Tobacco Documents Online

RESEARCH ADVANCES IN ALCOHOL AND DRUG PROBLEMS Series Editors Reginald G. Smart Howard D. Cappell Frederick B. Glaser Yedy Israel Harold Kalant Robert E. Popham Wolfgang Schmidt Edward M. Sellers A Continuation Order Plan is available for this series. A continuation order will bring delivery of each new volume immediately upon publication. Volumes are billed only upon actual shipment. For further information please contact the pub- lisher. Research Advances in Alcohol and Drug Problems Volume 8 Edited by Reginald G. Smart, Howard D. Cappell, Frederick B. Glaser, Yedy Israel, Harold Kalant, Robert E. Popham, Wolfgang Schmidt, and Edward M. Sellers Addiction Research Foundation and University of Toronlo Toronto, Ontario, Canada PLENUM PRESS • NEW YORK AND LONDON

308 BARRY S. BROWN Vagl~rn, P., and Fossheim, L, 1980, Differential troatment of young abusers: A quasi.experimental study of a "therapeutic coanmunity" in a psychiatric hospitut, J. Drug Issues 10:505. Waal, H., 1980, Unconventional treatment models for young drag abuses in Scandinavia, J. Drug l~ues 10:441. Wesson, D. R., and Smith, D. E., 1979, Treatment of the polydrng abuser, in: Handbook on Drug Abuse, (R, L. DuPont, A. Goldstein, J. O'Donn¢ll, and B. S. Brown, eds.), pp. 151-157, Government Printing Office, Washington, D.C. Wesson, D. R., Smith, D. E., and Lemer, S. E., 1975, Streetwise and nonsUeetwise polydrug typology: Myth or reality, J. Psychedel. Drugs 7:121. Wesso~, D. R., Smith, D. E., Lerner, S. E., and Keuner, V. R., 1974, Treatment of polydrug users in San Francisco, Am. J. Drug Alcohol Abuse 1:159. Wesson, D. R., Grant, I., Carlin, A. S., Adams, K. M., and Harris, C., 1978, Neuropsychological impairment and psychopathology, in: Polydrug Abuse (D. R. Wesson, A. S. Ca.din, K. M. Adams, and G. Beschner, eds.), pp. 263-272, Academic Press, New Wexler, H. K., and De Leon, G., 1977, The therapcntic community: Multivariate prediction of retention, Am. J. Drug Alcohol Abuse 4:145. Wikler, A., 1968, Diagnosis and treatment of drug deport@race of the barbiturate type, Am. J. Psychiatry 125:758. Winer, L. R., Lorio, .I.P., and Scraffo~, I., 1974, Effects of treatment on drug abuser and family, Special Action Office for Drug Abuse Preventiou Report 4 RGO03 (1974). Winn, L, 1982, Kukulu Kumuhana--Final Evaluation Repot, NIDA Grant Report HSI DA 02056 (March, 1982). Wunderlich, R. A., Lozes, L, and Lewis, J., 1974, Recidivisru rates of group therapy participants and other adole.set~ts processed by a juvenile court, Psychother. Theory Res. Pracffce 11:243. gZOgg9890g 11 Less-Hazardous Tobacco Use as a Treatment for the "Smoking and Health" Problem LYNN T. KOZLOWSKI Tobacco is a dirty weed. I like tt. It satisfies no normal need. I like it. It makes you thin, it makes you lean, It takes the hair right off your bean. It's the worst darn stuff I've ever seen. I like it. G. L. Hre.Ml~ff~OER, in Penn State Froth, 1915 1. INTRODUCTION The health care industry cares about tobacco use mainly because it causes death and disability in users and perhaps in their associates (e.g., U.S.D.H.E.W., 1979, U.S.D.H.H.S., 1982). The war against tobacco use is at root a war of messages and recommendations about conduct. Persuasion is important in this particular war on drugs because the product in question is neither illegal nor difficult to obtain. If tobacco kills or injures, then, assuming no redeeming values, the obvious message is to stop or not start using tobacco. Unfortunately, tobacco appears to have some redeeming value, if only to the dependent user who suffers without it (e.g., Schachter et al., 1977; Silverstein, 1982). Whatever the reasons, history shows that tobacco, once introduced to a culture, is never eliminated, even in LYNN T. KOZLOWSKI • Addiction Research Foundation, Toronto, Ontario, Canada. The views expressed in this publication are those of the author and do no/necessarily reflect those of the Addiction Research Foundation. 309

310 LYNN T. KOZLOWSKI LESS-HAZARDOUS TOBACCO USE highly coercive societies (Brooks, 1952). However, change or evolution has taken place in the types of tobacco that are most popular (Kozlowski, 1982a). Once one is forced to assume that tobacco has redeeming values (benefits), then one needs to open negotiations with the enemy to determine if some deal can be struck with those who do not wish or who are unable to give up tobacco use entirely. The goal of the negotiation is to make the best of a health risk by minimizing it, knowing that, for some, it can not be eliminated practically. Partial victory is substituted, where possible, for total defeat. The less-hazardous-tobacco-use message is directed toward those who are unwilling or unable to stop using tobacco completely. Ideally, it acts when the antitobacco message (stop or don't start) fails, and it complements the antitobacco message; in practice, these two messages are competitive and troubled allies. At present, both the less-hazardous and the antitobacco messages indicate goals that we are struggling to find out how to attain. We are still trying to discover how best to prevent, stop, or modify tobacco use. [Incidentally, the protobacco response to the above messages is that we really don't know yet if tobacco is dangerous, so continue to use tobacco if you care to (Friedman, 1975).] This chapter will argue that the use of less-hazardous tobacco, if prohibi- tionistic impulses can be put aside, may have an important role in the treatment of the smoking and health problem. Just as research efforts arc needed to try to improve prevention and cessation techniques, they are needed to try to improve the techniques of less-hazardous tobacco use. (For a review of issues related to the application of less-hazardous-tobacco-use treatments, see Kozlowski, 1984.) The phrase "less-hazardous tobacco use" is meant to be inclusive. Cigarettes, for example, are the most hazardous tobacco products overall; yet even cigarettes can become a less-hazardous use of tobacco, if only a little of a few cigarettes is smoked each day. On the other hand, some less-hazardous tobacco products are less-hazardous in certain respects no matter how they are used: chewing tobacco, for example, carries no risks of fire and essentially no risk of lung disease. For some workers concerned with smoking and health, the mission of this chapter is outrageous. For these individuals (and