Epidemic: The Cancer-Producing Society

This essay is reproduced here as it appeared in the print edition of the original Science for the People magazine. These web-formatted archives are preserved complete with typographical errors and available for reference and educational and activist use. Scanned PDFs of the back issues can be browsed by headline at the website for the 2014 SftP conference held at UMass-Amherst. For more information or to support the project, email sftp.publishing@gmail.com

Epidemic: The Cancer-Producing Society

by Samuel S. Epstein

‘Science for the People’ Vol. 8, No. 4, July 1976, p. 4–11

This article was one of the papers presented in the session on Priorities in Cancer Research, at the American Association for the Advancement of Science meeting in Boston, Feb. 20, 1976. (See SftP, May, 1976 for a review of this and other AAAS sessions arranged by members of Science for the People.]

Cancer is now a major killing and disabling disease of epidemic proportions. More than 53 of the 210 million U.S.A. population (25%) will develop some form of cancer, and approximately 20% of Americans now die from cancer.

It is estimated that 665,000 new cancer cases were diagnosed and that there were 365,000 cancer deaths in 1975. Thus, cancer deaths in 1975 alone were eight times higher than the total U.S. military deaths in the Viet Nam and Korean war years combined.

No age, sex, or ethnic group is spared from cancer. Cancer is a leading cause of death at all ages, including infancy and childhood. Cancer has also been induced following maternal exposure to carcinogens, as recognized in post-adolescent girls whose mothers had been treated in pregnancy with diethylstilbestrol. 

The total economic impact of cancer is massive. Estimates indicate that in 1969 the direct costs for hospitalization and medical care for cancer exceeded $500 million. It appears that the total direct costs for a particular patient range from $5,000 to over $20,000. The direct and indirect costs of cancer, including loss of earnings during illness and during the balance of normal life expectancy, were estimated at a total of $15 billion for 1971.


Cancer Deaths, 1975 365,000
World War II Battle Deaths 292,000
Viet Nam War Deaths (6 years) 41,000
Korean War Deaths (3 years) 34,000
Auto Accident Deaths (1969) 59,600
Polio Deaths (1952; worst year) 3,300

The Recent Increasing Incidence of Human Cancer 

The rate of recent increases of cancer deaths is more rapid than the rate of increase in population, and is even more rapid than the increase in the overall rate of death. It is of interest to note that there are only three major causes of death which have significantly increased in the recent past; these are cancer, homicide, and cirrhosis of the liver. This increase in new cancer cases is real and is over and above that due to any increase from age alone. The cancer death rate appears to have approximately tripled since the beginning of this century, in spite of advances in diagnosis and cure. It is of interest to note that the major improvements in 5-year cancer survival rates occurred prior to 1955, and appear to reflect advances in surgery, blood transfusion, and antibiotic therapy, rather than in cancer chemotherapy. 


Site Whites
Lung 403 238 908 435
Pancreas 61 37 200 326
Colon 48 15 100 214
Breast 14 37
Prostate 43 153
Uterus and Cervix -41 -55
Esophagus -23 0 225 200
Stomach -68 -77 -46 -56

The earliest year for which cancer death rates are available is 1900. These rates were crude, not adjusted for age, and were based on approximately half of the United States population living in 153 cities in 10 states. The mortality data from the National Center for Health Statistics, created in 1933, are based on the total U.S. population and are age-adjusted. Overall crude cancer death rates since 1933 have increased annually by about 1% until 1975, when provisional estimates indicate a 2.3% increase over 1974; the reality of the 1975 increase appears to have been independently confirmed by Metropolitan Life Insurance data. A substantial proportion of this increase since 1933 has been from lung cancer and is due to smoking. Similar large increases in the incidence of cancer have been noted in other organs, particularly in Blacks, in whom some of this may reflect the increased availability of diagnostic facilities. In spite of the .increase in overall cancer incidence and in various specific organ rates, there have been significant declines in incidence for other such organs as stomach and cervix (see Table 1). There is evidence of a recently increasing incidence of use of estrogens by post-menopausal women. The “spotty” changes in cancer incidence and death rates over the past few decades have, in fact, provided major epidemiological clues as to environmental causes of cancer in various organs.


