Welcome jayant

Correspondence

Volume 351, Number 9100 7 February 1998

		Tobacco sponsorship of Formula One motor racing

Sir--Anne Charlton and colleagues (Nov 15, p 1474)¹ claim to show that boys who like motor racing were more likely to start smoking, with the implication that the one caused the other (rather than just showing a coincidence of attitudes). One valid fact is that if either parents or friends smoke, a boy is more likely to start smoking.

In respect of their conclusion about motor racing, a review of the data shows that of the 1063 boys questioned, 125 liked motor racing. Of these, 16 started smoking. This base is so small that if only one of these had answered either question differently, the results would have lost their statistical significance--which is hardly a strong base on which to demand changes in public policy.

I find it disturbing that such weak research and ill-founded conclusions are not only published in The Lancet, but also treated as authoritative in an editorial, and are used to generate much media interest. I note that the research was completed in 1995, but was not published then. It would seem that its publication now is intended to embarrass the UK government, following its recent decision on allowing tobacco sponsorship to continue for Formula One motor racing.

P A Sadler

1 Charlton A, White D, Kelly S . Boys' smoking and cigarette-brand-sponsored motor racing. Lancet 1997; 350: 1474.

Sir--Your Nov 15 editorial ¹ endorsement of Anne Charlton and colleagues '² study of boys' smoking and cigarette-brand-sponsored motor racing, like the study itself, says more about your shaky knowledge of epidemiology, logic, and the Smee report,³ not to say the nature of sponsorship, than it does about the connection between tobacco sponsorship and adolescent uptake of smoking.

It is now widely accepted that there are multiple risk factors for adolescent smoking. Conrad and co-workers⁴ report on over 400 of these, whereas the Goddard report⁵ examines seven. Clearly if smoking is to be described as it increasingly is as a paediatric disease, it is a multifactoral disease. This means that a study, such as Charlton's, which fails to control for the multiple risk factors for adolescent smoking, can tell us nothing about the causes of any adolescent's beginning to smoke. A study of this type can in principle tell us only about the association of certain alleged predictors and adolescent smoking: it can never legitimately use the language of cause and effect nor provide evidence of causality.

Thus, Charlton and colleagues' logic that: "72764 non-smokers might be expected to watch motor racing"; and "of these boys, 9314 would take up smoking", leads to "any ban on tobacco advertising must include all sponsored sport", is specious. Even allowing that the results of the study can be generalised, the study provides neither logic nor evidence to support the claim that tobacco-sponsored motor racing has caused anyone to take up smoking. Indeed, it is just as reasonable to conclude, from the data presented, that both adolescent interest in motor car racing and smoking are caused by some other factor as it is to conclude that adolescent smoking is caused by tobacco sponsorship of motor-car racing.

Indeed, had Charlton et al been more attentive to the Smee report they would have discovered this before undertaking a study that by definition can demonstrate nothing. For instance, in reply to the comment that "these boys were significantly more likely than the others to name Marlboro . . . and Camel", Smee³ notes that "awareness of advertising is at most a necessary condition for coming under its influence. It is not reliable evidence that advertising increases consumption". Again Smee notes that although adolescents may well recognise tobacco advertisements it does not follow that such advertisements initiate smoking since "it is also possible that those children who react most positively to advertising are already disposed to smoke".

Finally, it is true that the "Smee report made it clear that cigarette advertising is associated with tobacco consumption", but not quite in the way that Charlton and co-workers claim. As to the effect of advertising on consumption in the UK, Smee noted that "advertising does not have a statistically significant effect in any form". Moreover, Smee carried out a detailed statistical analysis of whether tobacco advertising is "likely to influence the smoking prevalence among teenagers". A separate model was developed for both teenage men and women and "in neither model was advertising a significant factor".³

J C Luik

1 Editorial . Labour government's tobacco spin spins them off track. Lancet 1997; 350: 1411.

2 Charlton A, White D, Kelly S . Boys' smoking and cigarette-brand-sponsored motor racing. Lancet 1997; 350: 1474.

3 Smee C. Effect of tobacco advertising on tobacco consumption: a discussion document reviewing the evidence. London: HM Stationery Office, 1992.

4 Conrad L, et al. Why children start smoking cigarettes: predictors of onset. Br J Addiction 1992; 87: 1711-24.

5 Goddard E. Why children start smoking. London: HM Stationery Office, 1990.

Authors' reply

Sir--P A Sadler's first point concerns the causal nature of cigarette sponsorship on Formula One motor racing and boys' smoking. Our research was a cohort study that followed the same children over 5 years from the ages of 9 and 10 years to 13 and 14 years. Thus the study was longitudinal and not cross-sectional, and it was possible, as we said, to observe that the watching of motor racing preceded the onset of regular smoking by up to a year. We also examined the boys' self-perception in six domains¹ and in no respect were those who watched motor racing significantly different from other boys.