institutions), no tobacco product can be part of the treatment of the smoking and health problem; complete pre- vention and absolute cessation of all tobacco use are the only acceptable gods. The exclusive goal of exterminating all smoking and tobacco use is so prominent a feature of the contemporary discussion of "Smoking or Health" that it will be necessary to (1) try to account for the predominance of this goal and (2) confront the possible pitfalls of pursuing only this means of reducing the health conse- quences of smoking. To try to avoid some needless arguements, I will define how I am using some key terms, before entering into the debate. Less-hazardous means reduced in risk or not as dangerous; it does not automatically mean safe or without risk. Tobacco use can refer either to (1) the type of product or (2) the nature of its use. This chapter is not mainly about the so-called "less-hazardous cigarette." Low-yield cigarettes will not be referred to 6~0~9890E as less-hazardous or safer cigarettes. Though they may indeed be less hazardous than high-yield cigarettes, this point is still controversial (e.g., Kozlowski et al., 1982b; Gerstein and Levison, 1982). Though low-yield cigarettes are low-yield when placed in the ports of smoking machines, they are not necessarily low- yield when placed in the mouths of smokers. In fact, the lowest-yield cigarettes (1 mg tar, 0. lmg nicotine) can turn into medium- or high-yield cigarettes when a smoking machine assay is adjusted tO simulate better the smoking behavior seen in a human smoker apparently bent on compensating for the reduced yields (Kozlowski et al., 1982c). Although treatment is often a medical term, it is employed here in its more fundamental meaning as a way of dealing with something. The smoking problem and the tobacco problem are more general than the smoking and health problem. Some individuals view any form of tobacco use as a serous waste of time and resources and as an activity to be discou~ged; these views would hold even if tobacco use posed no risk of disease or disability. If one believes that tobacco use, per se, is a problem to be eliminated, then less-hazardous tobacco use presents at least one problem too many. If one believes that tobacco is a problem primarily because of serious effects on health, then the reduction of the toxic consequences of tobacco use is a worthwhile goal. The smoking and health problem focuses on the damage to health caused by cigarettes. In this chapter, addiction or compulsive drug use, per se, is not considered a major health problem, unless the drug-taking behavior causes serious physical, social, psychological, or behavioral disturbance. According to the Diagnostic and Statistical Manual of the American Psychiatric Association (DSM-III) (Amer- ican Psychiatric Association, 1980), smoking becomes an official disorder 005. lx Tobacco Dependence) if serious attempts to stop or reduce tobacco use have been unsuccessful, if tobacco withdrawal occurs during tobacco abstinence, or if the tobacco use continues despite a serious physical condition (e.g., respiratory or cardiovascular disease) that the user knows is exacerbated by tobacco use. No mention is made of "impairment in social or occupational functioning." Tobacco dependence is, in fact, alone among the several substance use disorders described in DSM-III (alcohol, barbiturates, cocaine, opioids, amphetamines, phencyclidine, hallucinogens, cannabis) in that impairment in social or occu- pational functioning is not judged to be an "immediate and direct" result of the use of the substance. 2. BACKGROUND Compassion and Venom According to an Arabian story (Bain, 1896), the Prophet, Mahomet, rescued a snake from freezing by warming tbe snake against his body. The thankless snake bit him, but Mahomet sucked the venom from his wound and spat it upon

312 LYNN T. KOZLOWSKI the ground. On that spot, it was said, grew the first tobacco plant, combining the compassion of the prophet with the venom of the serpent. The quest for less- hazardous tobacco products has been directed toward reducing or eliminating the "venom" of tobacco, while at the same time keeping its "compassion." From a hedonistic perspective, then, less-hazardous tobacco use strives to maintain pleasure and minimize pain. From the perspective of the marketplace, the ideal less-hazardous product sells well, but does not kill the customers. Low-yield cigarettes lack one of the key requirements of a less-hazardous tobacco product, in that they remove the "compassion" along with the "venom." Low-yield cigarettes---as designed, not necessarily as smoked---are little more than placebo cigarettes. Ultra-low-tar cigarettes are ultra-low-smoke cigarettes. As much as 80% of the smoke in each puff of a cigarette yielding 1 mg tar can be diluting air (Kozlowski, 1981b). On the assumption that some of the phar- macologic actions of tobacco are responsible for the "compassion" (nicotine is most often thought to be tbe key ingredien0, an across-the-board reduction in drug delivery hardly qualifies as a practical strategy for producing an acceptable less-hazardous tobacco product. Tobacco use is often understood in a much too simplified way. That harm and benefit, venom and compassion, can reside in the same product is readily appreciated in many areas of applied research, yet researchers in the tobacco area have tended either to identify tobacco as a killer or to deny that claim. Though many individuals die from both the direct and indirect use of automobiles, I can not recall heating the argument that, therefore, all automobile use should be prevented or stopped. Because of the widely appreciated benefits of the automobile, the less-hazardous automobile movement has been more prominent than the antiautomobile movement. Jumping to Exclusions In their classic book on logic, Cohen and Nagel (1962) note: "One of the most fruitful sources of intellectual confusion is the too facile assumption that any two propositions which are not equivalent are mutually exclusive" (p. 68). At least one of the ramifications of this confusion can be seen in the ready employment of false dichotomous questions. Such questions make a practice of opposing issues that are neither exhaustive nor mutually exclusive (Fischer, 1970). Notice that the Royal College of Physicians in the United Kingdom entitled their recent monograph on the health