[ Pobierz całość w formacie PDF ]
. Thecommittee did not observe that no simple cause-and-effect relationship existsbetween any etiological factor, including bacterial pathogens, and a disease.31Many of the committee s conclusions reflected its ambivalence aboutstatistical correlations.It concluded that  cigarette smoking is causally re-lated to lung cancer in men and cited the dose-response relationship:  TheCigarette Smoking and Statistical Correlations 249risk of developing lung cancer increases with duration of smoking and thenumber of cigarettes smoked per day, and is diminished by discontinuingsmoking. It then stated that  the data for women, although less extensive,point in the same direction, but did not explain why the statistical find-ings for men could not be generalized to women.The committee s difficul-ties were exacerbated in its conclusions regarding coronary heart disease,where it refused to accept causal relationships:It is established that male cigarette smokers have a higher death rate fromcoronary artery disease than non-smoking males.Although the causativerole of cigarette smoking in deaths from coronary heart disease is not proven,the Committee considers it more prudent from the public health viewpointto assume that the established association has causative meaning than tosuspend judgment until no uncertainty remains.32Subsequent Surgeon General s committees on smoking were morereceptive to statistical correlations and even borrowed the concept of therisk factor from life insurance in order to stress the multifactorial nature ofdisease etiology.The Surgeon General s second report in 1967 evaluateddevelopments in the three years since the first report:Prospective morbidity studies confirm the relationship between cigarettesmoking and coronary heart disease.These studies also provide the opportu-nity to evaluate the effect of smoking independently and in combinationwith other known  risk factors, such as high blood pressure and high serumcholesterol that are also important in the pathogenesis of coronary heartdisease.It has been demonstrated that cigarette smoking not only operates asan independent  risk factor but that it may combine with other  risk fac-tors to produce even greater effects on cardiovascular health.33The 1968 report went even further:  The acceptance of a multiplefactor causation hypothesis for coronary heart disease emphasizes the needfor more sophisticated statistical analyses of appropriate data. It addednew risk factors for coronary heart disease, including  sociological, psycho-logical, and personality variables, as well as  genetic and constitutionalfactors. The 1968 report included new kinds of statistical analyses and the1971 report extended the concept of risk factors to peripheral vascular dis-ease and other diseases associated with smoking.34A multifactorial concept of disease etiology dominated the 1986 U.S.Public Health Service report, The Health Consequences of Involuntary Smok-ing, which concerned the inhalation of tobacco fumes released into the air250 Risk Factors and Coronary Heart Diseaseby burning cigarettes.Based on statistical dose-response relationships, thereport concluded that  involuntary smoking is a cause of disease, includinglung cancer, in healthy nonsmokers. Concepts of causality were not men-tioned and the legitimacy of epidemiological and statistical research meth-ods never questioned.The transformation in concepts of disease etiologybetween the 1964 and the 1986 reports was as great as that produced bythe germ theory of disease a century earlier.35Public Policies Concerning SmokingFor many years government health agencies, voluntary health associations,and the health professions also believed that statistics could not establish acausal relationship between smoking and disease.Wynder, who was a leaderin the anti-smoking movement, reminisced in 1988:  In retrospect, it isdifficult to comprehend why it took health professionals and society solong to grasp the full extent of the causative association between lung can-cer and smoking. Recognizing the importance of economic interests, hecontinued:  The position of the tobacco industry is understandable as is itsinfluence on groups depending on its financial support, such as the media,and even governments. But, he insisted, economic interests could not ex-plain why most of those directly involved in health activities failed to act.36The political and economic influence of the tobacco industry was asignificant obstacle for the anti-smoking movement.About 1970, tobaccofarming was a $1 billion industry and tobacco sales a $10 billion consumerproduct that produced $4 billion in tax revenues to federal, state, and localgovernments.Tobacco farming and cigarette production were concentratedin a few states, which increased their political power.Cigarette sales con-tributed to the earnings of hundreds of thousands of retailers.Tobacco adver-tising was a major source of revenue for newspapers, magazines, billboards,and radio, and generated 8% of television advertising revenues about 1970.Advertising agencies remained loyal to the industry: in 1969 19 of 23 NewYork agencies without cigarette accounts said they would unhesitatingly acceptthem.The leading voluntary health associations relied on the mass media topresent their messages without cost and received support from companies thatbenefited from cigarette sales.Medical societies at the national, state, and locallevels were sensitive to economic interests.37The cigarette industry was a highly cohesive economic and politicalforce dominated by six firms.The larger companies established pricingpolicies that were adopted by the smaller ones, assuring high profits for all.Cigarette Smoking and Statistical Correlations 251Competition occurred primarily through advertising and the introductionof new brands and products.The patterns of close cooperation enabled thecompanies to adopt a unified strategy in contesting the evidence.38The tobacco industry appealed to the ideological views of many bio-medical scientists and physicians by funding laboratory investigations anddenigrating epidemiological studies.In 1954 it created the Tobacco Indus-try Research Committee (renamed the Council for Tobacco Research in1964) to sponsor research, and its director promised  complete freedom ofthought and action to investigators.The committee proposed  glandulardisturbances as causes of both heavy smoking and cancer, a hypothesiscomfortably amenable to years of futile but well-funded laboratory investi-gation by credulous biomedical scientists [ Pobierz całość w formacie PDF ]