Year Cancer Death Rate
1900 64
1933 100
1960 147
1971 161
1972 167
1975 176

Environmental Carcinogens as Major Cause of Cancer

There is now a growing consensus that the majority of human cancers are due to chemical carcinogens in the environment and that they are hence ultimately preventable.2 Numerous estimates by expert national and international committees indicate that 70-90% of human cancers are environmentally induced; the Director of the NCI recently concurred in these estimates, and placed a figure of 90% as the incidence of environmental cancer. The basis for such estimates largely derives from epidemiological studies, in large community populations over extended periods, which have revealed wide geographic variations in the incidence of cancer of various organs. In some of these studies, the role of specific environmental carcinogens has been implicated or identified.3

There is also now general agreement that the U.S. population and workforce has been and is being continually exposed to a wide range of known and identified chemical carcinogens in their air, water, and food, besides, in all likelihood, to a greater range still of unknown or untested carcinogens. Potent new chemical agents are being synthesized and introduced into commerce and the workplace at an exponential rate, generally in the absence of adequate testing for carcinogenic and for other adverse public health and ecological effects. 

A recent National Cancer Insitute (NCI) atlas on cancer mortality rates, in different counties, has demonstrated marked geographical clustering of rates for various organs in the U.S. while population in heavily industrialized areas. Such data suggest associations between cancer rates in the general community and the proximity of residence to certain industries.

Apart from the importance of occupational factors in the incidence of cancer in the population-at-large, specific occupational exposures are a major cause of cancer deaths, particularly in males. Various estimates have indicated that approximately 10% of all current cancer deaths in males are occupational in origin. These include lung cancer and pleural mesotheliomas in insulation workers and in others, such as construction workers, exposed to asbestos; bladder cancer in the aniline dye and rubber industry, induced by such chemicals as 2-naphthylamine, benzidine, 2-aminobiphenyl, and 2-nitrobiphenyl; lung cancer in uranium miners of Colorado, in coke oven workers, and in workers even briefly exposed to bischloromethylether; skin cancer in drilling and shale oil workers; nasal sinus cancer in wood workers, cancer of the pancreas and lymphomas in organic chemists; and angiosarcoma of the liver, besides other cancers, in workers involved in the manufacture and fabrication of polyvinyl chloride. 

The toll of cancer in particular occupational exposures is overwhelming. For instance, it has been estimated that about 50% of asbestos insulation workers die of cancer, and that 20% of all long-term asbestos workers die of lung cancer. Approximately 30% of all premature deaths in uranium miners are due to lung cancer. Many other occupational groups are at high cancer risk, including steelworkers, miners and smelters, rubber workers, and workers in a wide range of petrochemical industries. 

Scientific Basis for Determination of Carcinogenicity 

The determination of carcinogenicity for a particular chemical or mixture is based on toxicological testing in experimental animals, or on epidemiological observations on human populations who have been exposed to chemical carcinogens. While each of these approaches has its own inherent problems, animal testing can enable the identification of carcinogens prior to their introduction to commerce and the workplace, rather than attempting their identification by retrospective epidemiological studies in human populations; the latter are generally based on identification of temporal or geographical clustering of specific organ cancers. 

Current toxicological techniques are relatively insensitive and limited in their ability to detect carcinogens, individually and in various combinations or mixtures, in concentrations realistically reflecting low or ambient levels and patterns of environmental exposures. Similarly, it is generally considered that epidemiological techniques are unlikely to detect weak carcinogens unless there are sharp differentials in exposure of the general population, as with cigarette smoking; even with smoking, the single largest cause of cancer deaths, several decades of investigation were required before causality could be established. For widely dispersed agents, including unintentional or accidental food additives, such as Dieldrin and DDT, to which the population-at-large is generally and ubiquitously exposed, human experience is unliKely to provide any meaningful indication of safety or hazard. 

There is an overwhelming consensus in the qualified scientific community that carcinogenecity data derived from valid and well-conducted animal experiments have a high degree of presumptive human relevance. Indeed, every chemical known to be carcinogenic to humans, with the possible exception of trivalent arsenic, is also carcinogenic to animals. Additionally, many chemicals now recognized as carcinogenic to humans were first identified by animal testing. These include diethylstilbestrol, bischloromethylether, vinyl chloride, and aflatoxins. There can be no possible scientific, besides other, justification for the continued insistence, by leading industrial representatives and some regulatory agency officials, that animal data must be validated by human experience as a prerequisite to regulatory action. The scientific validity of data derived from animal testing is legislatively recognized in the 1958 Delaney Amendment to the Federal Food Drug and Cosmetic Act,4 and in recent regulatory actions, such as the suspension of the major agricultural uses of Dieldrin, and, more recently, of Chlordane and Heptachlor, whose carcinogenicity has been clearly demonstrated in animals, but not yet in humans. 