consequences of tobacco use "Smoking or Health" (Royal College of Physicians, 1977). To return to the analogy with the automobile, it is as if a book were entitled "Driving or Health." In neither instance is the dichotomy justified. Not all drivers suffer ill-health as a conse- quence of automobile use; some do. Not all smokers suffer ill-health as a con- sequence of tobacco use; some do. Neither are all nonsmokers and nondrivers certain to be healthy. O80899890g LESS-HAZARDOUS TOBACCO USE 313 False dichotomies deny the crucial middle ground and emphasize the ex- tremes. They polarize a question. They add to the memorability of the question. They might provide an image around which to rally contributions and interest in a problem. Yet it is perilous for scientists to treat them as any more than slogans or entertainments. To use the false dichotomy to guide research on the problem is, in fact, to base one's exploration on an unsupportable premise. Importance of Beliefs Beliefs and values are the first principles from which the creation and eradication of social problems flow (Lindblom and Cohen, 1979). Outside of the sometimes idealized world of public health education, complicated beliefs and circumstances contribute to the valuation of tobacco. Those who thought that announcing that "smoke kills" would lead to an exodus from the bondage of tobacco use might also have predicted that the high risk of earthquakes should have emptied California. If a patient in your care or a loved one dies or suffers from a tobacco-related disease, the costs of the activity may overwhelm the benefits. If you support your family through the sale of tobacco (or if you have gone to college because of a scholarship from a tobacco company), the benefits of tobacco may be salient. If you are a smoker who feels some pleasure in smoking, then the threat of a future death from smoking might be countered persuasively with the conviction that one must, after all, die of something and that, despite the most pampered life in the world, an accidental death from any number of causes could lurk around the next corner. Cigarette smoking is argued to be the "largest preventable cause of death in America" (U.S.D.H.E.W., 1979). The term "preventable" is a problem. The wish to prevent should not be confused with the ability to prevent. In later sections of this chapter, it will be argued that there are limits to the preventability of tobacco use. The limits on the preventability of tobacco use become some of the strongest arguments for developing less-hazardous modes of tobacco use. Beliefs and values influence what one chooses to be preventable. Many activities are preventable, given enough effort to prevent; however, drug use has not shown itself to be an area of easy prevention, despite large investments in wars on this or that drug. It is doubtful whether drug use is, in practice, preventable in a free society. (Prevention and deterrence are quite different concepts.) I would revise the quotation that opens this paragraph by stating that cigarette smoking is the largest cause of death that authorities are trying to prevent. Scientific Haggling Scientific dispute does not take place at a level above the usual mire of human argument. The mantle of science is worn by people whose conflicts with a colleague share much with arguments with a spouse. Such maneuvers as

314 LYNN T. KOZLOWSKI intimidation, throat, insult, belittling, evasion, and flattery, to give a partial list, are as readily found in rational argument as they are in everyday argument (Lakoff and Johnson, 1980). Lakoff and Johnson assert that both rational argument and haggling are grounded in the metaphor "argument is war." Anyone familiar with the manufacture of the scientific bullets used in the war against (and for) tobacco use should he aware that all is fair in love and rational argument, especially when an emotionally and economically charged issue is involved. Some of the most striking reactions to the topic of less-hazardous tobacco use have not appeared in print, but have occurred behind the scenes at scientific and medical meetings. In 1980, for example, a scientific meeting was convened by a major voluntary agency in the United States, in part to help set research priorities on smoking and health. The chairman of the smoking group had pre- pared a position paper that was to focus discussion on a list of research topics. In the opening plenary session, the distinguished chairman was asked, "Why isn't the issue of doing research on less-hazardous tobacco use on the agenda?" With no hesitation, he responded, "Better men than I have been ruined for proposing such a thing!" Be assured that although this comment influenced the proceedings, it did not appear in them, and neither did less-hazardous tobacco use appear in the list of topics in need of research. Fortunately, this section will not have to rely on undocumented anecdotes to establish the needed background. In 1978, Science (Marx, 1978) reported on an event that was distressing the highest "smoking or health" officials in Wash- ington, D.C. The Secretary of Health, Education, and Welfare (HEW) had been mounting a vigorous, high-profile campaign against cigarette smoking. Plans were being made for the 1979 Report on Smoking and Health of the Surgeon General. This report was to be released on the 15th anniversary of the landmark 1964 Surgeon General's Report. The 1979 Report was to be roughly three times the size of the 1964 Report. The promotion of healthier "lifestyles" was fast becoming a popular activity: everywhere running shoes were filled with jogging feet. In this atmosphere, a government scientist, Dr. Gio B. Gori (no less than the Deputy Director, Division of Cancer Cause and Prevention, National Cancer Institute) published a paper (with Cornelius Lynch) indicating that low-yield cigarettes, especially modern low-tar brands, were less hazardous than high- yield cigarettes. The paper encouraged smokers to wean themselves progressively to less-hazardous cigarettes as "an alternative to smoking cessation that is perhaps more effective than the self-denial approaches of current anti-smoking messages" (God and Lynch, 1978). Although Gori and Lynch were careful to avoid calling low-yield cigarettes "safe," the publicity surrounding the publication in the pres- tigious Journal of the American Medical Association announced that "tolerable" cigarettes were at last available. As described by Marx (1978), "... the sug- gestion by a government scientist that smoking might be 'tolerable' was not 1808~9~90~ LESS-HAZARDOUS TOBACCO USE 315 well received by health officials who were afraid it would undermine their anti- smoking efforts." [In an equally notable example of understatement, God wrote in a summary essay: "Public policy in smoking and health has been dominated for years by idealistic approaches with moderate sympathy for less-hazardous cigarettes" (Gori, 1980).] Those who were upset about the Gori paper and its impact included: the Secretary of HEW, the Surgeon General, and the Directors of the National Cancer Institute (NCI) and the National Heart, Lung, and Blood Institute. Dr. Gori is reported to have told the press that the Secretary of HEW was trying to have the NCI fire or at least discipline him. At the time of the Marx report, Dr. God had been removed recently from command of the NCI Smoking and Health program. Dr. Gori no longer works for the NCI. The Gori and Lynch (1978) and the earlier Gori (1976) papers did offer encouragement that those smokers who would not stop smoking completely could benefit from a switch to lower-yield cigarettes. Although this may be true under some circumstances, the Gori research has been the object of a great deal of scientific criticism (e.g., Gart and Schneiderman, 1979; Warner, 1979), and I would be toward the front of the line in criticizing the Gori work. One of my key objections concerns the uncritical acceptance of the lower yield ratings of recent cigarettes. Such acceptance places unwarranted confidence in the adequacy of the simulation of human smoking behavior by standard smoking machines (Kozlowski et al., 1980, 1982c; Kozlowski, 1981a, c). Why were officials afraid that a "less-hazardous cigarette" message would undermine antismoking efforts? The official antismoking efforts were directed at smoking prevention and cessation. In other words, the only message they wished to present was, "If you don't smoke, don't start; if you do smoke, stop." The addition of a further clause, "If you must smoke, at least smoke a lower- yield cigarette," was intolerable. Why should this additional message cause so much trouble? Does such a message actually spoil an antismoking campaign? 3. ARGUMENTS AND EVIDENCE, NOT FACTS AND PROOF When one deals with social problem-solving, despite the fondest wishes of practitioners and politicians, truly objective facts are rarely found. And without such facts, no incontrovertible proofs will be forthcoming. At most, one can argue and give evidence in support of the arguments (Lindblom and Cohen, 1979). So, when policymakers ask what should be dorte about the smoking and health problem, they should expect arguments and evidence rather than facts and proof. Though the policymakers and their advisors may wish to act as if revealed facts can lead to a course of action, the process depends unavoidably on .arguments.

LYNN T. KOZLOWSKI 4. ARGUMENTS AGAINST ADVOCATING LESS-HAZARDOUS TOBACCO USE I will try to present arguments against less-hazardous tobacco use that contain more meat than straw. (My position is not as partisan as might be supposed: a diverse clientrle needs to be served, and, for some, antismoking messages are probably the most useful prescription.) I will, however, feel no obligation in this section to take on the pore role of devil's advocate; some arguments for advocating less-hazardous tobacco will be touched upon while presenting the arguments against less-hazardous tobacco use. Damage to Cessation and Prevention Efforts It is widely believed that to advocate less-hazardous tobacco use is to undermine antismoking efforts. The rationales behind this belief are no doubt complicated and several. The most prominent concerns the information-pro- cessing abilities and motivation of smokers. Although support could be cited from psychological research on human information-processing abilities (e.g., Nisbett and Ross, 1980), the prevailing belief probably rests more squarely on common truisms such as People beheve only what they want to beheve, People want to have their cake and eat it too," or "People will want to take the easy way out." The smoker, it is thought, will gather from the less-hazardous tobacco message that it is acceptable to continue to use tobacco and will tend to ignore the advice that less-hazardous use should be employed only by those who cannot or will not give up tobacco use entirely. A related argument is that complicated messages will not be as persuasive or as memorable as simple messages: the less-hazardous use message, then, complicates the overall message to the det- riment of the antismoking message. These first two arguments concern problems with the reception of the smok- ing and health message. Another line of argument holds that in a world of limited resources one cannot do all that one might like to do to reduce the smoking and health problem. In terms of priority rankings, prevention and cessation activities are seen then as more important than reduced-risk activities. Will Recruitment to Tobacco Use be Encouraged by the Availability of Less-Hazardous Tobacco Products? No one knows the extent to which con- cern about the health consequences of tobacco use acts to deter those who are otherwise tempted to take up tobacco. One line of research does indicate that women, in particular, find it easier to take up smoking, given the modern, "milder" low-yield cigarettes (Silverstein et al., 1980, 1982). If advocacy of less-hazardous tobacco use adds enough recruits to the ranks of tobacco users, then the reduced risks to the individual user could be outweighed by the greater g~OS~9~90g TOBACCO USE 317 number of individuals at risk (see the section on the Prevention Paradox below). Trends in recruitment to tobacco use should be monitored. Will Tobacco Users Use Less-Hazardous Products Instead of Quit- ting? No one knows how many smokers would have given up tobacco use entirely if they had not known of the option of less hazardous use. Some smokers might switch to low-yield cigarettes to allay the pesterings of associates about the health consequences of smoking. However, it is doubtful that these individ- uals would be willing to give up tobacco entirely, unless greater social pressure were put on them. The group of smokers to be most concerned