Tobacco Documents Online

ICOSI BACKGROUND BRIEFING PAPERS

---

A1 ICOSI BACKGROUND BRIEFING PAPER CLAIM International companies are usinq double standards in the selling of high 'tar' and nicotine cigarettes in developing countries. RESPONSE Consumer preference determines the type of cigarettes sold in any market. For cigarettes, as for food and other commodities, consumers in developing countries have in many casesdifferent tastes from consumers in the Western world. Furthermore, in most developing countries, the per capita consumption of cigarettes is extremely low when compared with Western countries. In some African countries, for example, it is as low as 12 or 13 cigarettes a month and a very high proportion of the cigarettes sold are sold not by the pack but by single units. It is not therefore surprising that in those countries consumers do not appear at present to obtain satisfaction from cigarettes with low yields which are gaining popularity in Europe and North America. The policy of the international companies is to offer the consumer a wide range of brands of varied yields but in all markets account must be taken of consumer preference. In the developing countries this is evidenced by the fact that the cigarettes produ- ced by cottage industries (e.g. Kreteks), which have a higher yield than international company brands, have very considerable success against international company brands. Furthermore, where local national companies are producing machine-made cigarettes, their brands are as high, or higher, in yield as the interna-" tional brands sold in their countries. It is thus totally unjust for the opponents of Smoking to accuse the international tobacco companies of using double standards. Where they sell brands in developing countries which have higher yields than the equivalent brands on their own domestic markets, they are doing so in response to consumer preference. Changes in consumer preference are a slow process. If international companies made rapid reductions in the yields of their brands in developing countries, any remaining higher yield brands would inevitably become the consumer's choice. If all brands in a developing country were heavily reduced in yield in a short time period (which would not be possible in any case in a competitive situation, particularly where local national companies and cottage industries as well as international companies were involved), some

A2 scientific reports h~ve suggested that consumers would most likely adjust their patterns of smoking, for example in terms of inhala- tion and puff frequency. Finally it should be pointed out that in a number of developing countries international companies are gradually reducing the yields of their brands to provide their customers with a wider range of choices, but this move would become impratical and self-defeating if it become out of step with consumer preference. llth May 1979

B1 ICOSI BACKGROUND BRIEFING PAPER CLAIM International cigarette companies do not include warning clauses on cigarette packs or in cigarette advertising in developing countries. RESPONSE In some developing countries cigarette packs do carry a warning clause. In those and in a number of developed countries both packs and advertising carry warning clauses as a result of Government legislation or heavy Government pressure on the tobacco industry. While respecting the right of health officials to offer opinions on matters relating to public health which are of interest to them, the tobacco industry holds and has always held the view that in developed countries, where intensive publicity in all media on smoking and health has ensured that consumers are aware of the issues involved, warning clauses are unnecessary. Furthermore, in developing countries with a multiplicity of language and dialect, warning clauses present almost insuperable difficulties for the industry. llth May 1979

Cl ICOSI BACKGROUND BRIEFING PAPER CLAIM In most developing countries international tobacco companies are opposed to the publication of'ta~ and nicotine tables and decline to reveal the 'tar' and nicotine deliveries of their brands. RESPONSE The publication of 'tar' and nicotine tables has been brought about in a number of developed and some developing countries by legislation or government pressure on the industry. The governments involved primarily consider that these two components are of prime importance in relation to certain deseases which the opponents of smoking have attributed to cigarette smoking. In addition, some governments have introduced official league tables to replace tables already published by independent consumer asso- ciations, the results of which had proved to be suspect and unreliable. 'Tar' is, in fact, a misnomer. Contrary to popular belief, ciSa- rette 'tar' is not something to which human smokers are exposed. 'Tar' more properly describes specific laboratory products obtained from several distinct laboratory procedures. There are a number of ways of defining and measuring 'tar', e.g. 'total particulate matter', 'dry particulate matter', and 'particulate matter, water and nicotine free'. Comparisons between league tab ie results in different countries can be misleading, and possibly invalid, if the definition of 'tar' is not the same in each table. Further, the popular but unscientific notion that the exposure of smokers to various cigarette smoke constituents can be deter- mined by laboratory analyses is untenable. The rate and amount of delivery of various smoke constituents depend upon too many uncontrolled factors which vary considerably among smokers, including, for example, the number, size and frequency of puffs, the depth and degree of inhalation, the length of the butt remaining and even the time of day a person smokes. In contrast, the measurements of 'tar' and nicotine are made from smoke produced by smoking cigarettes mechanically under a set of defined laboratory conditions. If laboratory conditions were standardized they may enable the comparison of two or more products, but they cannot reproduce the way in which the indivi- dual smokes a cigarette.

C2 As to the alleged 'harmful' constituents of cigarette smoke, critics claim that 'tar', nicotine and several other gases are the primary sources of toxicity. In the past, much effort and research has been devoted to understanding tobacco use and health. Yet, in spite of all the research that has been done, no one has ever established that any ingredient or group of ingredients as found in tobacco smoke is harmful to humans. Therefore, the scientific basis for the claim that certain levels of 'tar' and nicotine have health consequences is lacking. The international tobacco companies, therefore, oppose the publication of league tables because they are likely to be very misleading to the consumer in several major ways. The tables might well suggest that levels of certain ingredients in tobacco smoke have proven health significance. The consumer might be misled to believe that he is actually exposed to the published levels of the constituents when indeed there are too many variations in the smoking patterns of humans to draw such a conclusion about an individual's exposure. The problem of use of differing methodology and definitions for determining the levels of these smoke constituents would result in the publica- tion of noncomparable and, consequently, valueless information. llth May 1979

ICOSI BACKGROUND BRIEFING PAPER CLAIM International tobacco companies use in develop.in~ countries advertising methods not considered acceptable in the Western World. RESPONSE There are of course in the Western world restrictions on cigarette advertising related both to the methods or media used and to the content of advertisements. These restrictions were imposed by legislation or acceded to by the tobacco industry under pressure from Governments. While it may be said, as in this accusation, that these restrictions apply to methods unacceptable to the opponents of smoking in the Western world, the rationale behind the imposi- tion of such restraints is not valid in itself - and there is therefore no logical reason why these restraints should also be applied to developing countries. The falsity of the rationale is evidenced by the fact that : i. Studies have concluded that advertising has no significant effect on total cigarette consumption. A recent study carried out in the United Kingdom examined the effects of advertising on cigarette consumption there. The study included data on a quarterly basis over the last 20 years and the results showed that : a) The effects of advertising on cigarette consumption were statistically insignificant regardless of the definition of the advertising variable and of the time period chosen. b) The same as in a) above applied to the effects of Government.anti-smoking advertising. Evidence from other markets also supports these results. The experience of those countries prohibiting cigarette advertising has been that consumption was not reduced or, as in the case of Norway, consumption initially fell but these reductions were strictly temporary. D1

