John Luik

In his November 2008 article on tobacco regulations, my colleague George Gay quoted Oxford University scholar Ross McKibbin’s comment in the London Review of Books (September 2008) about the intellectual quality of the current U.K. government. "The present cabinet," wrote McKibbin, “has become the most lightweight in living memory. Some of its members are so lightweight they shouldn’t be in the cabinet at all …”

George wrote his article before the U.K. government announced its new tobacco “strategy.” This strategy is based on the results of a summertime consultation that produced more than 100,000 responses, though a large majority of these reportedly were simply uniform postcard responses prepared by the anti-tobacco lobby.

            But the new measures proposed by the U.K. government are in many respects a brilliant illustration of both how ill-evidenced most tobacco control policies are and also how heedless they are of perverse consequences.

            The U.K. measures, to take effect in 2011, include requiring customers wishing to use cigarette vending machines to purchase tokens that will only be available to those showing proof of their age. The main measure, however, is a ban on the retail display of all tobacco products in shops. Defending the government’s move on BBC Radio 4’s Today program, Health Secretary Alan Johnson claimed that there was “overwhelming evidence” supporting such a ban as a way to prevent children from smoking, “When they [children] see a point of display and as a result of seeing it they take up smoking … it’s the key evidence as why 200,000 11-15 year-olds are smoking.” (“Shops face ban on tobacco displays,” The Guardian, Dec. 9, 2008).  

           

Lack of success

But Johnson’s comments say far more about the levels of bias and ignorance that drive not only U.K. tobacco policy, but international tobacco control efforts as well, than they do about sensible policies for reducing tobacco use.

            To begin with, there is a central problem about the U.K.’s new tobacco policies that has dogged tobacco control from its inception. That problem is the stark lack of success of most public policies designed to curb smoking. Because most of these policies have failed, the anti-tobacco movement has been forced to come up with a continually changing menu of new policies, each promising to deliver what its successors promised but did not produce.

            For instance, when activists championed tobacco advertising bans, it was asserted that this was a major—if not the major—cause of youth smoking, and that legislating bans would dramatically reduce the number of adolescent smokers. When bans failed, it was argued that it was the enticing character of the cigarette package itself that led to smoking, and that bolder, larger and more graphic warnings would reduce smoking. When these warnings failed—as the evidence from Canada shows—the anti-tobacco lobby argued for hiding the pack behind the counter, to be followed by selling cigarettes only in plain packs.

            The common thread that runs throughout these failed policies is not only that they are based on little credible evidence of addressing the root causes of smoking, but also that when their failure becomes evident, it is unacknowledged, and even more arrogantly, it is used as the basis for further and more draconian regulations.

            Consider the case of the recent public smoking ban in the U.K. The government has consistently maintained that the smoking ban, introduced in July 2007, has forced record numbers of smokers to quit. According to a July 2008 report prepared by the chief medical officer, Sir Liam Donaldson, 234,060 people quit smoking in the months prior to and after the ban. In the foreword to the report, Donaldson claims that the “significance of the smoke-free laws cannot be underestimated.”

            However, as the recent Health Survey for England shows, these claims about the effectiveness of the smoking ban as a tobacco control policy are false. Indeed, not only has the public smoking ban failed to reduce smoking, its first year has seen an increase in cigarette consumption among males aged 18-34. As the survey notes about smoking after the ban: “There was no significant difference in cigarette smoking prevalence after the implementation of the smoke-free legislation on 1 July. Among smokers, the mean number of cigarettes smoked per day did not fall significantly overall …” In fact, smoking prevalence amongst male smokers increased from 23 [to] 24 percent.

            Even more crucially, among smokers from the lowest socioeconomic quintile—who have some of the highest smoking rates in the U.K.—the number of cigarettes consumed per person increased. Again, 30 percent of smokers reported that the ban had encouraged them to stay at home where they were free to smoke. This validates research by Adda and Corneglia of University College London (The Effects of Taxes and Bans on Passive Smoking, 2006), which found that public smoking bans increased children’s exposure to ETS.

            What the experience of the U.K. public smoking ban shows is that first, the measure was a failure in terms of one of its champion’s central arguments—that it would reduce smoking—and second, that this failure is denied by the health establishment even in the face of its own evidence. So the first problem with the new U.K. tobacco control policies is that they are part of a long line of policy failures, hidden and denied—failures that provide no basis for thinking the latest measures will indeed be somehow different and in fact work.

           

Unsupported

A second problem with the new policy of banning tobacco displays is that there is simply no compelling evidence that it will have any effect on smoking, either by adolescents or adults. This lack of evidence is true in two different areas, first in terms of the academic research about whether seeing tobacco displays leads young people to begin smoking or former smokers to relapse, and second in terms of the lack of reduced youth smoking in countries that have already tried display bans.

            Let’s begin with the evidence about the supposed effect of seeing tobacco displays on young people and adults. Health Secretary Johnson claimed that there was “overwhelming evidence” about the connection between tobacco displays and young people beginning to smoke. But this is not true. The U.K. Department of Health relies on three studies to support its argument that seeing tobacco in shops leads children to smoke. One study (Henriksen et al. 2004) is not even about tobacco displays but rather about tobacco advertising in California, which has little relevance to the U.K.

            A second study (Wakefield et al. 2006) refutes the government’s key contention that seeing cigarette displays leading to smoking. In this study, students were showed pictures of stores filled with tobacco displays. Other students were showed pictures of stores with no tobacco displays. There were, however, no statistically significant differences in approval of smoking between the students who saw stores filled with cigarette displays and those who viewed stores that had no tobacco displays.