about is those who would have been able to abstain if they had not been offered the promise of reduced-risk tobacco use. I know of no estimates of how many individuals have been lost to smoking cessation or prevention because of the availability of presumably safer ways to use tobacco. A high priority should be given to empirical research that would estimate the size of the problem. Also, a high priority should be given to de- termining how to present the risks of tobacco use to individuals in ways that will have the greatest impact on health care decisions and health care behavior (cf. Slovic et al., 1977). Even if many individuals are lost to tobacco abstinence because of the less-hazardous tobacco use message, it does not follow that, therefore, the costs of the treatment outweigh the benefits. Being Faithful to One's Job Description It can be as important to know what is not part of one's job as it is to know what is part of one's job. Although tobacco once was a product that was crucial to the practice of the healing arts (Stewart, 1967), modem physicians believe that it is not within their job description to, in effect, advocate the use of any tobacco product: if less-hazardous tobacco use is to develop, it is thought to be up to the tobacco companies to be the advocates and developers. For the medical profession in general, tobacco has become an evil substance that is totally unfit for human consumption, unlike certain other potentially hazardous products (e.g., eggs, whole milk, salt, and sugar) about which the medical profession is willing to make recommendations concerning less hazardous use. For some reason, the tobacco industry has been especially easy to identify as the enemy, perhaps because of deep-set Calvinist convictions about the sin of drug use. Though a physician might be comfortable advising a low-salt or low-sugar diet (knowing that a no-added-salt or no-added-sugar diet, though probably less-hazardous, would receive little compliance), this same physician could not recommend the use of a less-dangerous tobacco product (knowing that abstinence may also result in little compliance). I juxtapose the tobacco and the food industries to illustrate an ironic inconsistency in the practice of public health and medicine: all these products may be optional but some are much more optional than others.

3 t 8 LYNN T. KOZLOWSKI • The Less-Hazardous Message is Already Well-Known The antismoking messages of prevention and cessation may inadvertently and unavoidably support the cause of less-hazardous tobacco use. Smokers on their own might tend to adop.t less hazardous uses of tobacco in response to clear messages that they should stop using tobacco. In other words, the message of less-hazardous tobacco use might occur by default as the antismoking message is spread. Similady, it can he argued that the less-hazardous tobacco use message is already well-known, because of the publicity surrounding the tar and nicotine yields of cigarettes. Standard tar and nicotine ratings have been supplied by governments, in part to encourage the use of lower tar and nicotine cigarettes by those who do not stop smoking (Friedman, 1975). The modem "tar derby" emphasizes that lower yield is better. Even if one considers the low-yield cigarette as the paragon of less hazardous tobacco products, the less-hazardous-tobacco- use message has been spread mainly in a superficial and dangerous way. (See Kozlowski, 1984, for a discussion of applications of less-hazardous tobacco therapies.) It is not enough simply to point a tobacco user to different products: advice and assistance should be given to help the user reduce exposure to toxic tobacco products. Less-Hazardous Tobacco Use as Boondoggle One argument against advocating less-hazardous tobacco use is that the promise of reduced risks is more apparent than real. In other words, the rec- ommendation to use less-hazardous products should not be made because there are no truly less-hazardous tobacco products. Low-Yield Cigarettes. Low-yield cigarettes are at the same time the most popular and the most questionable of the presumably less-hazardous tobacco products. Some evidence indicates significant, but small, reductions in risk to be gained from a switch to low-yield cigarettes (e.g., Lee and Garfinkel, 1981; Vutuc and Kunze, 1982); other evidence indicates no reductions in risk (e.g., Castelli et al., 1981; Kaufman et al., 1983; Robinson et al., 1982). The present evidence is far from conclusive (Russell et al., 1980b; Kozlowski et al., 1982b; U.S.D.H.H.S., 1981). It is possible that long-term use of low-yield cigarettes is required before a beneficial reduction in smoke exposure is seen and that those who are forced to switch brands are less likely to compensate for reduced yields than those who switch on their own (Russell et al., 1982). All cigarettes, along with other smoking tobaccos, make it difficult for users to know exactly what they are ingesting from these products (Kozlowski, 1984). It is not possible simply to read a product label and thereby know what one is getting from a cigarette or pipe. Actual smoke intake depends more on the details of a smoker's behavior (number of puffs, volume of puffs, depth of inhalation): 880t~9890~, LESS-HAZARDOUS TOBACCO USE 319 more on the smoker than on the product (Kozlowski, 1983). Cigarette smoke is more often inhaled than any other kind of tobacco smoke (U.S.D.H.E.W., 1979), and therefore cigarette smoke presents special risks to the lungs. There is also the question of risks to those who associate with or are exposed to smokers when they are smoking (U.S.D.H.H.S., 1982). AIso, the use of smoking tobaccos carries the risk of fires, and resultant death and suffering for both active and passive smokers (Bed and Halpin, 1978). Given (1) the controversy over the epidemiologic effects (in both active and passive users of tobacco smoke), (2) the difficulty in monitoring dosage, (3) the problem of inhalation, and (4) the issue of fire hazards, it is not easy to he sanguine about low-yield cigarettes as a treatment for the health consequences of smoking. Especially in light of other, more promising options for less-haz- ardous tobacco use, I am inclined to he very.pessimistic about the value of low- yield cigarettes. Even if low-yield cigarettes are poor less-hazardous tobacco products, it does not follow that other less hazardous tobacco