D2 Further examples exist in Eastern Europe and in 3 countries where no advertising at all is permitted, there have been significant increases between 1970 and 1975, despite anti-smoking campaigns. These countries are Poland (increase of 24.5 %), USSR (increase of 15.7 %), and Hungary (increase of 11.4 %). It should also be noted that Hungary and Poland have a larger per capita consumption than West Germany where cigarette advertising is permitted. Advertising does not entice non-smokers to become smokers. Studies indicate that other influences such as peer pressure and the examples of others are far more important factors. In this connection the following statement by K. Waernberg included by the U.S. Department of Health, Education and Welfare in its publication (June 1975) entitled "Ban on Advertising - What Then ?" (Smoking and Health. Vol. II. Health Consequences, Education, Cessation Activities and Governmental Action.) is particularly pertinent. "There is no evidence to support the view that a ban on advertising would have a positive effect on smoking habits. No empirical research has been able to show that aggregate brand advertising leads to greater total ' tobacco consumption. Nor has anything been found to suggest that advertising entices non-smokers, young people in particular, into becoming smokers. It follows, therefore, that there can be no evidence showing that a ban on advertising would result in reduced tobacco consumption and fewer new smokers." The misconceptions behind the rationale for advertising restric- tions are thus evident and it follows that advertising methods are in themselves irrelevant, as is demonstrated by a recent study in the United Kingdom which shows no effect of advertising on total consumption over 20 years despite the severe curtailment of advertising methods over that period in the U.K. What is more relevant is that the tobacco industry adheres to its announced policy of not designing its advertising to encourage smokers to smoke more, to create new smokers or to appeal to children - and this the industry is doing both in the developed and the developing countries. Furthermore it is important that the advertising media in developing countries should continue to be available to the tobacco industry. Only thus can information be given to the smoking public in those countries concerning the availability of products or new product information. llth May 1979

ICOSI BACKGROUND BRIEFING PAPER E1 CLAIM Tobacco ~rowin~ in Third World countries inhibits the production of food crop~. RESPONSE In most Third world countries there is abundant land available and thus its usage for the growing of tobacco in no way inhibits the production of food crops. Since the early 1960's, increased emphasis has been placed on food production in the Third world, particularly in the development of new plant varieties, the use of fertilizers, improved methods of plant protection and hygiene, mechanisation and small-scale irrigation. The tobacco companies' principal contribution to the encouragement of more productive farming has been to demonstrate clearly the benefits of crop rotation systems. Through the application of new agricultural techniques to tobacco and, where appropriate, to rotation food crops, the level of agricultural and related business activity has thus risen faster in tobacco growing than in non tobacco growing areas. As a result of improved land cultivation and the use of residual fertilizer techniques, increased yields in tobacco areas of food crops have been recorded as follows : Chile Bangladesh India (Mysore) Kenya Guatemala Maize Rice Millet 9~eat Beans Jute 30% 30% 250% 100% 20/25% 20/25% 100% 100% Ihcreased yields from the use of fertilizers on tobacco crops have encouraged their use on other crops. A similar transfer of techno- logy has occurred in the use of crop chemicals. Tobacco growing has financed the purchase of sprayers, so giving farmers the means of applying crop chemicals to their other crops. In several Central American countries (e.g. Panama, Costa Rica and Nicaragua) tobacco is grown under irrigation. The purchase of the pumps and ancillary equipment has been financed by the tobacco companies for tobacco growing and the farmers use the equipment out of season to increase the yield of their rotation crops, the

E2 I0. scale of production of which would not otherwise have justified the expense of irriga.tion. A similar "spin off" effect has occurred with tractors and cultivation equipment in West Africa. In the State of Western Nigeria, the tobacco companies' sponsored tractor schemes are used both by tobacco and by non tobacco farmers and through their example has led to the establishment of similar schemes by Government and by other operators. In the more advanced agricultural economies of Venezuela, Argentina and Brazil, the tobacco companies have established soil analytical programmes, often conducted in their own laboratories. This enables advice to be given to farmers on lime and fertilizer requirements for the best crop yields. Where tobacco is grown in rotation with other crops, the soil analysis can be used to provide a fertilizer programme for the alternate crop. The rapid spread of intensive cropping in the Third world has brought with it problems of soil erosion and loss of natural soil condition and also weeds, pests and disease. Particularly in the tropics, this loss of organic matter and soil structure are accen- tuated where primitive tools and implements are used. Subsoiling and deeper ploughing have been advised and are now common practice in Central America and parts of East and West Africa as a means of breaking up the hard pan which had developed. Water permeabi- lity has been improved and the potential root zone of the soil increased. In Northern Nigeria, the tobacco companies are evaluating alterna- tive cultivation techniques which result in reduced wind erosion at the beginning of the wet season. On land which is undulating or hilly, as in Kenya, farmers are advised and encouraged to construct contours and plant on the contour. In Brazil, 95 per cent of the tobacco crop is planted by farmers who observe the principles and practive of soil conservation. Traditionally, farmers have removed or burned crop residues. However, the use of tractors and modern equipment had enabled them to plough in crop residues to maintain the level of organic matter -in the soil. In one country, Nicaragua, many farmers are indeed growing a green manure crop specifically for this purpose. llth May 1979

ICOSI BACKGROUND BRIEFING PAPER F1 CLAIM The tobacco companies encourage farmers in Third World countries to use wood for the flue-curing of tobacco thus depriving the poor man of his national fuel resources. RESPONSE The flue-curing of tobacco requires artificial heat, though other methods of curing do not. In Canada, the United States, the Caribbean, Central America, India and Indonesia, oil or coal or sometimes natural gas is used for flue-curing. The ideal unit for the production of flue-cured tobacco is however the family farm, and in some parts of the Third world the use of wood as fuel for flue-curing has had obvious advantages. Wood was available; it was cheap; its use demanded a lower level of techno- logy and did not involve the outlay of foreign exchange. In the earlier stages of tobacco production, wood was available from land clearance in many parts of Africa, from other agricultural activi- ties such as rubber plantations in Malaysia, from Government forest reserves in Sri Lanka, or from natural forests as in Brazil. Unfortunately, although Governments in these countries have generally encouraged the development of tobacco as a cash crop within their agricultural economy, with a few notable exceptions (such as Pakistan, where Government-initiated wood fuel plantations now produce 30.5 million cubic feet of fuel a year) their forestry Departments have been slow to see the need for reafforestation programmes. Whenever natural resources are used for any purpose, it is impor- tant to conserve them. For wood, reafforestation is a necessiry programme. Therefore, the tobacco companies have taken the initia- tive. It has not been possible for them to become directly involved in plantation. They have however encouraged farmers to plant trees, either on a co-operative basis or individually. Wood fuel colopera- tires jointly owned and managed by the farmers were very successful in Uganda but the agricultural systems in other countries have not allowed them to be set up elsewhere. Tree planting by individual farmers also encounters difficulties, sometimes because of the land tenure system, sometimes because of the limited size of the farmer's holding, and sometimes because he cannot envisage the future fuel requirement. Nevertheless, in a number of countries the tobacco companies have successfully encouraged farmers to plant trees, usually exotic fast-growing species, on any free land and using seedlings provided free by the company. As a result of 03678384