Safe levels of human exposure to chemical carcinogens cannot be predicted on the basis of animal or epidemiological data. Such considerations underlie the 1958 Delaney Amendment, which imposes a zero tolerance for carcinogenic food additives. The position subsequently expressed by HEW Secretary Flemming, that “Scientifically, there is no way to determine a safe level for a substance known to produce cancer in animals.” —reflects the overwhelming consensus of the qualified scientific community. 

Obstacles to Reducing Human Cancer

A. Absence of Toxic Substances Legislation

There seems to be strong a priori grounds for associating recent increases in cancer mortality rates with the increasing synthesis and usage of industrial chemicals, and concurrent exposure of large human populations, over the last four decades. (See Table II) 

Such increases are likely to have occurred in other industrialized countries, although perhaps later and less dramatically than in the U.S. 

It is not possible now to estimate the proportion of these novel chemicals which pose carcinogenic, besides other, hazards to humans. Except for some special purpose regulations in the area of pesticides, food additives, and drugs, however, this massive post-war efflorescence of petrochemical technology has occurred largely unrestricted by national, much less international, controls. There has been no general requirement for pretesting of chemicals, prior to manufacture or use, for carcinogenic or other adverse effects. As a consequence, it is likely that many carcinogens have been in wide use, whose effects may only manifest now, or in the next few years or decades. The case of vinyl chloride may well be a harbinger of other carcinogens from this generation of materials. Recently recognized as an occupational carcinogen, vinyl chloride was originally introduced into large-scale production in the 1950’s, and synthesis grew at about 15% per year until about four billion pounds were manufactured in the U.S. in 1970. Of the vinyl chloride workers identified by June, 1974 with confirmed diagnoses of hepatic angiosarcoma more than half, however, had received their first exposure prior to 1950. It is interesting to note that the chronology of the development of recent data on the carcinogenicity of vinyl chloride has been recently reviewed in a report of the AAAS Committee on Scientific Freedom and Responsibility.5This report documents the suppression of carcinogenicity data by the Manufacturing Chemists Association (MCA), allowing continued exposure without warning of tens of thousands of workers to high concentratons of vinyl chloride.


Class of Chemical 1970 Production
Billions of Lbs
% Increase
% Increase
Cyclic Intermediates 28.3 38 No Data
Plastics, Resin materials, and plasticizers 20.6 36 1,130
Synthetic rubber and rubber processing 4.7 16 350
Surface-active agents 3.9 12 810
Pesticides 1.0 -2 710
Others 79.9 33 550
Total 138.3 32 No Data

Toxic Substances Legislation is critically required to enforce the requirement for toxicological testing, in general, and for carcinogenicity testing, in particular, of new chemical agents prior to their introduction to commerce and the workplace. Failure to enact such legislation is likely to result in still further increases in the incidence of cancer in the coming decades. Various adverse economic impact analyses of such legislation, in general, have failed to consider or reflect the very substantial, and hitherto externalized, costs of human cancer. Illustrative is the Congressional testimony of 7/11/75, by the MCA and Foster D. Snell Co. who expressed “concerns on the huge costs and inflationary· pressures” inherent in such legislation.

 Questions have properly been raised as to the practical feasibility of expanding available national resources to permit adequate future carcinogenicity testing of new industrial chemicals, as would be required by Toxic Substances Legislation, quite apart from testing the numerous untested or inadequately tested chemicals currently in commerce and the workplace. The current Federal capability in the NCI bioassay program, largely based on contractors including the Frederick Cancer Research Center, allows carcinogenicity testing of 100-150 compounds per year; 400-600 compounds are now under test. The current capability is a significant expansion of the capability some 5 years ago for testing only 20 compounds per year. Current NCI protocols involved a total of 500 mice and rats, at a total cost of approximately $110,000, for a routine carcinogenicity test. 

Several thousand new compounds are now being introduced into commerce each year. Using appropriate systems of priorities and registration, and with the possible judicious use of short-term screening tests,6 it has been estimated that approximately 500 new compounds would have to be tested each year, and that this would necessitate some four- to five-fold expansion of current facilities. This should not represent any major problem, particularly as there are major untapped potential resources in the private sector, including universities, and the chemical and pharmaceutical industries. The National Laboratories, such as Brookhaven, Oak Ridge, and Argonne, also represent major potential facilities.

