Authors: Richard Kluger
As long as Baker ran the NCI etiology program, he kept a fairly tight rein on Gori. But when he succeeded Kenneth Endicott as director of NCI, his place running etiology and monitoring the Smoking and Health Program was taken by Frank Rauscher, who felt that the SHP was well launched and that Gori was qualified to keep it moving forward. “I never viewed him as a scientist,” Rauscher said of Gori, “but I was impressed by his managerial skill.”
After the passage of the National Cancer Act in 1971, NCI’s budget began taking quantum annual leaps that eased residual grumbling within the institute that the smoking program was siphoning needed funds from more worthy efforts. Increasingly Gori would socialize with Tobacco Working Group members, some of whom—like Wynder, Kensler, and the brilliant chemist Philippe Shubik, who ran a large laboratory at the University of Nebraska—were also prime contractors for the less hazardous cigarette program. Such a cozy arrangement, which under ordinary circumstances would have run afoul of conflict-of-interest standards, was countenanced in part because of the difficulty involved in finding suitable contractors and in part because the work turned out by their labs was of superior quality. And it should have been, since they were for the most part merely replicating their own earlier work but now using the SHP’s experimental cigarettes to conduct their studies. Nor did it go unnoticed that Gori began to affect a few of the traits of his role model, Wynder, driving a sporty car and traveling first-class while other NCI officers settled for coach. His overall conduct earned him the enmity of an Italian countryman at NCI with far more impressive credentials—Umberto Saffiotti, Rauscher’s assistant director in the carcinogenesis program and highly cautious about the efficacy of a “safer” cigarette. Saffiotti, among others, thought of Gori more as an impresario than as an objective manager of his contractors, playing the booster and even the propagandist for a program he was anxious to build into his own bailiwick.
Characteristic of Gori’s tub-thumping were remarks he delivered at the second world conference on smoking and health held in London in September 1971. By way of assuring his listeners that the goal of the NCI effort was no abstract construct, he said that based on “epidemiological studies, smokers of filter cigarettes delivering less tar and nicotine show a remarkably decreased risk of disease; these studies give unequivocal proof in man that reduced tar and nicotine provide a first model of a less hazardous cigarette.” But two of the three studies Gori cited in support of this gross overstatement had been led by Wynder and Oscar Auerbach, neither of whom offered such definitive judgments, and the third study, by I. D. Bross, reported a significant reduction in risk within two years after his subjects switched to lower-yielding brands—a claim so premature as to be dubious on its face.
Gori’s gift for currying favor with Congress earned him enhanced power at NCI. But critics feared that he had grown too chummy with the SHP contractors he was supposed to monitor: Gori’s name appeared as co-author of some twenty articles growing directly out of the contract work under his supervision. Such publications fed concerns that Gori and his contractors enjoyed a mutually useful and manipulative relationship rather than one in which he stood apart from them as a rigorous arbiter of their work and the progress of the whole enterprise.
Gori extended his overreach. By the spring of 1974, remarks attributed to him in the unofficial
Cancer Newsletter
stated that the carcinogenic properties of tobacco smoke had been identified, and when reduced, “less cancer will result.” He was further quoted as saying that the tobacco industry would be able to produce a cigarette “in no more than three years that could be smoked in moderation (up to a pack a day) with no harmful effects.”
That was too much for Saffiotti, who, along with other circumspect investigators, saw lung cancer as an immensely complex disease in which each individual’s genetic code and systemic vulnerability came into play along with environmental and occupational variables—all factors that, as Saffiotti later put it, “potentiate one another” and undercut the notion of panaceas. More research was required, along with confirming bioassays conducted over an adequate time span, Saffiotti asserted, dashing cold water on Gori’s glowing 1974 picture of SHP’s progress, and added, “Until we have those data, it is dangerous to lead people to believe there is a safer cigarette.” Another influential NCI scientist, TWG member Marvin Schneiderman, was equally unhappy with Gori’s facile evaluation, which he thought benefited the tobacco industry’s marketing program. “It’s not up to us to find a product they can market,” he declared. “Are we the research arm of the tobacco industry?”