treatments are therefore in- effective. The failure of one pharmaceutical is no grounds for closing down the pharmacopoeia. Pipes and Cigars. The available evidence suggests that pipes and cigars are less hazardous than cigarettes (Doll and Peto, 1976; U.S.D.H.E.W., 1979). People who start (and stay) with pipes or cigars tend not to inhale and hence reduce the exposure of their lungs to toxic smoke products. It is unclear whether smokers who turn from cigarettes to pipes and cigars continue the habit of inhaling. Some researchers have found inhalation among so-called secondary pipe or cigar smokers (Castledon and Cole, 1973; Turner et al., 1977, 1981); others have not found evidence of substantial inhalation by secondary pipe or cigar smokers (McCusker et al., 1982; Wald et al., 1981). Even if secondary pipe and cigar smokers do inhale, the epidemiologic evidence indicates that, while these smokers are at greater risk of dying than are primary pipe or cigar smokers, they are at a lower risk than those who continue to smoke cigarettes (Doll and Peto, 1976). If people were encouraged from the start to smoke pipes or cigars rather than cigarettes, this problem of inhalation among pipe or cigar smokers might not arise. Smokeless Tobaccos. There is really no dispute about whether smokeless tobaccos present fewer hazards to the user than do smoking tobaccos (Harrison, 1964; Russell et al., 1980a). Smokeless tobaccos expose the lungs to essentially no tobacco toxins. No carbon monoxide and no tar is produced. The oral cancers associated with oral smokeless tobaccos are substantially less lethal and are more easily diagnosed than lung cancers (U.S.D.H.E.W., 1979). In addition, smoke- less tobaccos pose no problems of second-hand smoke and no risks of fire. Clearly risks are reduced, but the residual risks are substantial enough to cause some authorities to refuse to advocate their use (Christen et al., 1979). A subclass of the boondoggle objection then is that reductions in risk are too small to warrant

320 LYNN T. KOZLOWSKI " support. (For a discussion of nicotine-containing chewing gum as possibly the least hazardous of the smokeless "tobaccos," see Kozlowski et al., 1982a, and Kozlowski, 1984.) 5. ARGUMENTS FOR ADVOCATING LESS-HAZARDOUS TOBACCO USE Job Descriptions Reconsidered Health Care Providers. Health care providers are sometimes preoccupied with their roles as opinion leaders and "moral forces" within their communities. Their function as authority figures can even obscure the more central parts of their job descriptions. Is it not best to try to do what one can to reduce death and disability in those who continue to use tobacco? Is not the reduction of death and disability a fundamental part of the job description of a health care provider? Tobacco Product Providers. For the health care provider to consider the support of less-hazardous tobacco use as a job for the tobacco industry may be naive. It could be risky to lee, re the development of less-hazardous products to an industry whose life's blood is the cigarette. Although the tobacco industry is best-suited technically to developing less-hazardous tobacco products, it should not be forgotten that it has a business to protect. Also, the tobacco industry has steadfastly denied that tobacco use causes any medical problems; this hardly puts them in a position to invest much in the development of products that reduce hazards that they assert are not there to begin with. Community Health and the Prevention Paradox Physicians who are not specially trained in public health and preventive medicine are apt to make a category mistake when considering the issue of less- hazardous tobacco use. This category mistake [i.e., allocating concepts to a category to which they do not belong (Ryle, 1949)] consists of mistaking the public health issue for a personal health issue written large. As a matter of personal health care, for a physician to recommend the less-hazardous use of tobacco can be seen (and felt) as a failure to use the positive powers of one's practice. It does not follow that a small benefit to the health of the individual will constitute a small benefit to the health of the community. Dr. William Castelli (1981) made the mistake of removing the less-hazardous tobacco use argument from the public health domain and placing it in the physician's office. Taking the most generous estimate from the report of Lee and Garfinkel (1981), Castelli noted that a pack-a-day smoker who switched from unfiltered to filtered cigarettes reduced his or her risk of lung cancer from 20 times that of a nonsmoker to 15 ~g0999890g iS-HAZARDOUS TOBACCO USE 321 times that of a nonsmoker. Castelli wrote: "I do not personally get much sat- isfaction encouraging someone to pursue a habit which increases the risk of lung cancer 15 times" (p. 642). This does describe the situation from the physician's perspective; however, from the perspective of one interested in community health, the satisfactions may be obvious: if 2000 pack-a-day smokers had been dying each year from lung cancer, now 1500 smokers would be dying. Five hundred people would still be alive; 25% fewer smokers would be dying from lung cancer. Rose (1981) describes the "prevention paradox .... a measure that brings large benefits to the community offers little to each participating individual." A treatment that is worthwhile and practical for the community may have trivial influence on the individual. Conversely, the treatment that may have the most benefit for the individual may be impractical and hence of little use for the community. Rose uses the treatment of hypertension as an example. Extremely high blood pressure can be controlled with drugs, but relatively few individuals have extremely high blood pressure. If the average diastolic blood pressure of the community were reduced by just 7-8 mm Hg (say, by altering the die0, then the number of disorders due to blood pressure would decline as much as if all those with pressures of 105 mm Hg or more were treated in a 100% effective way. The less dramatic therapy reaps appreciable net benefit because of the large number of people involved with the treatment. The continuing discussion of the low-yield cigarette has often ignored the relevance of this paradox for tobacco use (e.g., Marks, 1982). Basically, the principle behind the paradox is that small effects on a large enough