There are major inconsistencies between the philosophies and practices of various federal agencies with regard to their individual regulation of chemical carcinogens.7 The same carcinogen is thus likely to be regulated, or not, in a widely disparate manner, dependent on whether it is found in air, water, food, or in the workplace, and dependent on its source of discharge and dissemination into the environment, from point or nonpoint sources. Toxic Substances Legislation is likely to encourage the resolution of such disparities, while still recognizing possible special jurisdictional requirements.

B. Low Federal Priorities for Research on Environmental Carcinogens.

Even within the scope of the currently limited available federal resources, there appears to be relatively low priority accorded to research on environmental carcinogens and on the prevention of human cancer. Various estimates, ranging from 5-20% of the total budget, have been made as to the expenditure by the NCI on environmental carcinogens; of the $581 million for fiscal year 1974, it has been claimed that the real figure is close to 10o/o, based on $134 million spent on “Cancer Cause and Prevention”, which, however, includes approximately $70 million spent on viral causes of cancer;8 for the 1975 budget of $691 million, direct expenditures on environmental carcinogens again appear to be close to 10%. The low priority for environmental carcinogenesis in the NCI also appears to be expressed by the fact that in January, 1974 subject index of current NCI grants, only one of a total of 307 pages deals with epidemiological and population studies on cancer. Additionally, of the 3 members of the President’s Cancer Panel and of the approximately 23 members of the National Cancer Advisory Board, none appear to have significant professional qualifications or experience in epidemiology and preventive medicine, and only one is authoritative in chemical carcinogenesis; there also would appear to be disproportionately strong industria] representation on the Board and Panel, in the absence of labor and consumer representation. 

Questions have also been raised as to the appropriateness of the relatively low NCI expenditures on anti-smoking propaganda, and on their relatively high expenditures for research on “Safe Cigarettes”; the latter costs should perhaps more properly be borne by the tobacco industry, which now currently spends about $250 million annually on advertising. In this connection, it is also of interest to note that the U.S. Dept. of Agriculture spends approximately $50 million annually on various tobacco support programs, and that its Agricultural Research Service (ARS) assigns more laboratory space to research on tobacco, for producing a more marketable product and not for safety, than on food distribution. These federal policies do not appear consistent with very high national costs from the current epidemic of lung cancer, apart from bladder cancer and cardio-respiratory disease, due to smoking.


According to Benno Schmidt, chairman of the President’s Cancer Panel, speaking to the National Cancer Advisory Board, “The time is ripe … It is clear that Congress and the public want to get carcinogens out of our environment and keep them out … This is becoming enough of a question that it may become the subject of broad legislation soon -sweeping legislation which may be hard to live with. Do we want to ban the use of any agent until it is proved non-carcinogenic? … Legislation could lead to test requirements which would pre-empt the whole NCI budget . . . We’ve got to keep the initiative. At the next Congressional hearing, you’re going to have to say things that sound like you’re doing something, not just having organizational meetings and things under consideration.” (emphasis added).

—from Drug Research Reports, Nov. 26, 1975


C. External Pressures on Scientists

Pressures on industry scientists to develop and interpret data on chemical carcinogenesis to be consistent with short-term marketing interests, have resulted in a wide range of mythologies, calculated to minimize the significance or reality of the effects of human exposure to particular chemical carcinogens. While these mythologies cannot withstand elementary scientific scrutiny, they have been vigorously and effectively advocated at a wide range of arenas including Congressional Hearings, proceedings of the 1973 OSHA Standards Advisory Committee on Occupational Carcinogens, EPA suspension hearings on Dieldrin, and the more recent EPA suspension hearings on Chlordane and Heptachlor. These mythologies include the following:

“Tumorigens” are less Dangerous than Carcinogens

The identity of “tumorigens”, as opposed to carcinogens, has been vigorously proposed, particularly for chlorinated hydrocarbon pesticides, such as DDT and Dieldrin, which have long been known to induce “hepatomas” in mice. Presumably on the basis of such alleged distinctions, recent statements have been made, illustratively by a senior HEW spokesman, in response to repeated Congressional questioning, that “there is no evidence to my knowledge that DDT is a carcinogen.”9 The invalidity of such alleged distinctions has, however, been repeatedly and unambiguously emphasized by numerous expert national and international committees which have unanimously concluded that the terms tumorigens and carcinogens have synonymous implications.10 This proposition has, however, apparently ceased to be relevant with the recognition of pulmonary metastases resulting from a wide range of liver tumors induced in mice by “tumorigens” such as Dieldrin, which also induce extrahepatic neoplasms in mice and rats.

