It is over 30 years since the appearance of the first English language reports indicating that passive smoking is harmful to the respiratory health of infants and children,^1,2 and almost a quarter of a century has elapsed since publication of the first two papers pointing to an increased risk of lung cancer in non-smoking adults who live with smokers.^3,4 The first of these events passed without much discussion but, by the time the second occurred, the tobacco companies were ready with a strategy to oppose what they had been advised was “the most dangerous development yet to the viability of the tobacco industry that has yet occurred”.⁵ That strategy went beyond disputing the science to “playing the man”; Hirayama’s credentials as a scientist were called into question. Fully nine years later, he was spirited away from the 7th World Conference on Tobacco and Health in Perth, Western Australia, when it seemed that the tobacco companies were about to serve him with a writ. The Director of Action on Smoking and Health (Australia) was unsuccessfully hauled through the courts, and both the editor of this journal and the author of this editorial were served with legal papers designed to prevent them speaking in public about passive smoking⁶ while they were involved in preparation of the second report on the issue from Australia’s National Health and Medical Research Council.⁷ That tactic backfired, however, when we called a press conference and described the attempt to gag us.

PERNICIOUS MYTHS

Such direct confrontations are rare, and it may be that the industry’s taste for court action has been curbed by its major setbacks within the legal system of the USA. However, one does not have to be involved in the issue of passive smoking for very long to appreciate that the companies and their fellow travellers have a second, lower profile string to their bow; their persistent promulgation of pernicious myths about passive smoking is a campaign so sustained as to remind one of the “blood libel”, that Jews kill and eat the babes of Gentiles at Passover.

These myths fall into several categories. The first is that passive smoking is, at worst, a nuisance, and that the problem can be solved by a combination of courtesy and ventilation. There is abundant anecdotal evidence that from time to time a polite request that a smoke-free zone be kept that way is greeted with verbal abuse or even the threat of violence, depending on the sobriety of the smoker. Careful calculations by Repace and others show that ventilation is an impractical solution⁸—although it might prove difficult to light a cigarette in the hurricane-like conditions that would need to prevail! In any case, the equipment is expensive to install and run. An official enquiry in Western Australia revealed that the staff of some nightclubs in which smoking was permitted did not even know where to turn on their air conditioning system, let alone how to adjust it in response to particular levels of smoke in the premises.⁹ Thus, the efficacy of courtesy and ventilation is a myth.

The next myths, of impracticality and low compliance, are regularly trotted out when each extension of smoke-free policies is foreshadowed. Time and again these predictions have been proved wrong; indeed, public and employee support for smoke-free policies increase after their introduction, and, given adequate discussion and notice beforehand, compliance with smoke-free policies is usually very good.¹⁰

ECONOMIC IMPACT: A THREE HEADED DRAGON

The economic impact of smoke-free policies is a three headed dragon. One head anticipates that the introduction of a smoke-free policy will have dire economic consequences, particularly for the hospitality industry. The second speaks within days of the policy being introduced and claims that patronage has slumped. The third waits some weeks and proclaims that the bottom has fallen out of the bottom line. Enter St George, in the guise of Scollo, Lal, Hyland and Glantz whose seminal review¹¹ should see this beast in its death throes, with further supportive evidence now emerging from the Republic of Ireland.¹²

The myths mentioned thus far all concern smoke-free policies, but there is a parallel campaign waged against the science underpinning those policies. This tactic, too, draws directly on advice from the Roper Organization.⁵ Again, there are three principal assertions. The most nebulous of these concerns confounding. Like a weak undergraduate student who throws a technical term into an examination answer in the hope that it will attract some marks, the small, although systematic, differences in other aspects of lifestyle between non-smokers from households that include smokers and those from homes that are smoke-free¹³ are held somehow to explain why passive smokers get more lung cancer and heart disease than non-smokers. The exact mathematics are never fully laid bare, but the mud is regularly thrown in the hope that some of it will stick.

MISCLASSIFICATION

The next line holds, in essence, that at least some smokers are liars. More formally, that misclassification of exposure status, through some active smokers denying their habit, explains much of the apparent risk of passive smoking. Interestingly, of course, this argument implicitly concedes that active smoking is dangerous, something itself long contested by the tobacco industry. Misclassification is considered by Lam et al¹⁴ as a possible explanation for their novel finding, reported in this issue of Tobacco Control, that passively exposed active smokers have a significantly increased risk of various upper and lower respiratory tract symptoms compared with those who smoke “in isolation”. The findings are based on a cross sectional survey of policemen in Hong Kong, conducted almost 10 years ago. Lam et al¹⁴ concede that objective verification of passive smoking among active smokers is difficult but suggest that there is no reason why active smokers with respiratory symptoms should systematically over-report their passive exposure. Presentation of supportive crude odds ratios would have strengthened the case for a true effect, and classifying exposure of policemen on foot patrol according to estimated aggregate daily consumption of cigarettes by their colleagues during working hours is probably an over-simplification. Nevertheless, given the historical period and the prevailing emphasis of the effects of passive smoking on non-smokers, Lam’s results cannot be dismissed out of hand. If replicated, reducing the risk to continuing smokers will become another plank in the case for extending smoke-free policies to all workplaces.

RISKS FROM LIGHT SMOKING

The final charge is that the evident risks associated with passive smoking are implausibly large, given the amount of tobacco smoke that non-smokers inhale. An essential weakness here is that we still do not know for certain which elements of tobacco smoke are responsible for particular problems, notably cardiovascular diseases, in active smokers, and therefore do not know which biomarkers might be relevant in passive smokers.¹⁵ “Cigarette equivalents” of exposure have been proposed as a surrogate measure, with the detractors regularly claiming that passive smoking is at most equivalent to very light active smoking. Gori, for example, has published an elaborate argument that since epidemiological studies have struggled to demonstrate statistically significant excess risks associated with light active smoking, passive smoking must also carry no hazard.¹⁶ In a second paper in this issue of the journal, Bjartveit and Tverdal¹⁷ provide a direct refutation of that proposition through showing, in a large prospective study in Norway, that smokers of 1–4 cigarettes daily more than doubled their chances of dying from ischaemic heart disease and lung cancer, and significantly increased their mortality from all causes over the next 25 years. To paraphrase Glantz,¹⁸ tobacco smoke is dangerous—period.

The accepted definition of epidemiology speaks not only of what its practitioners study but also “the application of this study to control of health problems”.¹⁹ This direct link between evidence and action is central to epidemiologists’ seeming obsession with discounting chance, bias, and confounding as explanations for their results. In creating and promulgating untruths about passive smoking, the tobacco industry has put epidemiologists on their mettle, but the systematic slaying of those myths has only strengthened the case for smoke-free policies, to protect the health of smokers and non-smokers alike.