scale can produce more net benefits than can effects of heroic proportions on a small scale. This principle is also manifest in Russell's (Russell et al., 1979) advice on the benefits of physicians' advice on smoking cessation. Each physician will have relatively little success in persuading patients to give up cigarettes, but given the number of physicians available to spread the word, the net effects could be many times larger than the effect of more expensive alternative therapies. Of course, one of the key assumptions involved with employing the pre- vention paradox as an argument for less-hazardous tobacco use is that the number of people enjoying the small benefit must be large enough to add up to a sub- stantial net benefit. One might think that if the prevention and cessation efforts became highly successful, there would be few tobacco users left to enjoy the small benefits of less-hazardous tobacco use; however, it must be remembered that for those individuals who continue to use tobacco, cessation is, by definition, not an alternative treatment to less-hazardous use. The Limits of Prevention and Cessation None of the arguments for the advocacy of less-hazardous tobacco use should be used to argue against the deployment of prevention and cessation programs. The tess-hazardous tobacco use message has no war with the anti-

322 LYNN T, KOZLOWSKI smoking message; in fact, the relationship between the two is symbiotic. As noted above, the less-hazardous use message is best viewed as an effort to deal with the failures of other efforts. Prevention programs in the schools have received great attention in recent years (e.g., Evans et al., 1981). Though these programs have shown some success in reducing recruitment to smoking, at least during the school years, no one would argue that any program has discovered a certain technique for drastically reducing the number of smokers in high school, say, below the level of 10% of the students. Similarly, formal smoking cessation treatment programs find in general that 80% of their clients will relapse to smoking within I year (Raw, 1978). One estimate of how well smokers succeed at stopping smoking after repeated attempts on their own indicates that about 40% will fail to abstain in the long run (Schachter, 1981). One of the best studies of the overall impact of the antismoking campaign comes from a cohort analysis of smokers in the United States population (Warner and Murt, 1982). Based on the percentages of smokers in different age groups before the antismoking campaign really got started (before the 1964 Report of the U.S. Surgeon General), Warner and Mutt estimated how many smokers would have been expected in these same age groups had the antismoking cam- paign not taken place. In 1964, 67% of the 21- to 24-year-old men were smokers; in 1975, only 41% of the 21- to 24-year-olds were smokers. In 1964, 42% of the 21- to 24-year-old women were smokers; in 1975, 34% were smokers. They estimate that, if it were not for the antismoking campaign, 61% of men 18-27 years old would have been smokers in 1978; only 39% of this group were smokers in 1978, a difference of 22 percentage points. For women of the same age, they estimate that 49% would have been smoking; 37% actually were smokers in 1978, a difference of 12 percentage points. Though these figures indicate sub- stantial success for antismoking efforts, they also clearly show that a potential market exists for less-hazardous tobacco. The Promise of Diminishing Returns. Tobacco users differ in how de- pendent they are on tobacco. A number of studies have shown that cessation interventions are more successful with less-dependent tobacco users (e.g., Fa- gerstrom, 1982; Kozlowski et al., 198 I). One of the clearest implications of this finding is that the pool of continuing smokers is becoming more likely to contain more-dependent tobacco users. [The population of smokers is made up increas- ingly of fewer and heavier smokers, (U.S.D.H.H.S., 1981)]. In other words, the antismoking campaign has probably tended to remove those who are most easily removed from the ranks of smokers. Those who remain are likely to be a hard core of recalcitrant and perhaps "reactant" smokers. Reactance is a tech- nical term that refers to an individual's assertion of freedom of action when faced with attempts to restrict that freedom (Brehm, 1966). Less-hazardous tobacco use may be one of the few treatments available for these smokers. ~809E9B90E LI:SS-HAZARDOUS TOBACCO USE 323 A Question of Class. Recently, there has been a growing concern about the social inequalities of health care delivery systems (e.g., Morris, 1980). If one looks carefully at smoking statistics, it is apparent that, in general, those of lower socioeconomic status are more likely to use tobacco than are those of higher socioeconomic status. [An exception is that higher-class women are smok- ing more than lower-class women (U.S.D.H.E.W., 1979)]. If one looks at some especially disadvantaged groups, one finds, for example, that in Canada only 23% of teenagers (ages 15-19) who are still in school are daily smokers, whereas 48% of those who are no longer attending school (essentially high school drop- outs) are daily smokers (Health and Welfare Canada, 1981). The same report finds that those with low levels of education, and those who are unemployed or in low-status jobs, are more likely to be current daily smokers. Moody (1980) finds that those from lower socioeconomic groups also take more puffs per cigarette and are exposed to more daily tar than are those from higher socio- economic groups. Has the antismoking campaign been less successful in reaching the lower classes? Has the antismoking campaign been less successful with those of low socioeconomic status that it has reached? Are those of low socioeconomic status more likely to be dependent on tobacco? Certainly it is fair to say that a school dropout may miss out on many of the antismoking efforts in the schools. When one has lost a job, is one also inclined to hold on to the compassion to be found in tobacco? Whatever the reasons, socioeconomic lines have indicated systematic limits to the power of current antismoking efforts. 6. THE LIMITS OF LESS-HAZARDOUS TOBACCO USE Prevention and cessation efforts are not alone in having limited power and success. Much of the speculation about less-hazardous tobacco use as a treatment has not, in fact, received empirical test. Some of these good ideas may not work in practice. The epidemiology of tobacco-related diseases can only serve as a guide to possible treatments. Epidemiologic samples are self-selected as tobacco users: epidemiologic studies generally show no more than correlations between tobacco use and disease. No one knows, for example, how the health consequences of smoking might change in a group of cigarette smokers who were randomly assigned to take up pipe smoking. No one knows how many participants in such a study would be able to comply with their instructions. We do have evidence that secondary pipe and cigar smokers do have less risk of disease than those who continue to smoke cigarettes (Doll and Peto, 1976); but we do not know if the change to pipes or cigars is the cause of the reduced risk. Perhaps those cigarette smokers who do change to pipes or cigars are very different (e.g.,