            The third study, again by Wakefield (Wakefield et al. 2008), who appears to have done virtually all of the research supporting retail display bans, found that younger smokers were significantly more likely to notice tobacco displays than were older people. More crucially, daily exposure to tobacco displays was not associated with impulse purchases of tobacco—one of the key claims in favor of banning such displays. Indeed, the entire idea that tobacco displays lead ex-smokers to relapse through an impulse purchase of tobacco has virtually no support in the research literature. The U.K. health department’s own experts found that tobacco displays had no connection with impulse purchases of tobacco in the heaviest smokers. Further, attempting to quit smoking in the previous 12 months was not significantly associated with impulse purchase in the Wakefield study. Again, the government fails to include a study by Slater (Slater et al. 2007) that fails to find a statistically significant association between smoking uptake (puffer, experimenter etc.) and smoking relapse and tobacco displays. Finally, the literature on quitting and relapse is enormous, and it contains no support for the government and activists’ claims that seeing tobacco displays is a predictor of smoking relapse.

            Not only does the U.K. government’s evidence not support its assertions, other evidence—evidence from some of the same researchers cited by the government—undermines the government’s claims.

            For example, in another study, Wakefield (Wakefield et al. 2002) acknowledges that tobacco advertising in retail settings is not even focused on attracting new smokers—the key claim of the government—but rather on the branded decisions of existing smokers. In other words, there is here a recognition that becoming a smoker is in fact a two-stage process where the decision to smoke precedes the decision about which brand to smoke.

            And what of the real-world evidence from other countries that have experimented with display bans to reduce youth smoking? The U.K. government claims that other countries such as Canada and Iceland that have implemented display bans have seen impressive reductions in youth smoking as a result. But again, there is little evidence to support these claims.

            For example, the Canadian province of Saskatchewan banned tobacco displays in March 2002. The ban was in effect for 18 months until it was overturned by a court decision. During that period, youth smoking prevalence rates in the rest of Canada declined from 22.5 percent to 22 percent. But in Saskatchewan, with its display ban, youth smoking prevalence increased from 27 percent to 29 percent. In the neighboring province of Manitoba, which did not have a display ban, youth prevalence declined from 28 percent to 23 percent. Again, the comparative youth prevalence data from Canada shows that in 2006 there were no statistically significant prevalence differences between those provinces with display bans and those without. Display bans appear then not to have made a difference in youth smoking in Canada.

            As for Iceland, Statistics Iceland numbers show that in 2000, the year before the display ban, male adolescent prevalence (aged 15-19) was 12.5 percent. In the year the display ban began it was 19.1 percent and the year following that it was 22.5 percent. This hardly supports the contention that display bans have reduced youth smoking.

            A third problem with tobacco display bans is that like many other anti-tobacco measures such as graphic warnings and public smoking bans, they appear to generate reactance in which smokers become increasingly resistant to the external efforts of a judgmental society to force them to change their ways. The more society pushes, the more smokers push back by declining to quit. Curry et al. 1997, for instance, found that smokers who had decided to quit voluntarily and were strongly motivated to quit were far more successful at quitting than those who felt pushed to quit through social pressure or legislation. 

            And while we’re on the topic of unintended and perverse consequences, a display ban also works against sensible tobacco regulation by making legitimate taxed tobacco products look little different from contraband—something that is surely not in the interests of the industry, smokers or the government.

           

Root causes

The final problem with this latest part of the tobacco control menu is that banning tobacco displaysfocuses attention on something that has nothing to do with smoking uptake and thus diverts attention away from some of the genuine sources of youth and adult smoking. As a result some of the root causes of smoking continue to be ignored.

            In 1994, about 36 percent of U.K. women in the lowest socioeconomic group smoked. According to the British Household Panel Survey, a longitudinal survey of about 9,000 households, 28 percent of female smokers aged 18-49 had low-skilled manual jobs, 40 percent lived in social housing, and 30 percent were dependent on means-tested benefits. The most recent figures from National Statistics (Davy Socioeconomic inequalities in smoking: an examination of generational trends in Great Britain, 2007) found little change in smoking prevalence amongst these women over the last 14 years.

            Professor Hilary Graham of Lancaster University, who first wrote about these women in the late 1990s, observed that “Among women in the 18-40 age group who leave school without qualifications, 46 percent are smokers. Within this group, of those whose current or last job was a semi-skilled or unskilled one, 50 percent are smokers. When social housing was added to these two exposures, prevalence rises sharply to 67 percent. When the additional disadvantage of living on means-tested benefits was included, prevalence rises to 73 percent.” The greater a woman’s exposure to various forms of disadvantage, whether educational, occupational, social or economic, the greater her likelihood of becoming a smoker. Similar risks are to be found for young people. The more likely they are to be from poor homes, and do less well in school, the more likely they are to smoke.

            The implications of such statistics for tobacco control policy are significant. Even a cabinet room filled with lightweights should be able to grasp that preventing smoking, both in the young and the not so young, lies in addressing these sorts of socioeconomic factors that predict smoking uptake. Policies designed to keep children in school and allow them to be successful at school, policies to pull people up the socioeconomic ladder, as well as to vastly improve the services upon which they are disproportionately dependent, such as public transport and subsidized housing, offer some of the most practical steps that might be taken to tackle their high smoking rates.

            As Martin Jarvis and Jane Wardle observe about smoking and socioeconomic position, “Improvements in housing, education and employment would target the underlying social conditions which foster high levels of smoking. There is little doubt that substantial progress in this direction would greatly facilitate reductions in smoking ….” (“Social patterning of individual health behaviors: the case of smoking,” 1999)

            But instead of attempting to prevent and reduce smoking by addressing the root causes of smoking among those with the highest smoking rates in the U.K., the government proposes to ban shop displays of tobacco products. This is shameful even for a cabinet room full of lightweights.