this class of pollutants. l"lethods have been developed for sampling and analysis of a number of such compounds and considerable data have been accumulated as to their quantitative occurrence in the open atmosphere and in effluent gas streams from combustion and other processes. Available data AP/17 page 6 indicate that air pollution contributes a large proportion of the total exposure of humans to such contaminants. In the main~ they appear to deriv,~ principally from fuel combustion, whether in furnaces or engines, or from burning waste materials, and from some industrial processes. An urgent ~erest in these pollutants is justified because of the demon- strated cancerigenic potency of some of them, i.e. their ability to produce cancers in experimental animals. As in the case of radiation exposure, it is thought that there is no "threshold" of exposure for biological effect by such radiomimetic chemicals. This concept of "no threshold" is incorporated in Federal law in the United States, in the Food and Drug Act, which specifics that, for foods in interstate commerce, there shall be no detectable amount of any ad- ditive known to be capable of producing cancers in experimental animals. What, then, do we do in developing criteria of air quali~ for such air pollutants? In the present status of knowledge, the only completely "safe" level of which we can be sure is zero. But this is patently impossible of at- tainment in our present society. What, then, should be done? In such instances, it may well be necessary to establish some formal mechanism to exercise judgment on "benefits" versus "risk" to accomplish the same functions as are now being dealt with in the field of radiation exposure. One additional area of need which I would like to mention is that concerned with the practical application of air quality standards. If the advantages of the rational approach to the control of air pollution possible through the air quality criteria approach are to be realized, all of the parties currently sup- portingthe development of air quality criteria should give equal support to the steps necessary to enable the control agencies to apply them in their operations in an adequate manner. The urgency for setting air quality criteria in the United States stems from the States and communities and from industry, all of which look to the Federal Government fo1' guidance with respect to the setting of ambient air standards upon which to base sound emission standards. To serve a useful pur- pose, therefore, we believe that air quality criteria must truly be comprehen- sive guides which point the way to standards which will be realistic, techni- cally feasible, and compatible with other requirements of our society. AP/17 page 7 The fact thatibave chosen to point out some fundamental problems which face us in the establishment of air quality criteria does not mean that they should not or cannot be established. Neither should the fact that I have not discussed specifically such things as the desirability of reaching agreement on methods of measurement and on nomenclature be taken to mean that I do not consider such matters implicit in any discussion of air quality criteria. We are well aware of the magnitude and complexity of the philosophical~ scientific and sociological problems with which all countries are confronted in attempting to promulgate air quality criteria and to set ambient air stan- dards for the control of air pollution. Therefore, we welcome the efforts of the World Health Organization to stimulate international cooperation for the purpose of achieving the most economic utilization of the limited manpower~ facilities~ and funds with which all countries are attempting to cope with this problem.