DISCUSSION

T. M. Gocke. For the relatively few years that our medical school has been in Jersey City, we have been rather frequently plagued with statements both in the press and from our scientific colleagues that New Jersey pollutes the air of New York City. We did not take this seriously at first, but in the past few years we decided that we ought to study in a little bit greater detail the air pollution problem in our community and also to study the effects of air pollution on health. We also are interested in relating our findings to the findings of the British by using techniques similar to theirs. Therefore, about a year ago we undertook a survey which consisted of a respiratory questionnaire, a simple pulmonary function test, and a 70-millimeter chest X-ray. This survey was aimed at determining the prevalence rates of chronic respiratory symptoms among men living in the Jersey City housing project. Our epidemiologic definition of chronic bronchitis was adopted from the British as a chronic productive cough present during most of the days of the month for at least 3 months of the year. We interviewed men between the ages of 40 and 59 years. The population was relatively small; 95 percent of our defined population (435 men) cooperated in the survey. Twenty-one percent of these men had chronic bronchitis. Eight percent of them had chronic bronchitis, plus a history of recurrent chest illness for the past 3 years. Five percent of them have a history of bronchitis plus disabling dyspnea. Three percent had chronic bronchitis, plus both disabling dyspnea and a history of recurrent chest illness. On the simple pulmonary function test, the 1-second forced expiratory volume showed that men with chronic respiratory symptoms had a significant decrease in this reading compared to nonsymptomatic men. In Jersey City one aspect of air pollution that we studied in detail was average surfur dioxide levels, which over a 9-month period in 1961 and 1962 ranged from about 0.18 to 0.20 part per million. These levels are at least four times higher than those observed in suburban areas in northern New Jersey. This seems significant to us, but when we look at our population, we also must admit that 30 percent of the cigarette smokers have chronic bronchitis, whereas not one of the noncigarette smokers was so troubled. Comparing our results with those of the British, we found that the prevalence rates of chronic bronchitis for Jersey City men fell between

the rates recorded in London and those in rural England. This study then has been aimed to define a population risk. We have multiple factors in that the saturation of cigarette smoking in our population makes it a little difficult for us to determine just what role air pollution does play. We feel that only by following the natural history of chronic bronchitis in the men that we have identified as having it shall we be able to answer this problem. We are interested in relating not only the environment factor of respiratory symptoms but also the role of bacterial respiratory infections.

Geoffrey L. Brinkman. Dr. Spicer beautifully demonstrated the complexity of relating environmental factors to human health, and I want to mention a study we've done in Detroit which illustrates some of the difficulties.

Table 1 shows the group we studied, around 1,300 men, all full-time workers, ages 40 to 65. Group 1 is a group with no industrial exposure, hospital personnel and executives, who had no known occupational exposure to dust. Group 2 is a conventional industrial group. Groups 3 and 4 had both worked over 20 years in a foundry which is not well known for its cleanliness. Figure 1 demonstrates the prevalence of bronchitis, which was defined as having a cough productive of at least a teaspoonful of sputum a day for at least 4 of the previous 6 months. Notice that, contrary to what you might expect, the top group, Group 1, is the executive group which had no known exposure to occupational dust, yet its members had by far the highest incidence of bronchitis. Now why should Group 1 have the highest incidence of bronchitis? One factor was that this group had the highest incidence of cigarette smoking. In the foundry the men are not permitted to smoke for at least 8 hours of the day, whereas the man sitting at his desk is free to smoke throughout the day. Figure 2 shows the importance of cigarette smoking.

Frank L. Rosen. What is the incidence of bronchial asthma in the Los Angeles area as compared to other large cities in California, such as San Francisco, and to other cities in the United States?

Goldsmith. First, I have to clarify what Dr. Rosen means by incidence. This is a technical term and I assume he refers to frequency, which is customarily described as prevalence. But this is a minor distinction. The answer is that there is no discernible difference between the rates of asthma as reported by the California Health Survey, for example, between Los Angeles and San Francisco. If anything, San Francisco is higher; the difference is of dubious significance. As a further answer, to something which was not asked, we have also studied the prevalence of asthma in relation to the amount of pollution in specific areas in Los Angeles, and again, since the figures are of modest magnitude, I would not allege that they are necessarily significant. We find, and we have speculated upon this, that in the heavily polluted areas in Los Angeles there are very few cases of asthma. This leads us to wonder whether or not some of the people who have had the illness may have chosen to move from these areas. In reference to other cities in the United States, I can report that the highest level of asthma I know of in the United States is in Honolulu. I mention this without discussion.