Gori’s position was by now strong enough to weather such criticism. Saffiotti was regarded by some at NCI as a classic academic stickler who needed every last bit of rock-solid evidence in place before venturing a judgment, while Gori was arguing that public-health considerations demanded action as soon as possible, even if science had not yet reached a definitive conclusion. Still, Gori’s remarks were of dubious scientific soundness. Nobody could say that smoking up to a pack of cigarettes a day constituted “moderation,” and his endorsement of a threshold level of carcinogenic dosage—“no harmful effects”—was a hotly debated concept. Uneasiness persisted over Gori’s incessant and as yet unwarranted optimism, and tensions surrounding his perceived deference to the industry within the SHP occasionally overflowed. The most notable occasion was a TWG meeting in October of 1974, when remarks by Robert Hockett, associate scientific director of the Council for Tobacco Research, in defense of the industry’s resistance to the health charges against it, were assailed by member Philippe Shubik. “It shocks me that twenty years
later,” Shubik told the startled meeting, “you have not joined the community of men. You will go down in history denying facts well known to the scientific community.” It angered him, Shubik thundered, that people “who are scientists and know better, offer the arrant nonsense that cigarettes are not a health threat,” and he doubted that any among them truly believed that smoking was not “the primary cause of lung cancer.”
By 1976, Gori had outflanked Saffiotti, his ranking
bête noire
within NCI, and taken over the added post of deputy director of the Division of Cancer Cause and Prevention. But in the December 17 number of
Science
, Gori crossed his Rubicon with an article bearing the almost sensational title of “Low-Risk Cigarettes: A Prescription.” In it, he reported that the technology was now in place to reduce toxic components in cigarette smoke, and thus, “The feasibility of less hazardous cigarettes raises the question of whether there are limits of cigarette and smoke composition that may approach relative safety”—a threshold level, though Gori abstained from using the term. Such limits, he went on, “can be defined as smoke intake dosages at which the risk of disease in smokers approaches that of nonsmokers.” These “critical values,” as Gori termed them, could be estimated by dose-response analysis of various epidemiological studies, including those that formed the vital bases for the 1964 report to the Surgeon General. That threshold range for all smoke-caused diseases, he calculated, was a dosage equivalent to what a smoker derived from consuming two pre-1960 cigarettes, each containing 43 mg. of tar and 3 mg. of nicotine, per day or a daily pack of a brand yielding 4.3 mg. of tar and 0.3 mg. of nicotine per cigarette. The tables providing these figures were accompanied by extensive footnotes with mathematical formulas and abstract analytical analyses showing how he had reached these “critical values”.
But having taken pains to establish the grounds for his calculations in defining “low-risk cigarettes,” Gori then befogged the whole argument by stating, “It would be erroneous to interpret these critical values as indicators of safe smoking levels, since the experimental and statistical uncertainties of the [supporting epidemiological] studies are well known … ”—he had alluded earlier in the article to their differences with respect to design, quality of data, number of subjects, and demographic characteristics. Those same difficulties had confronted statistician William Cochran when he assessed and synthesized them in making his essential contribution to the original Surgeon General’s report. But these studies either were valid or were not valid in calculating the relative risks of consuming given dosages of tobacco smoke; Gori seemed to want to have it both ways. The same tendency confused his discussion of whether cigarettes satisfying his “critical values,” which he equated with “the maximum number of cigarettes that the average individual could smoke daily without apparently increasing his expected risk of mortality significantly above that of the nonsmoker,” were in fact safe or not. Those qualifying words “apparently”
and “significantly” were his hedges. “Uncertainty, however,” Gori went on, “should not be allowed to dilute the implication of these data, namely, that a rapid shift in cigarette consumption habits toward the proposed range of critical values would make it reasonable to expect that the current epidemic proportions of smoking-related disease could be reduced to minimal levels in slightly over a decade.” But wasn’t the certainty or uncertainty of the data the heart of the matter in determining whether one should risk smoking even relatively diluted cigarettes?