324 LYNN T. KOZLOWSKI constitutionally) than those who continue as cigarette smokers (Seltzer, 1972). Despite these reservations, the epidemiologic literature does form the basis for predictions about less-hazardous tobacco use. Technical versus Behavioral Interventions Technical interventions depend upon changes in the tobacco product. Be- havioral interventions depend upon changes in conduct. If, for example, a less- toxic tobacco tar could be developed, then, one might find reductions in lung cancer incidence even if there were no changes in tar intake. It has been argued that modern tars are less toxic (milligram for milligram) than the tars of the 1950s and that this reduced toxicity might account for the reduced incidence (e.g., Gori, 1976). If modern tars are less toxic or can be made less toxic, one would have a less-hazardous tobacco use treatment for the smoking and health problem that would (assuming that the compassion remained) pose essentially no problems of patient compliance. The ideal technical intervention involves the modification of a product that the tobacco user is already using. Being able to reduce the intrinsic risks of a product that tobacco users will not use provides little treatment for the tobacco and health problem. Some behavioral interventions might be directed to per- suading the tobacco user to use a less-hazardous product. Other behavioral interventions will be directed to the less-hazardous use of the product currently being used. Each of these kinds of behavioral intervention is truly easier described than done. For a discussion of some of the challenges involved with behavioral interventions, see Kozlowski (1984). Diet, Drugs, Occupation, and the Risks of Tobacco There may be adjunctive ways to engage in less-hazardous tobacco use. The epidemiologic literature suggests that it would be advisable for smokers to change other behaviors to reduce the health consequences of tobacco use. This literature is, for the most part, suggestive rather than conclusive. Those who work with asbestos and smoke cigarettes are at especially high risk of lung disease (see U.S.D.H.H.S., 1982, for a review). Similarly, cigarette smoking and birth control pills may act synergistically to increase the risk of cardiovascular disease in women (see U.S.D.H.H.S., 1980). Tobacco and al- cohol appear to act synergistically to increase the risk of cancers of the mouth, pharynx, larynx, and esophagus (see U.S.D.H.H.S., 1982, for review). It is possible that a continuing tobacco user could reduce the health consequences of tobacco use by being careful to avoid alcohol, asbestos, and birth control pills: in terms of practicality, it should not be prejudged which of these activities is optional for given individuals. As a positive measure to reduce the risks of cancer in the tobacco user, there is growing evidence that a diet rich in pro-vitamin A 9g0939990g LESS-HAZARDOUS TOBACCO USE 325 has a protective effect against lung cancer (Doll and Peto, 1981; Shekelle et al., 1981). Less-hazardous tobacco use might, then, be established by modifying (1) the tobacco use, (2) a cofactor for risk, or (3) both. 7. LEGITIMIZING THE TOPIC AND THE NEED FOR RESEARCH The arguments for and against advocating less-hazardous tobacco use have certainly not been exhaustive. This chapter has tried to legitimize the study of less-hazardous tobacco use as a beneficial treatment for the smoking and health problem. Despite the impression that some antismoking readers might have, I am uneasy about a blanket endorsement of the less-hazardous-tobacco-use ther- apy. Data may indeed emerge in the future that will show the less-hazardous movement to have been ill-advised. Current research should, however, not fear to show both the advantages and disadvantages of all aspects of the war against tobacco-related maladies. As is the case in many areas of applied research, in this area it is not possible to wait until all of the data are in to decide what should be done about less- hazardous tobacco use. As a self-administered therapy, "less-hazardous tobacco use" exists already. Many tobacco users will not be persuaded to give up their use of tobacco, despite the best efforts of the antismoking campaign. If there are ways to reduce obvious errors in this self-administered therapy, then the consumers of these therapies should know about them (Kozlowski, 1982b). Without research on the would-be forms of less-hazardous tobacco use, we are not able to establish their actual, rather than supposed, net worth. ACKNOWLEDGMENTS The author thanks R. Frecker, C. P. Herman, S. Herling, L. Jelinek, M. Pope, and K. Wagner for their assistance. REFERENCES American Psychiatric Association, 1980, Diagnostic and Statistical Manual of Mental Disorders, 3rd. Ed. American Psychiatric Association, Washington, D.C. Bain, J., 1896, Tobacco in Song and Story, Caldwell, New York. Bed, W. G., and Halpin, B. M., 1978, Human fatalities from unwanted fires, The Johns Hopkins University--Applied Physics Laboratory, December, p. 8. Brehm, J. W., 1966, A Theory of Psychological Reactance, Academic Press, New York. Brooks, J. C., 1952, The Mighty Leaf: Tobacco through the Centuries, Little Brown, Boston. Castelli, W. P., 1981, Filter cigaretles and heart disease, Lancet 2:642. Castellt, W. P., Dawber, T. R., Feinleib. M., Garrison, R. J., McNamara, P. M., and Kannel, W. B., 198 I, The filter cigarette and coronary heart disease: The Framingham study, Lahcet 2:109.