S. S. Epstein. I wish to report briefly on some current studies at the Children's Cancer Research Foundation in Boston, the object of which is the development of a simple and rapid biological technique for assaying the carcinogenic potential of polluted atmospheres in terms of their content of 3,4-benzpyrene. 3,4-benzpyrene, in particular, and other polycyclic hydrocarbons to a varying but lesser extent, possess the property of photodynamic toxicity, as evidenced by their ability to sensitize cells to otherwise nontoxic longwave ultraviolet radiation. The photodynamic toxicity of pure compounds such as 3,4-benzpyrene is simply demonstrated by incubating aqueous suspensions of these materials with motile cells such as Paramecium caudatum in the dark for 2 to 3 hours, followed by irradiation at 3,660 Å. Under these conditions a lethal effect, proportional inter alia to the carcinogen concentration over a range of 10-5 to 10-11 gm/ml, can be directly observed.

Using standardized techniques, a series of crude benzene extracts of air particulates and chromatogramed fractions thereof (kindly made available by Drs. Falk and Sawicki) have been tested for photo

dynamic toxicity, using 3,4-benzpyrene as controls. Information obtained in this manner is being correlated with available chemical and biological data. The following aspects of the photodynamic assay may be contrasted with the conventional mouse skin carcinogen bioassay: (1) rapidity and simplicity, and the fact that information may be obtained in the course of a day and not a year or so later; (2) economy; (3) the small size of the samples required (10 milligrams of a tar is more than adequate); (4) the test does not appear to be influenced by anticarcinogens, promoting agents, and nonspecific irritants.

Theron G. Randolph. I am speaking as a practicing internist and allergist.

Attempts to appraise the clinical significance of the air pollution problem, as far as chemical exposures are concerned, must begin with the realization that there are two major divisions of the subject. One has to do with outdoor chemical air pollution, which you have heard discussed from various standpoints. Equally important is the subject of indoor chemical air pollution. In my experience, this indoor air contamination is such a frequent cause of chronic illness in susceptible persons that it must be evaluated and preferably controlled before one is justified in drawing deductions as to the clinical significance of chemical contamination of the outdoor atmosphere.

Although outdoor chemical air contaminants enter homes located in contaminated areas, chemical air contamination arising within a patient's living quarters may occur irrespective of the location of such dwellings. The major sources of such contaminants are odors and fumes arising from leaking utility gas or the combustion products of gas, oil, or coal. The gas kitchen range, gas-panel heating units, and fuel-oil space heaters are the most significant, although appreciable air contamination is also derived from hot-air furnaces (irrespective of the type of fuel used), sponge rubber padding, bedding and upholstery, insecticides, paint odors, disinfectants, and various other odorous household materials. Contaminated air from the garage also often enters the house.

These relatively constant sources of indoor chemical air pollution are rarely suspected as inciting and perpetuating causes of chronic illness. Only as patients are maneuvered in respect to these exposures may acute reactions be observed which demonstrate cause-and-effect relationships. For instance, on the basis of observing clinical effects

in moving patients in and out of their homes and, later, of maintaining the patient in his home but removing and then replacing his gas kitchen range, over 800 such devices have been removed permanently from the homes of highly susceptible

persons.

A wide range of chronic illnesses result from such day-in and day-out hydrocarbon exposures. The most serious are depressions and other advanced psychotic states (Randolph, T. G.: Ecologic Mental Illness-Levels of Central Nervous System Reactions, Third World Congress of Psychiatry, vol. 1, University of Toronto Press, 1962). Lesser grade cerebral reactions manifest as mental confusion and brain fag as well as physical fatigue. Closely related manifestations are rheumatism, arthritis, myalgia, neuralgia, headache, and related musculoskeletal and neurological syndromes. Any of the responses ordinarily considered as allergic, especially stuffy nose, coughing, bronchitis, bronchial asthma, are also commonly on the basis of susceptibility to airborne chemical contaminants.

Finding and avoiding these home environmental incitants impinging on the physical and mental health of susceptible persons is opening a new experimentally orientated medical approach to many chronic illnesses.

