by Allan Brandt
Such an ecumenical conclusion was precisely what Hill & Knowlton and the industry sought. Garland’s conclusion reads as if it had been written at least a decade earlier, prior to the publication of extensive peer-reviewed studies from epidemiology, pathology, and laboratory investigation. He had reached far more definitive conclusions about the harms of smoking as early as 1953, when he wrote that the previous year’s scientific results from Doll and Hill had provided “an association between cigarette smoking and lung cancer so strong as to be considered proof within the everyday meaning of the word.”133
One can only conclude from the dispassionate tenor of Garland’s 1961 editorial, placing the harms of smoking back into the domain of doubt, that the tobacco industry had gotten its money’s worth from Hill & Knowlton.
III
POLITICS
No reasonable person should dispute that cigarette smoking is a serious health hazard.1
SURGEON GENERAL
LUTHER TERRY, 1964
In the meantime (we say) here is our triple, or quadruple or quintuple filter, capable of removing whatever constituent of smoke is currently suspect while delivering full flavor—and incidentally—a nice jolt of nicotine. . . . And if we are the first to be able to make and sustain that claim, what price Kent?2
ADDISON YEAMAN, 1963
GENERAL COUNSEL AND
VICE PRESIDENT,
BROWN & WILLIAMSON
We believe there is no connection or we wouldn’t be in the business.3
JAMES C. BOWLING, 1963
PHILIP MORRIS
Doubt is our product, since it is the best means of competing with the “body of fact” that exists in the mind of the general public.4
BROWN & WILLIAMSON, 1969
CHAPTER 7
The Surgeon General Has Determined
IN THE FACE OF THE “continuing controversy” over the health effects of cigarettes—widely and loudly broadcast to an expanding I media by the TIRC—researchers and public health representatives tried to make the public aware of the actual state of scientific knowledge. They expressed growing exasperation at the toxic combination of industry denials and reassuring claims for new filtered cigarettes—which radically distorted popular understanding of the harms of smoking. The industry, meanwhile, well understood that its carefully maintained posture of scientific uncertainty provided a shield against new regulatory initiatives.
In 1960, most public health policies still centered on the control of infection. From compulsory vaccination to new mechanisms of case reporting and tracking, the major interventions of public health had focused on the use of state police powers to limit communicable diseases.5 But by mid-century, systemic chronic diseases had overtaken infection as the major causes of death, and public health officials were forced to adjust their priorities. The control of “noncommunicable” diseases posed a new and entirely different set of problems. The identification of the cigarette as a cause of serious disease marks a critical turning point in the history of public health.6
Many public health officials, though committed to the notion that smoking caused disease, were nonetheless unsure of how to approach the problem. Was it the province of public health to regulate personal behaviors, especially those deemed of little risk to others? For behavioral risks, such as smoking, the role and authority of public health officials had yet to be defined. Throughout the twentieth century, the state’s responsibilities in disease detection, prevention, and health promotion had centered on the notion of communicability. Thus, the state was charged with the identification of infectious organisms, contact tracing of infected individuals, and, when appropriate, the detention or quarantine of those who might pose risks to the wider public.7 These activities and the boundaries of public health had, in general, been scrupulously policed by the medical profession to assure that all aspects of clinical care (and remuneration) remained the hallowed prerogative of physicians.
Now, however, the uneasy relationship between public health and medicine would receive new scrutiny. Tobacco explicitly raised the question about the role of public health officials in addressing important health risks. Many within public health were reluctant to enter the exclusive turf of clinical medicine by addressing matters of individual behavior.8 But as the importance of cigarettes as a cause of illness and death became increasingly apparent, these officials began to recognize that like it or not, smoking was already a public health issue. In 1956, at the urging of U.S. Surgeon General Leroy Burney, a study group on smoking and health was organized by the American Cancer Society, the American Heart Association, the National Cancer Institute, and the National Heart Institute. This group met regularly to assess the scientific evidence relating to tobacco and health. In its 1957 report the group noted that sixteen studies had been conducted in five countries, all showing a statistical association between smoking and lung cancer. Collectively, these studies demonstrated that• lung cancer occurs five to fifteen times more frequently among smokers than nonsmokers
• on a lifetime basis, one of every ten men who smoke more than two packs a day will die of lung cancer
• cessation reduces the probability of developing lung cancer9
These epidemiological findings were supported by animal studies in which malignant neoplasms had been produced by tobacco smoke condensates. Further, human pathological and histological studies added evidence to strengthen the “concept of causal relationship.” “Thus,” the authors explained, “every morphologic stage of carcinogenesis, as it is understood at present, has been observed and related to the smoking habit.” All three domains of scientific and medical knowledge now confirmed the epidemiological findings of the early 1950s. The significance of the data was powerful:The sum total of scientific evidence establishes beyond reasonable doubt that cigarette smoking is a causative factor in the rapidly increasing incidence of human epidermoid carcinoma of the lung.