To understand the thin ice Gori was skating on, one may compare his article with another appearing at the same time, on the same subject, and pointing in the same general direction—“ Tar’ and Nicotine Content of Cigarette Smoke in Relation to Death Rates” in
Environmental Research
by an American Cancer Society team of epidemiologists including Cuyler Hammond, Lawrence Garfinkel, and Herbert Seidman. Drawing their conclusions from a study of more than a million subjects over a dozen years starting in 1959, the investigators reported that for those smoking the same number of cigarettes each day, users of lower-yielding cigarettes (defined as less than 17.6 mg. of tar and 1.2 mg. of nicotine) showed a 16 percent overall lower mortality risk than those using higher-yielding brands (of 25.8 mg. of tar and 2.0 mg. of nicotine or more), a 14 percent lower risk of heart disease, and a 26 percent lower risk of lung cancer. This was a far cry from claiming a reduced risk level approaching that of nonsmokers; indeed, the data in this latest ACS study explicitly stated that nonsmokers had a 34 percent lower overall mortality risk than smokers of lower-yielding cigarettes, a 40 percent lower risk of heart disease, and an 85 percent lower risk of lung cancer. “It is quite apparent that the reduction in the tar and nicotine content of cigarette smoke did
not
make cigarette smoking ‘safe’ for the men and women in this analysis,” Hammond and his colleagues declared—and their data were based on actual smoking histories, not arithmetical computations drawn from hypothetical constructs about total daily dosages.
To be sure, Gori had his defenders, especially among industry-paid scientists and consultants like Charles Kensler of Arthur D. Little, who felt that Gori was unfairly criticized for simply following through on the ideas of Wynder and other ardent supporters of the less hazardous cigarette. Whether there was such a thing as a threshold level of safety in carcinogenic dosage was a matter of debate and opinion, Kensler argued, “but Gori knew all this and said, ‘Let’s look at the data,’ and since all the population studies had dwelt on dose-response as the key supporting correlation, Gori focused on the bottom end of the exposure spectrum, where disease creation was minimal.” To Kensler, Gori was simply being logical in pushing hard for attenuated cigarettes. “It was a damned courageous thing for him to do … [yet] all the establishment
outfits jumped on him,” primarily because Gori’s program was viewed as perpetuating smoking, in Kensler’s view.
At the end of January 1977, a month after his article ran in
Science, The New York Times
featured Gori’s program in an article headlined “U.S. Trying to Help Out with a ‘Safe’ Cigarette.” It spoke of the 150 varieties of experimental cigarettes that NCI had developed for the purpose of finding one within what Gori now termed “the tolerable range of risk.” Indeed, he elaborated, there were some “ultra-low” tar brands already on the market—none was named—that if smokers used them exclusively, “We’d be in good shape, we’d have no problem.” Did that mean they were safe to smoke? It surely sounded that way. But Gori put in his disclaimer: The only truly safe cigarette, he said, was “the one you don’t smoke.” Then why all this dancing around about “tolerable” levels of risk? The
Times
answer came in the comment that Gori and his project “are eager and compassionate propagandists for weaning the incurable smoker from high-risk smoking to low.”
But who yet knew if the ultra-low-yielding brands on the market that Gori referred to would leave their users with, as he put it, “no problem”? Or whether the effect of the lower tar and nicotine levels, as recorded by smoking machines, was lost through the real-life habits of smokers? You could get a lot higher yield from a “low-yield” cigarette if you took more puffs and inhaled them longer and more deeply. And there were other ways to foil such manufacturers’ devices for making cigarettes less dangerous as the addition of ventilation holes around the filter to dilute the delivery of carbon monoxide. Psychologist Lynn T. Kozlowski discovered that some smokers kept all those little holes covered with their fingers in order to derive the maximum flavor kick out of each puff. Such hole-blocking, Kozlowski reported in the November 1980 number of the
American Journal of Public Health
, could increase the yield of toxic by-products of smoking by as much as 300 percent. And who knew how much greater the increase was, if any, in toxicity from flavorings or anything else that cigarette makers felt they had to add—without reporting them to any outside regulatory body—to make the taste of their allegedly less hazardous cigarettes more acceptable to smokers accustomed to “full flavor”? A whole new generation would have to come of age before epidemiologists could track those smoking only lower-yielding brands and compare their mortality rates with those of nonsmokers and smokers of higher-yielding brands in order to calculate risk levels of allegedly less hazardous cigarettes. That did not mean that, in the interim, the effort to cut down yields was unworthy—only that high caution ought to be exercised in speaking of the promise such a modified cigarette might hold for smokers.