“Animal Carcinogens” are less Dangerous than “Human Carcinogens”: 

This thesis proposes that valid distinctions, from a regulatory standpoint, can be drawn between chemicals shown to be carcinogenic in experimental animals and those known to be carcinogenic in humans. It is further proposed that less stringent regulatory standards should be promulgated for “animal” carcinogens such as ethyleneimine, dichlorobenzidine, and 4,4′ -methylene(bis)-2- chloroaniline, unless and until their carcinogenic effects can be validated by human experience, based on deliberate human exposure. The thesis was, surprisingly, reaffirmed by a senior EPA official at a Research Triangle Park conference convened, on 1/6/76, to review recent data on the identification of relatively high concentrations of the highly potent carcinogen dimethylnitrosamine in the air of cities including Baltimore, where it is known to be escaping from an FMC plant manufacturing rocket fuel, and New York City and Belle, West Va., where it is presumably formed by interaction of atmospheric nitrogen oxide pollutants and azines, from sources probably including automotive emissions.11 

The EPA official who has major responsibilities in this area was unwilling to entertain substantive discussions on this carcinogen as a “hazardous pollutant”, on the grounds that its human carcinogenicity had not yet been established. 

There is in fact no evidence for the existence of “species-specific” carcinogens. All chemicals known to produce cancer in humans, with the possible exception of trivalent inorganic arsenic, also produce cancer in experimental animals, generally in rodents. Recent experience with carcinogens such as bis-chloromethyl ether, diethylstilbestrol, and vinyl chloride monomer, moreover, amply confirms the predictive value of animal carcinogenicity data to humans.

Human Experience has Demonstrated the Safety of Occupational Exposure to Certain Carcinogens:

Such claims have been repeatedly made for a wide range of “animal carcinogens” including Dieldrin, a-naphthylamine, ethyleneimine and dichlorobenzidine, and for “low levels” of exposure to acknowledged “human carcinogens”. These claims are generally made on the basis of lack of positive documentation of excess cancer deaths or on the basis of undisclosed or partially accessible records on small populations at risk, and on short periods of follow up. Clearly, such data do not permit development of valid epidemiological inferences. “Safe Levels” of Exposure to Occupational Carcinogens can be Determined: It is alleged that no or negligible risks result from exposure to “low levels” of occupational carcinogens, such low levels generally being set on the basis of technical expediency, or on other poorly articulated concepts. Illustratively, the ACGIH has assigned acceptable levels, threshold limit values (TLV’s) for asbestos, bischloromethyl ether, and nickel carbonyl. Numerous expert national and international committees and bodies have unanimously attested to the fact that there is no mechanism for determining the existence of biological thresholds for chemical carcinogens, and hence that the TLV concept is totally inapplicable to chemical carcinogens. 

Most chemicals are carcinogenic when tested at relatively high concentrations: 

This is totally inconsistent with available information. In a major NCI contract to Bionetics, approximately 130 industrial compounds and pesticides, selected because of suspicions as to their possible carcinogenicity, were tested at maximally tolerated doses in two strains of mice, with commencing exposure in infancy; less than 10% of these compounds were found to be carcinogenic.12 Of a total of some 6,000 compounds listed in the NCI “Survey of Compounds Which Have Been Tested for Carcinogenic Activity”, approximately 1000 were reported to be carcinogenic; by current standards, only some 3000 of those tests could be considered valid, and a total of only 500 compounds could now be accepted as carcinogenic. It must be emphasized that these compounds were highly selected, many of them being chemical derivatives of known carcinogens, synthesized for basic studies on carcinogenicity.

Apart from possible conflict between corporate loyalty and scientific freedom and social responsibility, as recently discussed in an AAAS report,13 sometimes industry scientists are even supposed to express opinions. At public hearings on September 11, 1973, on the proposed standards for the 14 occupational carcinogens, a lobbyist, representing the major manufacturers of those chemicals stated that the corporate scientists on the Committee were not qualified in regulatory problems, which were the exclusive responsibility of corporate management.