F2 co-operation between the tobacco companies and the farmers, 115 million trees have been planted in Brazil since 1977 alone. Other examples are in Kenya, where 3 million seedlings have been issued during the past three years, and in Sri Lanka where 40 seedlings are issued free to each farmer every year. Ii has also been possible to reduce fuel requirements by recommen- ding improved curing techniques and by improving the efficiency of wood furnaces. Technical improvements developed to meet problems in one country have been transferred to other countries where similar problems occur. '" Several tobacco companies are conducting research into the use of waste products as fuel. For example, the conversion into briquet- tes of waste material from coir, rice husk and sawdust is currently under investigation in Sri Lanka and the results could prove useful in other countries including Brazil. Trials are being oonducted in Kenya of solar energy as a supplementary source of heat for flue- curing. llth May 1979

ICOSI BACKGROUND BRIEFING PAPER SMOKING AND HEALTH -- A PERSPECTIVE For centuries, millions of people throughout the world have enjoyed smoking tobacco in various forms. The explorer, Christopher Columbus, encountered natives in the West Indies smoking tobacco when he discovered the New World in 1492. Tobacco was introduced to Western Europe in the sixteenth century, and it was not long before the custom of smoking began to spread. Nor was it long before smoking began to be attacked. G1 Today, tobacco use continues to be opposed primarily on the grounds that it is a health hazard--a claimed cause of lung cancer, cardiovascular disease and other disorders. The claims that are made about imoking and health rely mainly on reported statistical associations, but it is a scientific principle that such associations cannot establish causal relationships. Even the 1964 U.S. Terry Report noted that "statistical methods cannot establish proof of a causal relationship in an association." What statistical associations can do, however, is point to the need for further clinical and laboratory research to determine the precise relationships.l Moreover, considerable scientific data are inconsisten~ with the smoking-disease hypothesis. Indeed, there is ample scientific evidence showing that the smoking and health question is unresolved. 0~783~6

G2 This is illustrated by an examination of the accusation that tobacco use is a major cause of premature death. Some of the major statistical studies of select populations contain findings inconsistent with this accusation. In separate studies of Canadian and U.S. veterans, it was found that veterans with longer smoking histories did not necessarily experience higher mortality rates. One showed that persons who had smoked for 25-34 years had lower mortality rates than persons who had smoked for 15-24 years, and the other that smokers of 15-29 years had lower rates than persons who smoked for less than 15 years.2 Also of interest is the 10-year study of residents of towns in Denmark and England.3 In the Danish town, researchers found that nonsmokers had a higher death rate than those who smoked cigarettes, pipes and cigarettes, and other forms of tobacco. In the English town, they found that the "heavier" smokers had lower death rates than the "lighter" smokers. The reported statistical associations between smoking and higher mortality rates may be more consistent with a genetic hypothesis than a causal one, according to certain highly respected scientists. Studies of identical human twins provide support for this theory. Because identical twins have the same genetic makeup, they are excellent subjects for the study of extrinsic factors which may affect mortality. These studies have indicated~ that in identical twin pairs who have different smoking habits, no excess mortality has been observed in the twins who smoked more heavily.4 ~

In view of such scientific uncertainties, it is unfortu- nate that discussions of smoking and health questions have frequently become strident. Rationality is often replaced by emotionalism, a situation commented on recently by Dr. Gary Huber of Harvard University: "When it comes to tobacco, opinions are given often with such emotionalism that there is very little discussion, much less scientific objectivity."5 However, scientific objectivity is needed in considering complex questions of disease causation. Despite the many claims made about smoking and health, a legitimate and continuing scientific controversy surrounds the subject. The smoking and health question is unresolved and answers will be found cnly through unbiased scientific inquiry. May 1979

G4 References Berkson, J., "Smoking and Cancer of the Lung," Mayo Clin Proc 35: 367-385, 1960. Berkson, J.,. "Mortality and Marital Status: Reflections on the Derivation of Etiology from Statistics," Amer J Pub Hlth 52(8) : 1318-1329, 1962. Berkson, J., "Smoking and Lung Cancer," Med Proc i0: 327-336, 1964. Burch, P. R. J., "Problems in the Interpretation of Cancer Statistics with Special Reference to Lung Cancer," J Soc Occup Med 25(1): 2-10, 1975. Fisher, R. A., "Dangers of Cigarette-Smoking," Brit Med J 2(5039): 297-298, 1957. Fisher, .R.A., "Cigarettes, Cancer and Statistics," Centennial Rev Arts Sci 2: 151-166, 1958. Fisher, R. A., "Lung Cancer and Cigarettes?" Nature 182(4628): 108, 1958. Fisher, R. A., Smoking: The Cancer Controversy.' London: oliver & Boyd, 1959, pp. 7-47. Rigdon, R. H., "Cigarette Smoking and Lung Cancer: A Consideration of This Relationship," South Med J 62(2): 232-235, 1969. Rigdon, R. H., Statement presented at Hearings before the Committee on Interstate and Foreign Commerce, House of Representatives, April 15 - May I, 1969. Serial No. 91-12, pp. 1018-1025. Schoolman, H. M., et al., "Statistics in Medical Research: Principles Versus Practices," J Lab Clin Med 71(3): 357-367, 1968. Seltzer, C. C., "An Evaluation of the Effect of Smoking on Coronary Heart Disease," JAMA 203(3): 193-200, 196S. Yerushalmy, J., "On Inferring Causality from Observed Associations," Controversy in Internal Medicine, F. J. Ingelfinger, et al. (eds.).--Philadel~hia: W. B. Saunders Co., pp. 659-668, 1966.