Goldsmith. I should like to say one thing about Dr. Randolph's very interesting observations, which doesn't in any sense conflict with them. That has to do with the value of staying at home during periods of severe air pollution. It's important, as Dr. Randolph emphasized, from all the facts we know, and they of course are not adequate, that in a severe air pollution episode in any major metropolitan area, including London or Los Angeles, there is every reason to recommend as a public health measure that the people who may be most susceptible to air pollution do in fact stay at home. Perhaps they should keep their ranges off, but at least they should remain indoors.

Monroe G. Sirken. My question was touched upon by the two previous speakers. Much interest is centered on the air pollution outside, but people live most of the day inside. To what extent can we get a measure of the exposure inside by knowing what the air pollution is outside? I don't mean the additional inside air pollutants such as Dr. Randolph discussed, but rather whether or not we are getting an accurate measure of the exposure of a person who spends 16 hours of the day inside a building when we measure the level

of air pollution outside the building. Dr. Landsberg, what about the person who's working in an air-conditioned building and living in an air-conditioned apartment? This seems to be the growing trend in our larger cities.

Landsberg. First of all, we do have some parallel measurements between pollutants indoors and outdoors, at least on particulate matter. Generally one thing is very obvious and that is that the indoor pollutants follow the trend of the outdoor pollutants. They may not be at the same level, but the variation is almost exactly the same. Apparently the buildings of normal construction in which we live have a high diffusion coefficient and sooner or later there is an adaptation to the outdoor condition. You cannot seal a building; building codes prohibit you from sealing a building in fact, because of the danger of asphyxiation. As to the second part of the question, on an air-conditioned building, that depends very much on the type of air conditioning. We do have some information on that, too. If you have an air-cooling system which simply cools and circulates the air, the level of pollution in a ventilated building may be a little higher than outdoors, because everything that usually would settle is "stirred" by the airspeed inside the building. Hospitals in particular are horrible in that respect. We have a few measurements on filtered air which show that much depends upon the filter. Most filters are quite efficient with respect to pollen, and that was the purpose for which they were constructed. Some filters moisten the air, making the particles heavier, so that they sink to the ground and, if you do not stir them up, pollution is lessened. We have very little information on the exact chemical composition of indoor air. From what we do know, it seems that indoor air is worse air than outdoor air even when the latter is polluted, because of the gas ranges, the fireplaces, the house dust, and other allergens that are present. Hygroscopic particles, which exist in large quantity outdoors, are eliminated by the filters in the air-conditioning systems. Some of the modern. air-conditioning systems which continually recirculate the air and refilter it efficiently leave it almost free of pollutants, but we must remember that even then some of the gaseous pollutants stay. Carbon monoxide is not eliminated. Sulfur dioxide is generally altered, but it stays as sulfur trioxide or something of that type, so we are not completely protected by air conditioning. Biometeorologically speaking, we do not know whether the elimination.

of the diurnal rhythm is really good for people. In other words, whether or not living in a completely even environment is healthy has not been established.

Sidney Weinhouse. Would Dr. Kotin or Dr. Wynder estimate the approximate number of lung cancer deaths attributable to urban air pollution, and the extent to which stationary and automotive sources contribute to lung cancer?

Kotin. I'll vouch for the fact that polluted urban air as reproduced in the laboratory will induce lung cancer in experimental species. Whether or not any environmental source unequivocally alone produces lung cancer requires a degree of speculation and, as I said some years ago, it depends on how heroic you are in attempting to extrapolate. Certainly, elimination of the two most widely suspected environmental sources of carcinogenic agents would unequivocally result in a reduction in the lung cancer rate. We have no basis for outright guessing, and it would be guessing to estimate what the actual reduction would be with the elimination of one or the other factor. I suspect that a significant reduction would occur with the elimination of either, a fantastic reduction with the elimination of both. You also asked that we comment on the incidence of lung cancer of the adenocarcinoma type now presumed due to chemical carcinogens, and I emphasize the word "presumed." Certainly, the house of cards that related morphology to etiology over the past few decades is beginning to tumble. We know of no unique morphologic response to an environmental agent in which, by looking at the response in the intact host, you can identify the agent. There are certain paradoxes in terms of carcinogenesis as a discipline when one looks at the whole problem of lung cancer. For example, squamous cancer is termed "environmental cancer," and in studies it has been shown consistently to occur at a later period of life than do the other forms of cancer to be presumed nonenvironmentally related. This in no way negates the association. It does indicate that the complexity to which we give lipservice is perhaps more complex than we realize. There really is no pure type of lung cancer histologically, and as a corollary, relating it through. specific environmental factors is, at least with pathologists, extremely hazardous. A lot of this sounds like doubletalk. That is because the information does not yet exist. To conclude, eliminating polluted urban air would result in a reduction in the incidence of lung cancer. Eliminating ciga

rette smoking would result in a reduction in the incidence of lung cancer. I don't care to crawl out on a numerical limb.