The record of the National Center for Toxicological Research (NCTR), Pine Bluffs, Arkansas, appears to reflect external pressures. The NCTR was created by Presidential order on January 27, 1971, is supported fiscally by the EPA and FDA, and is operated by the FDA. The NCTR is considered to be a major national source of research on the scientific aspects of regulation of drugs, food additives, pesticides, and other consumer products. From its inception, however, it appears clear that the FDA planned to use this resource to attempt to develop data with which to invalidate the Delaney Amendment, and also to establish “safe levels” of exposure to chemical carcinogens. At hearings before Congressman’s Whitten’s Subcommittee on Agriculture, in April 1971, then Commissioner Edwards stated that:

The Pine Bluff testing facility will provide FDA with the scientific basis on which the Delaney anticancer clause may be changed … [reiterating his long-held view that the agency is] locked into an ‘all-or-nothing’ position because of the Delaney box … [He said] FDA didn’t want to make it more difficult by recommending changes until it has the scientific data to justify a modification. 

The scientific programs of the NCTR, including a “Mega Mouse” experiment and other large scale experiments, designed to establish “safe levels” for human exposure to such known carcinogens as DES, benzidine and other aromatic amines, were severely criticized in a report of 8/31/73 by an expert extramural ad hoc NCI committee, under the distinguished chairmanship of Dr. H.L. Stewart, although this report was subsequently dismissed by the Director of the NCI. As confirmed by Dr. U. Saffiotti, Associate Director of the NCI Division of Cancer Cause and Prevention, in a response to Senator Tunney of 4/21/75, these criticisms of NCTR programs still remain relevant.

Recently and more disturbingly, there are growing indications of attempts to re-write established principles of chemical carcinogenesis to suit alleged regulatory needs. Examples include a newly created Carcinogen Assessment Group (CAG) of EPA, and a sub-committee of the National Cancer Advisory Board (NCAB), charged by the NCI, on 9/19/75, to develop recommendations on the scientific basis for the regulation of environmental carcinogens. 

It is of interest to note that the CAG group and a parallel group, formed to assess “benefits” of carcinogens, were formed in partial response to claims of a senior EPA official that the “scientific credibility” of the Agency had been threatened by the actions of the Office of General Counsel in Suspension proceedings on Dieldrin, and on Chlordane and Heptachlor; these proceedings were largely based on the unequivocal demonstration of carcinogenicity of these pesticides, for which no human carcinogenicity data are available.

Draft documentations of the CAG, 11/4/75 and 12/9/75, and of the NCAB subcommittee, 11/11/75, have attempted to draw scientific and regulatory distinctions between the experimental induction by chemical carcinogens of apparently benign tumors and malignant tumors; such distinctions are at major variance with an overwhelming consensus to the contrary by expert national and international committees. The CAG documents also appear to indicate that the scientific assessment of risk from chemical carcinogens should be responsive to “socio-economic consequences and costs of regulation” and to “benefit-risk” considerations. The chairman of the NCAB subcommittee issued in public a draft document which had not been seen, let alone approved, by the subcommittee members and which contained major scientific distortions and inaccuracies. The draft was then introduced by Velsicol Chemical Co. into the EPA Suspension Hearings on Chlordane and Heptachlor. Subsequently, the chairman and subcommittee members stated to EPA that this draft was preliminary and should not be used for regulatory purposes. Nevertheless, this draft appears to have influenced the administrative law judge in his decision, 12/12/75, not to recommend the suspension of these pesticides, for which there are unequivocal carcinogenicity data in rodents; this recommendation was subsequently rejected by the EPA Administrator on 12/24/75.

D. Distorted Analyses on the Economic Impact of Carcinogen Standards 

A now apparently standard response by certain sectors of industry to attempts by regulatory agencies to promulgate standards limiting environmental and occupational exposure to chemical carcinogens is to forecast, generally on the basis of procured studies, major economic disruption and unemployment attendant on compliance. Apart from the questionable economic validity of such forecasts, they do not address themselves to the externalized costs, economic and otherwise, of carcinogenic and other toxic effects. 