G5 U. S. Public Health Service. Smoking and Health, Report of the Advisory Committee to the Surgeon General of the Public Health Service. Washington, U. S. Department of Health, Education, and Welfare, Public Health Service Publication No. 1103, p. 20, 1964. 2. Doll, R. and A. B. Hill, Mortality Study in British Doctors, and Best, E. W. Ro, et al., Mortality of Canadian War Pensioners, as Reported in: U. S. Public Health Service. Smoking and Health, Report of the Advisory Committee to the Surgeon of the Public Health Service. Washington, U. S° Department of Health, Education, and Welfare, Public Health Service Publication No. 1103, p. 93, 1964. 3. Cole, T. J., et al., "Bronchitis, Smoking and Obesity in an English and a Danish Town: Male Deaths After a 10-Year Follow-up," Bull Physio Path Resp 10(5): 657- 679, 1974. 4. Cederlof, R., et al., "Morbidity Among Monozygotic Twins," Arch Environ Health 10(2): 346-350, 1965. Cederlof, R., et al., "Respiratory Symptoms and 'Angina Pectoris' in Twins with Reference to Smoking Habits. An Epidemiological Study with Mailed Questionnaire," Arch Environ Health 13(6): 726-737, 1966. Cederlof, R., et al., "Hereditary Factors and 'Angina Pectoris,'" Arch Environ Health 14(3) : 397-400, 1967. Cederlof, R., et al., "Hereditary Factors, 'Spontaneous Cough' and 'Smoker's Cough,'" Arch Environ Health 14(3): 401-406, March, 1967. Cederlof, R. and L. Friberg, "Tobacco Smoking and Health: Results of Epidemiologic Studies in Twins," Lakartidningen 65(27): 2727-2734, July 3, 1968. Cederlof, R., et al., "Cardiovascular and Respiratory Symptoms in Relation to Tobacco Smoking: A Study on American Twins," Arch Environ Health 18(6): 934-940, 1969. Cederlof, R., Statement presented at Hearings Before the Committee of Interstate and Foreign Commerce, House of Representatives, April 15 - May i, 1969. Serial No. 91-11, pp. 873-882.

G6 De Faire, U., et al., '"Concordance with Respect to Mortality in Ischaemic Heart Disease and Cerebrovascular Disease, A Study on the Swedish Twin Registry," CVD Epidemiol Newsl 18(1): 21, 1975. Friberg, L., et al., "Smoking Habits of Monozygotic and Dizygotic Twins," Br Med J 1(5129): 1090-1092, 1959. Friberg, L., et al., "Mortality in Smoking Discordant Monozygotic and Dizygotic Twins," Arch Environ Health 21(4): 508-512, 1970. Friberg, L., et al., "Mortality in Twins in Relation to Smoking Habits and Alcohol Problems," Arch Environ Health 27(5): 294-304, 1973. Huber, Gary, "State of the Art on Tobacco and Health," Paper presented at the American Thoracic Society Annual Meeting, May 15, 1978.

ICOSI BACKGROUND BRIEFING PAPER HI Smoking and Lung Cancer The accusation that smoking causes lung cancer is well publicized. What is perhaps not so well publicized is the considerable amount of scientific evidence that questions this causal hypothesis. The claim that smoking causes lung cancer is based mainly on studies in which smoking was reported to be statistically associated with lung cancer mortality. But this reasoning ignores the important question of whether or not such associations have causal significance. Scientists generally agree that statistical associations do not establish causal relationships. Dr. Joseph Berkson, the distinguished medical statistician now retired from the famed Mayo Clinic in the United States, has stated that "Cancer is a biologic, not a statistical, problem" and that "if biologists permit statisticians to become the arbiters of biologic questions, scientific disaster is inevitable.''I However, such associations do point to the need for further ..~ 2 clinical and laboratory investigations. A number of laboratory and clinical findings have raised serious questions about the alleged causal relationship between smoking and lung cancer. These include the fact

H2 that no ingredient or group of ingredients, as found in tobacco smoke, has been established as producing lung cancer in humans.3 In addition, despite numerous laboratory experiments over many years utilizing thousands of animals, the type of lung cancer in humans most frequently associated statistically with smoking has never been induced in research animals by 4 inhalation of tobacco smoke. Inconsistencies in the epidemiological evidence also raise questions about a causal relationship between smoking and lung cancer. For example, large variations in lung cancer death rates reported from various countries cannot be explained by differences in tobacco consumption. The United States and Canada have the highest per capita consumption of cigarettes in the world,5 but rank tenth and sixteenth, respectively, in lung cancer mortality for white 6 males. Conversely, while the United Kingdom, Finland and the Netherlands have lower per capita consumption than the United States and Canada, they have substantially higher 7 lung cancer death rates. Further, a recent study found that the incidence of lung cancer in Vienna has remained constant since 1962 despite the fact that tobacco consumption has been steadily 8 rising.

Other research also challenges the smoking-lung cancer hypothesis. Studies conducted in England and in the United States have shown that the age when lung cancer is detected does not depend upon the age at which smoking begins, or the duration of smoking, or even whether or not 9-11 one smokes. In addition, the vast majority of smokers 12 do not develop lung cancer, and persons who have never smoked develop the same type of lung cancer most frequently 13 associated with smoking. H3 In recent years, there have been reports of a so- called "epidemic" of lung cancer. However, many scientists have noted that this apparent increase in lung cancer deaths may well reflect both the availability of more sophisticated techniques for detecting a disease that has existed for many generations, and a tendency to overdiagnose the disease in 14 recent years. The Swedish pathologist Dr. T. Berge of the University of Lund, in commenting on the incidence of lung cancer, said recently: "It is obvious that the frequency figures based on clinical methods are too low in earlier series, whereas in recent series the figures may be too high because of overdiagnosis. This may give a false impression of the real increase.''15

H4 During the ear.ly part of the century, doctors were limited in their ability to diagnose lung cancer. The principal methods for detecting lung cancer--x-ray, bronchoscopy and sputum cytology--did not become widely available to 16 physicians until after 1930. Indeed, there was not even a specific disease category for lung cancer in the International Classification of Diseases until 1939. If only a very small percentage of cases diagnosed as tuberculosis and other infectious or respiratory diseases at the turn of the century were actually lung cancer, the apparent increase in this disease since 17 1900 would be virtually eliminated. In fact, recent studies in England, Austria, Sweden, and Japan have shown that lung cancer still is being confused with respiratory 18-21 diseases such as tuberculosis, pneumonia and influenza. In addition, many scientists have recognized that lung cancer has been erroneously diagnosed in recent decades. For example, autopsy studies have indicated that in cases in which lung cancer was listed as the cause of death, only 40% of clinicians' diagnoses could be confirmed by microscopic . .. 22-25 examlnatlon. Professor Rudolph Hoppe also recently reviewed 20,000 clinically diagnosed cases of lung cancer in Northern Germany and reported he could confirm only 62% of the diagnoses among males and 44% among females.26