The authors went on to call for a public health response to these findings: “The evidence of a cause-effect relationship is adequate for considering the initiation of public health measures.” Although “additional research is needed to clarify many details and to aid in the most effective development of a program of lung cancer control,” they felt that public health initiatives did not have to wait for this supplemental work to be completed.10 They explicitly addressed the relationship of scientific knowledge to public health action. After these conclusions were published, Surgeon General Burney released a statement on behalf of the Public Health Service (PHS) that “there is increasing evidence that excessive cigarette smoking is one of the factors which can cause lung cancer.”11
But Burney and his colleagues soon questioned whether such statements were sufficient, especially when they were so vigorously contested by the powerful combination of tobacco public relations and marketing. What could the Public Health Service, with its limited authority and limited budget, do to reduce disease associated with smoking? It could provide a systematic and definitive assessment of the evidence. And it could challenge the disinformation campaign waged by the industry. Given ongoing public confusion and the widespread uncertainty among physicians about what to tell their patients, the PHS began evaluating its role in addressing the harms of smoking. Though relatively weak, the PHS could still command the expertise to strike a blow against science-by-public relations.
Certainly, it was within the purview of public health to assess medical and scientific evidence and make these evaluations available to the public. But Burney’s review raised essential questions about what additional steps might be taken. Burney’s staff was characteristically sensitive about the implications of moving public health into the realm of advocating changes in personal risk behaviors. If public health were viewed as usurping medicine’s traditional role of counseling patients about how to avoid disease, it could easily incur the wrath of well-oiled and well-heeled organizations like the American Medical Association. Moreover, the very idea of risk reduction, simply because it was a ne
w development in health care, was contested terrain in the long-standing divide between public health and medicine.12
At a February 1956 meeting to plan the PHS policy regarding the tobacco findings, two views emerged. Most of the surgeon general’s staff concluded that it was time to “get this message out,” especially to school-children. Chief assistant to the surgeon general, James Watt, however, expressed concern that the PHS should avoid “a missionary statement . . . designed to influence personal habits.” Lewis Robbins, a physician in the PHS with responsibility for relations with the medical profession, shared some of Watt’s concerns. According to Robbins, “Public health should never take a position which gets ahead of the medical profession.” To the PHS, the boundary between clinical care and public health was something to be very closely watched. Among other reasons, their modest funding was dependent on a Congress heavily lobbied and supported by the representatives of organized medicine. Robbins estimated that only half of all practicing physicians had become convinced of the relationship between smoking and cancer. As a result, he concluded that the surgeon general should work to influence the “scientific community through appropriate channels.”13 These anxieties led a weak and uncertain public health community to conceive only a limited notion of its role in one of the biggest health issues of the century.
As a result, consensus panels about the harms of smoking proliferated, each offering state-of-the-art assessments of the evidence. In January 1959, yet another distinguished group of cancer researchers, led by statistician Jerome Cornfield of the NCI, offered a substantive review of the accruing evidence linking cigarettes to lung cancer. Cornfield and colleagues also noted that the persistent “debate” about the science was driven by the tobacco industry:It would be desirable to have a set of findings on the subject of smoking and lung cancer so clear-cut and unequivocal that they were self-interpreting.
The findings now available on tobacco, as in most other fields of science, particularly biologic science, do not meet this ideal. Nevertheless, if the findings had been made on a new agent, to which hundreds of millions of adults were not already addicted, and on one which did not support a large industry, skilled in the arts of mass persuasion, the evidence for the hazardous nature of the agent would generally be regarded as beyond dispute.14
Just as the tobacco industry in the late nineteenth century had developed the technology for the mass production of cigarettes, so now it had developed techniques for the mass production of controversy and doubt. The industry insisted on scientific criteria that it knew full well could not be attained then, or ever—when it was still willing to admit such criteria at all. C. C. Little, as we saw, explicitly refused to do so. As Cornfield explained, despite the impressive data—not only from epidemiology but from laboratory and clinical science—bringing the controversy to resolution would prove no easy matter.
To confront the doubt fomented by the industry, Cornfield’s group carefully considered the range of alternative hypotheses to account for the significant rise in cases of lung cancer. They scrupulously read the standard critiques offered by skeptics, such as Joseph Berkson and Ronald Fisher, knocking down each question in turn. Their conclusions went even further than the 1957 report initiated by Surgeon General Burney. “The consistency of all the epidemiologic and experimental evidence,” they wrote, “also supports the conclusion of a causal relationship with cigarette smoking. . . .”15 Like the 1957 study group, these authors felt an imperative for action, stressing that the available findings were “sufficient for planning and activating public health measures.”16
Following the Cornfield article, Surgeon General Burney offered yet another comprehensive assessment. He revisited the epidemiologic data as well as the animal and pathological investigations, and came to the following categorical conclusion, published as the official “Statement of the Public Health Service”:The Public Health Service believes that the following statements are justified by studies to date [1959]:
The weight of the evidence at present implicates smoking as the primary etiological factor in the increased incidence of lung cancer.
