• End game
• Prevention
• Public opinion
• Public policy
• Tobacco industry

Discussion of a tobacco ‘endgame’, and the growing body of scholarly research and commentary on how to achieve it,1–21 encourage innovative next-generation ideas for ending the tobacco epidemic. Yet serious endgame discussion also creates risks. How we respond to these risks will determine what 21st century tobacco control can accomplish.

An endgame addresses tobacco as a systems issue, rather than an individual behaviour; addresses health and political implications; reframes strategic debates; advances social justice; and is fundamentally transformative in changing how tobacco use and the tobacco industry are regarded. An endgame is not merely more of the same, in that it requires an authentic public policy commitment to achieving a true endgame, as opposed to continuing to envision the public health challenge as an ongoing war of attrition. Incremental change cannot fix this public health emergency, at least not absent a vision of an endpoint when the threat will be eradicated. Thus vision and goals are in some ways more important than specific tactics. What remains astonishing is the degree to which the social construction of tobacco as normal and desirable, accomplished over the last century by a savvy industry, still blinds many to the urgency of our task and the contradictions inherent in our own messages about tobacco.

For example, it becomes more and more impossible to reconcile the ‘danger: don't do it’ message with tobacco products’ ubiquitous availability and legality. The inconsistency with which tobacco is regulated, compared with other far less deadly products, is inexplicable.

Endgame strategies are the means to reach an endpoint. That endpoint should include envisioning the future of tobacco products, tobacco users, the tobacco industry, the role of government and the role of the tobacco control movement. Assuming an endpoint in which death and disease from tobacco would be virtually eliminated, several key questions emerge.

For example, will that envisioned endpoint include no commercially sold tobacco products at all, or only tobacco products less deadly than cigarettes? Or, as some suggest, will only ‘clean’ nicotine be available commercially? Will tobacco users all have transitioned to other products? Will the tobacco industry still exist as such, or will it focus on other types of nicotine products, as the proliferation of ‘alternative’ nicotine products suggests is already occurring? Or will the industry move into other businesses? How will we replace government funds generated through tobacco taxes, when the savings from reduced rates of death and disease may be unrealised for decades? Will there still be a tobacco control movement at the endpoint?

Even if we find consensus, our ability to rationally and methodically plot a route to the endpoint is limited. Policy literature22–25 demonstrates that social change results from multiple factors, many serendipitous and unpredictable. The point is to be ready—to believe it is possible to permanently change what tobacco means in our societies, and to talk about it openly.

At least three tensions must be navigated. First, should transitions be gradual or abrupt? Abrupt change makes it more difficult for the tobacco industry to adapt and thereby reduce a policy's effectiveness. However, abrupt change can also create public anxiety and increase resistance.

Another tension is whether public health should compete or cooperate with the tobacco industry. The industry has no wish to kill people; it simply wants to make money. Massive death is simply an accepted externality of doing business. Incentivise that business differently, goes this argument, and the industry may become part of the solution. Others, notably those concerned with implementation of Article 5.3 of the WHO Framework Convention on Tobacco Control,26 which seeks to protect public health policy from tobacco industry interference, assert that the industry in many respects is the tobacco problem. A true endgame scenario must ultimately end commercial promotion of tobacco products. These might be characterised as accommodationist versus abolitionist perspectives.

Third, how harmful is nicotine addiction in the absence of smoking? Is our endpoint reached when we have eliminated use of combustible commercial tobacco, or tobacco in any form? Are incentives to move addicted tobacco users toward medicinal or other forms of ‘clean’ nicotine the means to an end, or the end in itself? Some suggest the latter. Yet there is considerable evidence that nicotine is not benign, but may play a critical role in disrupting cell regulation and cell-mediated immune function,27 and that even products considered far less harmful than smoked tobacco, such as Swedish snus, are associated with elevated death rates from cancers.28 The complexity of addiction means that this is also not merely a question of biochemistry. Should we encourage use of products such as e-cigarettes that powerfully model smoking behaviour but presumably expose users to fewer dangerous compounds?

The risks of not envisioning an endpoint are clearly far greater than the risks of attempting any of these solutions and failing. In the USA, it is hard to imagine that trying and failing at any endgame could result in nearly 400 000 deaths a year. Any endpoint will only come after many years of paying the price for current failures to respond more aggressively to this epidemic. But it is worth tackling radical ideas, because even short-term failures may contribute to achieving long-term goals: witness the first attempts in California to achieve smoking sections in restaurants, which failed—but which initiated a public conversation that eventually led to 100% smoke-free spaces.29 Thus we should not too quickly foreclose conversations about more radical endgame strategies, even if they do not at first succeed.

Different solutions will prove possible in different places, and may unfold in unique ways. In Singapore, Finland and the Australian state of Tasmania, for example, there is serious discussion of implementing a law banning cigarette sales to those born after a particular year, usually 2000 (see elsewhere in this issue).30 I believe that all endgame approaches would require continuing, explicit and aggressive industry denormalisation31 to enhance existing measures and accompany product modifications, limits on retail outlets, and an eventual phasing out of cigarettes from the market.

In the US, where regulatory regimes are more stubbornly incrementalist, an effective endgame will likely involve combining approaches across different levels of government, as I outline here in brief. Initially, I would focus on the cigarette alone, as the single most deadly tobacco product. Nationally, the US Food and Drug Administration (FDA) could require lower nicotine levels in all cigarettes, reducing their addictiveness, and ensure that any restrictions on conventional cigarettes apply to quasi-cigarette products such as little cigars. This is feasible, within the mandate of the FDA, and as discussed elsewhere in this issue, evidence suggests that it might make it easier for smokers to quit and would not lead to compensatory smoking. Further restricting the marketing of other products could help reduce dual use and youth uptake—again, measures that lie within the federal purview.

US states could restrict tobacco outlets, either through first limiting cigarette sales to state-run outlets, as some states now manage ‘hard’ alcohol, or by capping the number of retail outlets within a geographic area. Again, this is feasible and there are models on which such a system could be constructed. Display bans could reduce visual cues that elicit purchase. The goal, however, should be to move toward phasing commercial cigarettes off the market, with reasonable advance notice. No other product that killed half its normal users would be permitted on the market today; it is simply unacceptable to allow the current situation to continue.

Cities and towns could initiate the first cigarette sales bans, just as they led on secondhand smoke policies, but they would have to anticipate industry litigation. In some ways, however, even if they passed such measures and then could not afford to fight the subsequent lawsuits, the media coverage of the issue in town after town could advance the goals of tobacco control by keeping the industry in the news fighting against local public health initiatives.

There are ample arguments to explain why all this won't work. But every single effective tobacco control policy ever proposed was initially met with scepticism. When people in Richmond, California sought smoke-free air in their apartment complexes, proponents were told it couldn't be done. When flight attendants wanted smoke-free airplanes, they were told that smokers would riot and planes would crash. When advocates sought smoke-free Irish bars and pubs, their own colleagues scoffed. But the seemingly impossible happened in all these places.

It happened because someone imagined it could, and kept talking about it.