Research has i~plicated many other factors in the development of lung cancer. These include occupational exposures, viruses, diet, genetic influences, food additives, air and water pollution, stress, hormones, aging and impaired body defense mechanisms. The late Sir Ronald Fisher of England, world-famed statistician and geneticist, proposed in the 1950s that constitutional factors might be far more important than smoking in lung cancer. He too emphasized that statistics alone do not prove causation. Professor Philip Burch, a well-known British scientist, addressed the Royal Statistical Society in 1978 on the subject of "Smoking and Lung Cancer: The Problem of Inferring Cause." He said that "many eminent persons, committees and commissions have unanimously concluded that lung cancer 'is almost entirely due to cigarette smoking.' I once shared that view, but having now studied the evidence in more detail and from new angles I feel unable to reach a definitive conclusion . I find myself forced back to Fisher's earlier verdict: 'the data so far do not warrant the conclusions based upon them.'"27 H5

Clearly, the scientific picture surrounding smoking and lung cancer is complex. An adequate understanding of lung cancer causation continues to elude the scientific community. Obviously, much more research is needed to fill the many wide gaps in knowledge. May 19 79

References H7 i. Berkson, J., "Smoking and Lung Cancer: Some Observations on Two Recent Reports," J Am Stat Assoc 53(281) : 32, 1958. 2. Berkson, J., "Smoking and Cancer of the Lung," Mayo Clin Proc 35: 367-385, 1960. Berkson, J., "Mortality and Marital Status: Reflections on the Derivation of Etiology from Statistics," Amer J Pub Hlth 52(8): 1318-1329, 1962. Berkson, J., "Smoking and Lung Cancer," Med Proc I0: 327-336, 1964. Burch, P. R. J., "Problems in the Interpretation of Cancer Statistics with Special Reference to Lung Cancer," J Soc Occup Med 25(1): 2-10, 1975. Fisher, R. A., "Dangers of Cigarette-Smoking," Brit Med J 2(5039): 297-298, 1957. Fisher, R. A., "Cigarettes, Cancer and Statistics," Centennial Rev Arts Sci 2: 151-166, 1958. Fisher, R. A., "Lung Cancer and Cigarettes?" Nature 182(4628): 108, 1958. Fisher, R. A., Smoking: The Cancer Controversy. London: Oliver & Boyd, 1959, pp. 7-47. Rigdon, R. H., "Cigarette Smoking and Lung Cancer: A Consideration of This Relationship," South Med J 62(2) : 232-235, 1969. Rigdon, R. H., Statement presented at Hearings before the Committee on Interstate and Foreign Commerce, House of Representatives, April 15 - May i, 1969. Serial No. 91-12, pp. 1018-1025. Schoolman, H. M., et al., Statistics in Medical Research: Principles Versus Practices," J Lab Clin Med 71(3): 357-367, 1968. Seltzer, C. C., "An Evaluation of the Effect of Smoking on Coronary Heart Disease," JAMA 203(3): 193-200, 1968.

H8 Yerushalmy, J., "On Ihferring Causality from Observed Associations," Controversy in Internal Medicine, Ingelfinger, F. J., et al., (eds.). Philadelphia: W. B. Saunders Co., 1966, pp. 659-668. U. S. Public Health Service. Smoking and Health, Report of the Advisory Committee to the Surgeon General of the Public Health Service. Washington, U. S. Department of Health, Education, and Welfare, Public Health Service Publication No. 1103, p. 20, 1964. Lijinsky, W., Testimony, Hearings Before the Intergovernmental Relations Subcommittee of the Government Operations Committee, U. S. House of Representatives, p. 135, March 1971. Rosenblatt, M. B., Testimony, Hearings Before the Committee on Interstate and Foreign Commerce, U. S. House of Representatives, p. 1260, April 1969. 4. Furst, A., Statement for the Subcommittee on Health, Committee on Labor and Public Welfare, U. S. Senate, 94th Congress, Second Session, Cigarette Smoking and Disease, 1976, February 19, 1976, pp. 100-116. 5. Lee, P. N. (ed.), Tobacco Consumption in Various Countries. London: Tobacco Research Council, 1975, pp. 4, 5. 6. Segi, M., et al., Cancer Mortality for Selected Sites in 24 Countries No. 6 (1966-1967), Japanese Cancer Society: November 1972, pp. 76, 77, ii0, Iii. 7. loc. cit. supra, refs. 5-6. 8. Parkash, O., "Lung Cancer: A Statistical Study Based on Autopsy Data from 1928 to 1972," Respiration 34: 295-304, 1977. 9. Passey, R. D., "Some Problems of Lung Cancer," Lancet II: 107-112, 1962. i0. Pike, M. C. and R. Doll, "Age at Onset of Lung Cancer: Significance in Relation to Effect of Smoking," Lancet I: 665-668, 1965.

12. 13. Herrold, K. McD., "Survey of Histologic Types of Primary Lung Cancer in U, S. 'Veterans," Path 7: 45-79, 1972. Farris, J., Statement presented at Hearings before the Committee on Commerce, United States Senate, on S. 559 and S. 547, March 22, 23, 24, 25, 29, 30 and April i, 2, 1965. Serial No. 89-5, p. 973. Le Roux, B. T., "Presentation," Chapter II, Bronchial Carcinoma. E. & S. Livingstone Ltd., Edinburgh and London, pp. 12-21, 1968. 14. Rosenblatt, M. B. and J. R. Lisa, "Diagnostic Progress in Lung Cancer: Historical Perspective," J Amer Geriat Soc 16: 919-929, 1968. 15. Berge T. and N. G. Toremalm, "Bronchial Cancer -- A Clinical and Pathological Study. II. Frequency According to Age and Sex During a 12-Year Period," Scand J Resp Dis 56: 120-126, 1975. 16. Rosenblatt, M. B., Statement presented at Hearings before the Committee on Interstate and Foreign Commerce, House of Representatives, April 15 and May i, 1969. Serial No. 91-12, pp. 1255-1271. 17. Rosenblatt, M. B., Statement presented at Hearin'gs before the committee on Interstate and Foreign Commerce, House of Representatives, April 15-May i, 1969. Serial No. 91-12, p. 1257. 18. Heasman, M. A. and L. Lipworth, "Accuracy of Certification of Cause of Death. Studies on Medical and Population Subjects No. 20," H. M. Stationery Office, London, 1966, pp. i0, 12-13. 19. Bankl, H. and R. Krepler, "Das Klinisch Fehldiagnostizierte Karzinom (Untersuchungen am Obduktionsgut)," Wien Klin Wochenschr 83: 1-5, 1971. 20. Britton, M., "Clinical Diagnostics: Experience from 383 Autopsied Cases," Acta Med Scand 196: 211-219, 1974. 21. Hiyoshi, Y., et al., "Malignant Neoplasms Found by Autopsy in Hisayama, Japan, During the First Ten Years of a Community Study," J Natl Cancer Inst 59: 13-19, 1977. H9