The links between children's smoking and their parents' and best friend's smoking are well-known, as Sadler points out.² We were well aware of this when we designed the study and planned the analysis, which is why these factors were included. The fact that multiple influences operate on children's uptake of smoking does not negate or reduce the strength of sponsored motor racing as an independent significant factor.

With respect to the small numbers involved, we emphasise that if the scenario that he suggests is projected to the estimated population of 606365 non-smoking boys in this age group the odds ratio would be 1·81 (85% CI 1·77-1·86). In the study as reported, the findings with regard to the onset of regular smoking in relation to watching Formula One motor racing were still significant. The boys in question did more than just experiment with smoking, which would have represented a much larger group: they became regular, soon to be addicted,³ smokers.

As previously stated, the research spanned 5 years. The findings were not held back or specially produced to embarrass the UK government, following its recent decision, as Sadler suggests. The analysis of 5 years' complex data takes time and we recently covered the topics of advertising and sponsorship spanning the whole period. We were so struck by the link between watching motor racing and the onset of regular smoking in boys at the age of 13 years--the age the cohort reached in 1995--that we decided to publish it as a separate item. The age of 13 years is known to be a critical one in the establishment of regular smoking.⁴ In fact, the letter reached The Lancet on Oct 15, 1997, and was accepted for publication well before the government's announcement.

In reply to J C Luik, we make the point that cigarette-brand sponsorship in connection with Formula One motor racing is a specific type of advertising. The product is strongly associated with fast cars, risk taking, danger, the macho image, and fame, and therefore cannot be equated with other advertising strategies.

*Anne Charlton, David While, Sheila Kelly

1 Harter S. Manual for the self-perception profile of children. Denver: University of Denver, Colorado, 1985.

2 Royal College of Physicians. Smoking and the young. London: Royal College of Physicians, 1992.

3 McNeill A. The development of dependence on smoking in children. Br J Addiction 1991; 86: 589-92.

4 Jarvis L. Smoking among secondary school children in 1996: England. London: Office for National Statistics, 1997.

Sir--You rightly criticise the donation of £1 million to the Labour Party by the Chief of Formula One Constructors Association (Car Racing) and the exemption of Formula One motor racing from tobacco advertising ban in your Nov 15 editorial .¹ In Goa, this news brings to mind the tactics used by the tobacco industry to delay, scuttle, and dilute the provisions of the The Goa Prohibition of Smoking and Spitting Bill, 1997.

This bill was supported by about 20 medical and other voluntary organisations under the auspices of National Organisation for Tobacco Eradication. The Bill bans smoking and spitting in public places and vehicles, advertisements of all types, direct or indirect, sale of tobacco products to individuals younger than 21 years of age (age for consent for marriage in India), and to anyone within 100 m of an educational establishment or a place of worship. Violation of the clauses on advertisement and sale to minors will lead to imprisonment for up to 3 months; other offences will result in various levels of fines. Non-governmental organisations will have some powers in the execution of these provisions.

The Goa Government first tried to delete the advertisement-ban clause, but later, on the recommendation of the select committee, the bill was passed unanimously by the Legislative Assembly. Sadly, the Governor's consent, which usually is a mere formality, was declined. The bill has now been referred to the President of India and it will become law only with his assent. However, there is a precedent of a similar but less stringent bill in New Delhi that was passed by the President and this is encouraging.

The influence of the tobacco industry extends from local government to Union government. Since 1992, the Indian Tobacco Company (ITC), formerly Imperial Tobacco Company (a subsidiary of British American Tobacco), has made contributions to the Prime Minister's and Chief Minister's Relief Fund of up to 27 lakhs (2·7 million Rupees, £50 000). The ITC has also contributed generously to the Rajiv Gandhi Foundation. These are some of the ways that the industry ensures inaction and silence and buys respectability and a good corporate image for the promotion of tobacco products. Do we want our health and our children's health to be sold in this way? Nevertheless, we hope that the UK government that can so easily ban the sale of beef because of the death of 23 people since 1995,² can lead the world in banning the sale of tobacco, which kills 120 000 people in the UK every year.³

*Sharad G Vaidya, Jayant S Vaidya

1 Editorial . Labour Government's tobacco spin spins them off track. Lancet 1997; 350: 1411.

2 Department of Health web page. http://www.coi.gov.uk/depts/GDH/ coi529od.uk

3 Health Education Authority press release. Dec 10, 1997.