Cigarette smoking particularly is associated with an increased chance of developing lung cancer.17
In Burney’s view, the evidence of cigarette smoking’s harms was overwhelming and certainly worthy of government attention. Nonetheless, the TIRC continued to disparage such consensus statements. TIRC Scientific Director Little asserted, “scientific evidence is accumulating that conflicts with, or fails to support, the tobacco-smoking theories of lung cancer.”18
Burney was especially disappointed by the response at JAMA. An editorial, actually drafted in the surgeon general’s office with some modifications, shaded Burney’s message, much to his chagrin. “A number of authorities who have examined the same evidence cited by Dr. Burney do not agree with his conclusions,” argued John Talbott, editor of JAMA.19 Not surprisingly, JAMA offered a defense of clinical authority, emphasizing the role of the physician as guide and counselor:Neither the proponents nor the opponents of the smoking theory have sufficient evidence to warrant the assumption of an all-or-none authoritative position. Until definitive studies are forthcoming, the physician can fulfill his responsibility by watching the situation closely, keeping au courant of the facts, and advising his patients on his appraisal of those facts.20
Even as the industry attempted to obscure scientific results on cigarettes and lung cancer, government agencies had begun to recognize and publicize the cigarette’s harms. American scientists and public health officials who were convinced by the evidence were joined by colleagues in other nations. Between 1957 and 1962, the Medical Research Council of Great Britain, the Royal College of Physicians, the World Health Organization, and public health officials in the Netherlands and Norway publicly acknowledged that cigarette smoking caused lung cancer.21 The British Royal College of Physicians, after a two-year investigation, stated, “Diseases associated with smoking now cause so many deaths that they present one of the most challenging opportunities for preventive medicine today.”22 The report concluded:The strong statistical association between smoking, especially of cigarettes, and lung cancer is most simply explained on a causal basis. . . . The conclusion that smoking is an important cause of lung cancer implies that if the habit ceased, the death rate from lung cancer would eventually fall to a fraction, perhaps to one fifth or even, among men, to one tenth of the present level. Since the present annual number of deaths attributed to lung cancer before the age of retirement is some 12,000 . . . a large amount of premature shortening of life is at issue.23
The Royal College of Physicians clearly recognized that lives were at stake. Repeatedly, independent critical evaluation of the scientific findings that cigarettes caused lung cancer reached the same conclusion.
The Royal College of Physicians’ report marked a crucial step in the legitimation of the link between cigarettes and disease.24 E. Cuyler Hammond, director of the statistical research section of the American Cancer Society, wrote the preface to the American edition. He noted the esteem in which the Royal College was held and pointed out that “the reader is asked to accept nothing on faith.” The huge amount of scientific research presented in the report, Hammond explained, provided “evidence from which [readers] can draw their own conclusions concerning the effects of cigarette smoking.”25
That such statements had to be repeated so many times reflected the power and resources the tobacco industry committed to the production of uncertainty in the face of knowledge. Each new research report implicating tobacco as a cause of disease elicited a denial that any conclusive proof had been found. Typical of such volleys was Little’s response to Burney’s statement of 1957:The Scientific Advisory Board questions the existence of sufficient definitive evidence to establish a simple cause-and-effect explanation of the complex problem of lung cancer.26
Such language—typical of TIRC statements—reflected the careful wordsmithery of the experts at Hill & Knowlton. Burney had not
claimed that the evidence was “definitive” nor that the relationship of smoking and lung cancer was “simple.” Nor would he have disagreed that the problem of lung cancer was “complex.” But such equivocations did not alter the fact that smoking constituted a demonstrated health risk of great significance.
News accounts of new medical findings were generally accompanied by a statement from the TIRC insisting that “nothing new” had been found and that the studies were “merely” statistical. The TIRC was very effective in mobilizing a relatively small group of skeptics and amplifying their views as if they were equal in numbers and significance to the broad scientific consensus. But these skeptics, and the tobacco industry that trumpeted their views, produced no new research.
These persistent industry denials helped to generate a major innovation in medicine and public health: the consensus report. To a degree, unprecedented in the history of medicine, thorough and objective statements reviewing the findings gained increasing significance to medical and public health groups wishing to acknowledge resolution in the face of the widespread perception of an ongoing scientific “controversy.” The development of consensus reports would have long-range implications for establishing public health knowledge, clinical guidelines, and what would eventually come to be known as evidence-based medicine.27 Consensus reports typically seek to systematically and critically evaluate data and come to a conclusion about its medical and public health significance. They ask what is known and what is the best practice. In the case of smoking, the government found itself responsible for adjudicating a scientific dispute while the industry sought to maintain the notion that adjudication was the domain of individual clinicians, smokers, and would-be smokers. Irresolution was crucial to the industry’s interest; uncertainty the basis of its future livelihood.