HI0 22. 23. 24. 25. 26. 27. Rosenblatt, M. B., et'al., "Validity of Lung Cancer Mortality Data," Bull NY Acad Med 45(6): 519-527, 1969. Rosenblatt, M. B., et al., "Causes of Death in 1,000 Consecutive Autopsies," NY State J Med 71: 2189, September 15, 1971. Rosenblatt, M. B., et al., "Prevalence of Lung Cancer: Disparity Between Clinical and Autopsy Certification," Med Count 3(10): 53-54, 56-59, 63, 1971. Rosenblatt, M. B., "Diagnostic Accuracy in Cancer as Determined by Post Mortem Examination," Ariel, I. M. (ed.), Progress in Clinical Cancer, Vol. V, Grune & Stratton, Inc., 1973. Hoppe, R., "20 000 Mal Verdacht Auf Lungenkrebs," Prax Pneumol 31(10): 872-884, 1977. Burch, P. R. J., "Smoking and Lung Cancer: The Problem of Inferring Cause (with Discussion)," Read before the Royal Statistical Society, May 17, 1978.

ICOSI BACKGROUND BRIEFING PAPER Smoking and Cardiovascular Disease The opponents of smoking rely principally on data derived from statistical studies when they claim there is a causal relationship between smoking and cardiovascular disease. However, as many scientists have emphasized, statistical associations are not proof of causation. Rather, statistical associations point to the need for further research to ascertain the precise relationships between suspects and diseases. There is considerable scientific research that disputes such a causal assumption. The cause or causes of cardiovascular disease have not been established. As recently as 1976, the British Medical Journal acknowledged that an understanding of the causation and natural history of atherosclerosis is lacking.1 Scientists generally agree that cardiovascular disease is associated with many factors. These include dietary fats, obesity, high blood pressure, lack of exercise, heredity, and stress. Which factors, if any, play a causal role is still unclear. It may well be that some are not causes by themselves but actually characterize underlying traits of persons susceptible to cardiovascular disease. For example, individuals who choose a diet with high fat Ii

I2 intake may have a genetically-determined tendency toward cardiovascular disease. Numerous studies have in fact reported finding such genetic tendencies.2-II This genetic theory is further supported by data derived from the human twin studies conducted at the Karolinska Institute in Stockholm.12 Scientists there did not find an excess of cardiovascular disease deaths among the heavy smoking twins in identical twin pairs with different smoking habits. The researchers said their data suggest that smoking does not cause coronary heart disease, but rather that both smoking and heart disease are genetically determined.13 This theory is also supported by a recent Norwegian study which reported that the degree of risk of heart attack was dependent largely on family history and that cigarette consumption was not a significant factor.14 Some researchers contend that certain behavior patterns that have been classified according to personality type may be another important factor in determining whether an individual" will develop cardiovascular disease.15 These researchers divided their study groups into either Type A personalities -- highly competitive, time conscious, aggressive, confident, always working under pressure, Or Type B personalities -- more relaxed and less aware of the passage of time.16

Their studies showed that individuals classfied as Type A personalities tended to have higher cardiovascular disease rates than those classified as Type B personalities. A study of Swedish women reported a significant correlation between cardiovascular disease and Type A women, a type the author also found more common among smokers 17 than nonsmokers. Another study found that cardiovascular disease rates in heavy smokers of Type B personality were similar to or lower than those of nonsmokers and ex-smokers of Type A personality.18 It is difficult to reconcile trends in ca}diovascular disease mortality rates with the smoking causation theory. For example, in the United States, deaths from heart disease-- a broad category that includes both infectious and non- infectious diseases--have fallen slowly but steadily since 1950 for white males and since 1936 for white females. Even the non-infectious heart diseases that have been statistically associated with smoking have decreased in recent decades in both males and females.19 During this time, the popularity of cigarette smoking has increased. Also inconsistent with the smoking causation theory are studies indicating that stopping smoking does not necessarily result in a lower risk for cardiovascular disease I3

I4 among former smokers. A study of British doctors found that the incidence of cardiovascular disease among ex-smokers did 20-22 not decrease. Dr. Carl C. Seltzer of Harvard University, after examining many of the major studies, reported that men over 65 years of age who gave up smoking did not lower their risk for cardiovascular disease. Also, he said, in women age 65, the risk of cardiovascular disease rose upon cessation. 23-24 Some persons have claimed that carbon monoxide (CO) in cigarette smoke is responsible for causing cardiovascular disease in smokers. This claim is based largely upon studies by Professor Poul Astrup of the University Hospital in Copenhagen which reported that animals chronically exposed to carbon monoxide had arterial changes indistinguishable 25-30 from early atherosclerosis. However, very recently Astrup has reported that he has been unable to reproduce his 31 previous results. He concluded that his more recent study suggested that "no direct toxic effect of CO can be demonstrated.''32 It should be noted that CO is a normal body constituent created by metabolism. Without any exposure at all to external CO, the blood of an individual contains from 0.2% to 1.0% of carboxyhemoglobin (COHb) -- the combination formed by the gas with the red blood pigment.33 Many studies in the last half-century have indicated that a 10% COHb

I5 level, which people can reach when caught in heavy city traffic, is rarely even noticed. Although smoking may lead to small increases in the carboxyhemoglobin level in smokers, there is no scientific 34-37 evidence that proves these increases are harmful. An apt description of the smoking-cardiovascular disease situation may have been given by Dr. Theodore Cooper, former director of what is now the U. S. National Heart, Lung and Blood Institute, when he was asked whether cigarette smoking causes heart disease. Dr. Cooper responded that he considered smoking a risk factor but not a cause: Senator Hart: ". . I would merely ask if cigarette smoking causes heart disease?" Dr. Cooper: "No." Senator Hart: "It does not?" Dr. Cooper: "No.''38 May 1979