Estimates in 1974 by A.D. Little, under contract to the Society of Plastics Industry (SPI), and by Foster D. Snell, under contract to OSHA, on the impact of proposed occupational standards for vinyl chloride predicted costs as high as $65 billion and losses up to 1.6 million jobs. It is clear that such estimates are gross distortions, as most PVC producers are now in compliance, in the absence of major economic disruptions.14 Illustratively, capital costs of the B.F. Goodrich Co. for compliance came to about $34 million. Furthermore, this industry is considering leasing its “clean-up” technology and has found that the installed compliance technology actually cuts labor costs. It is of interest to note that B.F. Goodrich raised its prices in June, 1976, on PVC products, claiming that costs of meeting safety standards were partly responsible. A Union Carbide official has recently expressed surprise as to the ease of current compliance, in the absence of economic disruption, with the 1 part per million (ppm) standard. It should be noted that the 1 ppm standard is far from stringent and that commercially available equipment can monitor down to 1 part per billion (ppb), which should more properly be the current standard. 

In Summary 

The incidence of human cancer is rising dramatically and we are now experiencing a major epidemic of cancer which is killing one in five Americans. The economic costs of cancer are minimally $15 billion annually. The majority of human cancers are environmental in origin and are therefore preventable. Apart from a wide range of chemical carcinogens already contaminating our air, water, food and the workplace, new carcinogens are being synthesized and introduced into commerce in increasing numbers and in a largely unregulated manner.

The constraints to a major reduction in the incidence of cancer are largely political and economic, rather than scientific. As yet there is no comprehensive Toxic Substances Legislation, a critical element in developing national policies for reducing the incidence of human cancers. Such legislation should include coherent policies for the regulation of environmental carcinogens by federal agencies. The NCI should assign higher priorities to research on environmental cancer, and regulatory agencies should control environmental carcinogens more vigorously. 

Scientific research on chemical carcinogenesis, in industry, the NCI and other federal agencies, must be insulated from political and economic pressures. Otherwise the principles of chemical carcinogenesis and attempts to protect the public health will continue to be subverted by considerations including short-term marketing interests and alleged regulatory needs.

Samuel S. Epstein

Samuel S. Epstein is a physician and faculty member at the Case Western Reserve School of Medicine, Cleveland. He wrote the original draft of the Toxic Substances Control Act, five years ago, requiring pre-marketing testing of new substances. This Act has yet to be passed by Congress. Epstein took on the petrochemical industry in fighting to ban Dieldrin and other carcinogenic pesticides.

>>  Back to Vol. 8, No. 4  <<


  1. Cutler, S. J. and Devesa, S. S. Trends in Cancer Incidence and Mortality in the U.S.A. p. 15, in, Host Environmental Interactions in the Etiology of Cancer in Man. Doll, R. and Vodopija (Eds), International Agency for Research on Cancer, Scientific Publication No. 7, Lyon, 1973.
  2. Epstein, S. S. Cancer Research, 84: 2425, 1975.
  3. Epstein, S. S. Cancer Research, 84: 2425, 1975.
  4. ” … no additive shall be deemed to be safe if it is found, after tests which are appropriate for the evaluation of the safety of food additives, to induce cancer in man or animals. . . “
  5. Edsall, J. T. Report of the AAAS Committee on Scientific Freedom and Responsibility, Science, 188: 681 (1975).
  6. McCann, J. et al. Proc. Nat. Acad. Sci., 72: December, 1975.
  7. Epstein, S. S. Public Health Hazards from Chemicals in Consumer Products. p. 45, in, Consumer Health and Product Hazards. The Legislation of Project Safety. Vol. 1, Chemicals, Electronic Products, Radiation, Eds. Epstein, S. S. and Grundy, R., MIT Press, Cambridge, Mass., 1974.
  8. Greenberg, D. Washington Post, January 19, 1975; Science and Government Report, April 1, 1975.
  9. Tepper, L. Hearings Before a Subcommittee of the House Committee on Appropriations, p. 63, May 6, 1974.
  10. Epstein, S. S. Cancer Research, 84: 2425, 1975.
  11. Fine, D. et al. Submitted to Science, 1975. Fine, D. N-Nitrosamines in Urban Community Air, Progress Report, EPA Contract 68-02-231214, January 5, 1976.
  12. Innes, R. et al. J. Nat. Cancer Inst. 42: 1101, 1969.
  13. Edsall, J. T. Report of the AAAS Committee on Scientific Freedom and Responsibility, Science, 188: 681 (1975).
  14. Rattner, S. New York Times, December 28, 1975.