Wynder. The effect of a reduction in cigarette smoking has been measured in more detail than a possible reduction of air pollution. Drs. Hammond and Horn, and Drs. Doll and Hill of England, have shown that giving up smoking cigarettes reduces the likelihood of getting lung cancer and that the reduction becomes greater with length of abstinence. The point of urban-rural risk has been tackled by Hammond and Horn in another study. They standardized for the amount of cigarettes consumed and discovered that a city population of 50,000 or more has a death rate from lung cancer that is about 20 percent greater than that for the whole population, for both cigarette smokers and noncigarette smokers. It is possible that if they had standardized in addition to amount of cigarettes smoked, for degree of inhalation, and for butt length of cigarettes smoked and for possible occupational differences, the differences might have disappeared altogether. It is hoped that Drs. Hammond and Horn will produce in their projected new study enough detail on all these points so that they can determine whether the person who smokes, let's say, 20 cigarettes a day, lives in a rural area, and does inhale, and smokes his cigarettes three-fourths down, has the same rate of lung cancer as the same individual living in the city, or a different rate. Of course, matching for occupations would also be necessary. We know that certain occupations affect the rate of lung cancer. Like Dr. Kotin, I also would not dare to make a prediction. It has been well established that among noncigarette smokers today, lung cancer is a most unusual disease, in this country and in Great Britain. The question really is: How much does air pollution in addition contribute to your chances of getting lung cancer if you are already a cigarette smoker? As I indicated, there is an urban factor, in the United States, as well as in England, but in my opinion, there is no established proof, at least in the United States, that it is due to air pollution. It is possible that in England with special climatic and coal-burning problems we have a very select set of circumstances. Frequent lack of central heating and a habit of sleeping with open windows even in winter must also be considered in this regard. The high prevalence of chronic bronchitis in England may well be a reflection of these circumstances.

The influence of these various factors in England deserves more attention.

Adenocarcinoma is, as you know, Dr. Weinhouse, one type of cancer that has interested us for many years. One difficulty is that few investigators have interrogated enough patients with adenocarcinoma of the lung to obtain statistically significant results. This has been further complicated by the tendency to group adenocarcinoma together with terminal bronchiole carcinoma. We are continuing to study this problem. The data suggest that terminal bronchiole carcinoma is relatively more frequently seen. in a nonsmoker than is epidermoid cancer. The adenocarcinoma patient is also more often a nonsmoker than the patient with squamous carcinoma, but we also find more heavy cigarette smoking among patients with adenocarcinoma than we find in the control population.

Weinhouse. Is chronic exposure to carbon monoxide a serious health problem? How often in general do serious toxic levels occur in cities? And is there a development of tolerance in chronic exposure?

Goldsmith. The use of the word "chronic" makes the first question more difficult to answer. In my opinion, exposure to carbon monoxide in cities is a health problem of some consequence. When you use the word "chronic," I believe you are suggesting a chronic carbon monoxide toxicity. This is a matter of dispute. My own view is that it has not been proved to exist. This should not be interpreted as meaning that it could not exist, and it should not be interpreted as meaning that it isn't worth looking for. But the evidence so far doesn't support it.

About serious toxic levels, my statements will refer to the standards set by the California State Health Department-the middle level, which is called "serious" because by definition it implies an important interference with the bodily function, in this case the function of transporting oxygen by hemoglobin. From reliable data, it may be stated that 30 parts per million exposure averaged over 8 hours will produce approximately 5 percent carboxyhemoglobin, which means that 5 percent of the body's circulating hemoglobin is unavailable for the transport of oxygen. According to this definition, such serious levels occur several times a year in Los Angeles. I don't recall the exact figures, but we checked statistics for 2 or 3 years and we found 14 such episodes. So I believe the California standards are satisfactory. Such levels do occur in cities,