I6 References i. Editorial: "Progression and Regression of Athero- sclerosis," British Medical Journal i: 481-482, February 28, 1976. 2. Corday, E. and S. R. Corday, "Prevention of Heart Disease by Control of Risk Factors: The Time Has Come to Face the Facts," Amer J Cardiol 35(2): 330-333, 1975. 3o De Faire, U., et al, "Concordance with Respect to Mortality in Ischaemic Heart Disease and Cerebrovascular Disease, A Study on the Swedish Twin Registry," CVD Epidemiol Newsl 18(1): 21, 1975. Ibrahim, M. A., et al., "Screening of the Coronary Prone by Study of Offspring," HSMHA Health Rep 86(6): 577-582, 1971. McIlhany, M. L., et al., "The Heritability of Blood Pressure: An Investigation of 200 Pairs of Twins Using the Cold Pressor Test," Johns Hopkins Med J 136(2): 57-64, 1975. Murphy, E. A., "Genetics in Hypertension: A Per'spective," Circ Res 32(5) (Suppl. i): 129-138, 1973. Pesonen, E., et al., "Thickenings in the Coronary Arteries in Infancy as an Indication of Genetic Factors in Coronary Heart Disease," Circulation 51(2): 218- 225, 1975. Sholtz, R. I., et al., "The Relationship of Reported Parental History to the Incidence of Coronary Heart Disease in the Western Collaborative Group Study," Am J Epid 102(4): 350-356, 1975. Sidd, J. J., et al., "Coronary-Artery Disease in Identical Twins; A Family Study," New Enq J Med 274(2): 55-60, 1966. i0. Schrott, H. G., et al., "Increased Frequency of Coronary Heart Disease (CHD) in Relatives of Hyperlipidemic School Children," Circulation Suppl 52(4): 11-43, 1975. ii. Steinbach, M., et al., "Familial Clustering of Degenerative Cardiovascular Diseases," Rev Roum Med Intern 13 i) : 13-18, 1975.

I7 12. 13. 14. Cederlof, R., et al., "Morbidity Among Monozygotic Twins," Arch Environ'Health 10(2): 346-350, 1965. Cederlof, R., et al., "Respiratory Symptoms and 'Angina Pectoris' in Twins with Reference to Smoking Habits. An Epidemiological Study with Mailed Questionnaire," Arch Environ Health 13(6): 726-737, 1966. Cederlof, R., et al., "Hereditary Factors and 'Angina Pectoris,'" Arch Environ Health 14(3): 397-400, 1967. Cederlof, R., et al., "Hereditary Factors, 'Spontaneous Cough' and 'Smoker's Cough,'" Arch Environ Health 14(3): 401-406, March, 1967. Cederlof, R. and L. Friberg, "Tobacco Smoking and Health: Results of Epidemiologic Studies in Twins," Lakartidningen 65(27): 2727-2734, July 3, 1968. Cederlof, R., et al., "Cardiovascular and Respiratory Symptoms in Relation to Tobacco Smoking: A Study on American Twins," Arch Environ Health 18(6): 934-940, 1969. Cederlof, R., Statement presented at Hearings Before the Committee of Interstate and Foreign Con~erce, House of Representat~es, April 15 - May i, 1969. Serial No. 91-11, pp. 873-882. De Faire, U., et al., "Concordance with Respect to Mortality in Ischaemic Heart Disease and Cerebrovascular Disease, A Study on the Swedish Twin Registry," CVD Epidemiol Newsl 18(1): 21, 1975. Friberg, L., et al., "Smoking Habits of Monozygotic and Dizygotic Twins," Br Med J 1(5129): 1090-1092, 1959. Friberg, L., et al., "Mortality in Smoking Discordant Monozygotic and Dizygotic Twins," Arch Environ Health 21(4): 508-512, 1970. Friberg, L., et al., "Mortality in Twins in Relation to Smoking Habits and Alcohol Problems," Arch Environ Health 27(5): 294-304, 1973. F~rde, O. H. and D. S. Thelle, "The Troms~ Heart Study: Risk Factors for Coronary Heart Disease Related to the Occurrence of Myocardiai Infarction in First Degree Relatives," Am J Epidemiol 105: 192-199, 1977.

I8 15. Rosenman, R. H., et al., "Coronary Heart Disease in the Western Collaborative Group Study: Final Follow-up Experience of 8 1/2 Years," JAMA 233(8): 872-877, August 25, 1975. 16. Friedman, M. and R. H. Rosenman, "Type A Behavior Pattern: Its Association with Coronary Heart Disease," Ann Clin Research 3: 300-312, 1971. 17. Bengtsson, C., "Ischaemic Heart Disease in Women: A Study Based on a Randomized Population Sample of Women with Myocardial Infarction in Goteborg, Sweden," Acta Med Scand Suppl 549: 1-128, 1973. 18. Jenkins, C. D., et al., "Cigarette Smoking: Its Relationship 19. to Coronary Heart Disease and Related Risk Factors in the Western Collaborative Group Study," Circulation 38: 1140-1155, December 1968. 20. U. S. Public Health Service, Vital Statistics of the United States: Mortality, U. S. Department of Health, Education, and Welfare, National Center for Health Statistics, Washington. Seltzer, C. C., "Critical Appraisal of the Royal College of Physicians' Report on Smoking and Health," Lancet 1(7744): 243-248, 1972. 21. Seltzer, C. C., "Smoking and Coronary Heart Disease," New Eng J Med 288(22): 1186, 1973. 22. Seltzer, C. C., "Cigarette Smoking and Longevity in the Elderly," Med Count 6(2): 29-33, 1974. Seltzer, C. C., "Are There CHD Risks for Elderly Cigarette Smokers?", Paper presented at 26th Annual Scientific Meeting of the Gerontological Society, Miami Beach, Florida, November 6, 1973. 23. 24. Seltzer, C. C., "Smoking and Coronary Heart Disease in the Elderly," Am J Med Sci 269(3): 309-315, 1975. 25. Astrup, P., et al., "Enhancing Influence of Carbon Monoxide on the Development of Atheromatosis in Cholesterol- Fed Rabbits," J Atheroscler Res 7: 343-354, 1967. Wanstrup, J., et al., "Acceleration of Spontaneous Intimal-Subintimal Changes in Rabbit Aorta by a Prolonged Moderate Carbon Monoxide Exposure," Acta Path Microbiol Scand 75: 353-362, 1969. 26.

Kjeldsen, K., et~al., "Ultrastructural Intimal Changes in the Rabbit Aorta After a Moderate Carbon Monoxide Exposure," Atherosclerosis 16(1): 67-82, 1972. 28. Kjeldsen, K., et al., "Effects of Carbon Monoxide on Myocardium. Ultrastructural Changes in Rabbits After Moderate, Chronic Exposure," Circulation Research 34(3): 339-348, March 1974. 29. Thomsen, H. K. and K. Kjeldsen, "Threshold Limit for Carbon Monoxide-Induced Myocardial Damage. An Electron Microscopic Study in Rabbits," Arch Environ Health 29(2): 73-78, August 1974. 30. Thomsen, H. K. and K. Kjeldsen, "Aortic Intimal Injury in Rabbits: An Evaluation of a Threshold Limit," Arch Environ Health 30(12): 604-607, 1975. 31. Stender, S., et al., "The Effect of Carbon Monoxide on Cholesterol in the Aortic Wall of Rabbits," Atherosclerosis 28: 357-367, 1977. 32. Hugod, C., et al., "Influence of Carbon Monoxide on Intimal Morphology," Paper presented at the International Conference on Atherosclerosis, Milan, November 9, 1977. 33. Forbes, W. H., "Carbon Monoxide Uptake Via the Lungs," Ann NY Acad Sci 174(Art. i) : 72-75, October 1970. 34. Eckardt, R. E., et al., "The Biologic Effect from Long- Term Exposure of Primates to Carbon Monoxide," Arch Environ Health 25(6): 381-387, December 1972. 35. Hofreuter, D. H., et al., "Carboxyhemoglobin in Men Exposed to Carbon Monoxide," Arch Environ Health 4(1): 87-91, January 1962. 36. Kensler, C. J., "Components of Pharmacologic Interest in Tobacco Smoke," Ann NY Acad Sci 90(Art. I) : 43-47, September 1960. 37. Stupfel, M. and G. Bouley, "Physiological and Biochemical Effects on Rats and Mice Exposed to Small Concentrations of Carbon Monoxide for Long Periods," Ann NY Acad Sci 174(1): 342-368, 1970. 38. Cooper, T., Statement presented at Hearings Before the Subcommittee on Health of the Committee on Labor and Public Welfare, U. S. Senate, February 19, 1976, p. 183. I9

ICOSI BACKGROUND BRIEFING PAPER Jl Smoking and Pregnancy In the often emotional discussion of smoking and pregnancy, smoking has been blamed for a wide array of problems ranging from low infant birthweight to increased infant mortality rates. The scientific basis for such sweeping conclusions has been seriously questioned. Certain major studies have reported that children born to smoking mothers do not experience higher neo- 1-9 natal death rates than children born to nonsmoking mothers. An important study of maternal smoking habits and low birthweight infants was conducted by the late Professor J. Yerushalmy of the University of California. He concluded that the higher incidence of low birthweight infants born to the smokers was "due to the smoker, not the smoking.''10 Professor Carol Buck of the University of Western Ontario in Canada stated that Professor Yerushalmy "deserved great credit" for testing the causal significance of the reported association between low birthweight and smoking.11

In a letter published recently in The Lancet, a British medical journal, two British scientists reported that their data showed no effect of smoking on fetal growth. The authors concluded that although they do not "condone smoking under any circumstances, it does seem that there is a case for reopening the file on its effects on fetal growth.''12 May 19 79

J3 REFERENCES 3o 5o 8o Bailey, R. R., "The Effect of Maternal Smoking on the Infant Birth Weight," NZ Med J 71(456) : 293-294, 1970. Downing, G. C. and W. E. Chapman, "Smoking and Pregnancy: A Statistical Study of 5,659 Patients," Calif Med 104: 187, 1966. Kizer, A., "The Effect of Habitual Smoking on Pregnancy, Delivery, and the Newborn," Rev Obstet Ginec Venez 27(4): 595-643, 1967. O'Lane, J. M., "Some Fetal Effects of Maternal Cigaret Smoking," Obstet Gynecol 22(2): 181-184, 1963. Peterson, W. F , et al., " " and • Smoking Prematurity: A Preliminary Report Based on a Study of 7740 Caucasians," Obstet Gynecol 26(6): 775-779, 1965. Savel, L. E. and E. Roth, "Effects of Smoking in Preg- nancy: A Continuing Retrospective Study," Obstet Gynecol 20(3) : 313-316, September, 1962. Underwood, P. B., "Parental Smoking Empirically Related to Pregnancy Outcome," Obstet Gynecol 29(1) : 1-8, 1967. Yerushalmy, J., "Mother's Cigarette Smoking and Survival of Infant," Obstet Gynecol 88(4): 505-518, 1964. 9. Yerushalmy, J., "The Relationship of Parents' Cigarette Smoking to Outcome of Pregnancy; Implications as to the Problem of Inferring Causation from Observed Associations," Am J Epidem 93(6): 443-456, 1971. i0. Yerushalmy, J., "Infants With Low Birth Weight Born Before Their Mothers Started to Smoke Cigarettes," Am J Obstet Gynecol 112(2): 277-284, 1972. ii. 12. Buck, C., "Popper's Philosophy For Epidemiologists," Int J Epidem 4(3): 159-168, 1975• Alvear, J. and O. G. Brooke~ "Effect of Smoking on Fetal Growth," Lancet I: 1158, May 28, 1977.

ICOSI BACKGROUND BRIEFING PAPER K1 Smoking and Youth The manufacturers of tobacco products adhere to the premise that smoking is an adult custom. The decision to smoke or not is best made by mature, informed persons. It is reasonable to educate the public, in particular youth, about scientific findings in connection with smoking. If governments, therefore, decide that campaigns of educa£ion and information are desirable, such campaigns must be based on sound scientific grounds. A balanced campaign of information should point out all known risk factors to human health and by this means show the public, particularly young people, that a multitude of risk factors may affect their health. Undue emphasis on the alleged hazards of smoking would create the false impression that the single act of not smoking would prevent disease. May 1979

L1 ICOSI BACKGROUND BRIEFING PAPER The use of the terms "evidence" and "proof" in papers relatinq to smoking and health American jurisprudence (second edition, 1967), a law encyclopedia, shows the following definitions for the terms "evidence" and "proof" : Evidence -- Proof -- (T)he legal acceptation of the word imports the means by which any matter of fact the truth of which is submitted to an investiga- tion may be established or disproved ... the object of all evidence is to inform the trial tribunal of the material facts which are relevant as bearing upon the issue, in order that the truth may be elicited and that a fair determination of the controversy may be reached. .~ (T)he effect of evidence or ... the establish- ment of a fact by evidence. Proof results as the probative effect of evidence and is the conviction or persuasion of the mind resulting from the consideration of the evidence. It is merely that quantity of evidence which produces a reasonable assurance of the existence of the ultimate fact. However, although the above definitions clearly distinguish .. between evidence and proof, the terms are often used synonymously. Among the several definitions of proof in Black's Law Dictionary (fourth edition, 1968) are the following : -- (T)he words "proof" and "evidence" may be used interchangeably. -- No material difference between terms exists for purposes of instruction. Evidence and proof are also listed as synonyms in Roget's Thesaurus (fourth edition, 1977). In smoking and health research, any finding relating to a specific claim may be considered evidence. Thus, on most issues it is inaccurate to say "there is no evidence." Since proof and evidence 036'7841.5

L2 are frequently defined and often used synonymously, the phrase "there is no proof" should also not be used in reference to smoking and health, claims, unless it would be accurate to say "there is no evidence." "It has not been proven" or "it has not been established" are usually more accurate terms to use in the smoking and health arena. The terms "proven" and "established", suggest certainty or the possession of conclusive evidence to substantiate a claim. May 1979