“The truth is rarely pure, and never simple,” railed the editorial writer from the Sunday Telegraph on 8 March, forgetting to acknowledge Oscar Wilde. Oh yes, they were on to a good one. The spin doctors down at British American Tobacco (BAT) had passed the newspaper an exclusive: “Passive smoking doesn’t cause cancer—official.”
According to the piece, the World Health Organisation’s International Agency for Research on Cancer (IARC) had discovered that environmental tobacco smoke was just epidemiological hot air. But worse still, the infamous nanny factory, “whose institutional raison d’etre is to interfere as widely as possible in the day to day life of as many people as possible,” had been caught by BAT trying to withhold this embarrassing stuff from us all, slipping it discreetly into a few paragraphs in their biennial report. The Sunday Times picked up the story from an early edition of the Sunday Telegraph and uncritically accepted it in a late night editorial decision. The claims were repeated prominently in media in several other countries.
Truth in matters of passive smoking is rarely pure when we hear from the tobacco industry. A BAT memo from a 1988 London meeting talked of selection of industry funded scientific consultants, who “ideally ... have no previous record on the primary issues,” to “cooperate within the confines of decisions taken by Philip Morris scientists to determine the general direction of research, which apparently would then be filtered by lawyers to eliminate areas of sensitivity” (BMJ 1997;314:1569).
BAT’s head of science, Dr Chris Proctor, expressed “unconcealed delight” at the IARC report. So what was the cause of Proctor’s public relations optimism? The study of 650 lung cancer cases and 1542 controls, conducted in seven European countries, found a 16% increased risk of lung cancer in non-smokers whose spouse smoked and a 17% increased risk if they worked in a smoky workplace. It estimated there were up to 1100 deaths from passive smoking in the European Union each year. As if their reports had come straight from a BAT press release, none of this was so much as mentioned by the two London newspapers.
Instead the big “news” was that the lower 95% confidence limit on the estimate was 0.93, thus nudging it a whisker below statistical significance and allowing the wide eyed Victoria Macdonald from the Sunday Telegraph to go a step further: passive smoking might actually be good for us (“not only might there be no link between passive smoking and cancer but that it could even have a protective effect.”) Curiously, her impartial report did not find room to note that the upper confidence limit was 1.44. 
As yet there are no reports of Victoria Macdonald following up her insights with inquiries to Britain’s neonatal wards as to why they are not considering piping tobacco smoke into the wards to “protect” babies. Had the mesmerised hacks at the Sunday Telegraph and Sunday Times bothered to consider all the studies on passive smoking and lung cancer, they would have seen the IARC report in perspective. A meta-analysis of studies of lung cancer and passive smoking published in the BMJ last year (1997;315:973-80, 980-8) examined 4626 cases and 477 924 controls in 39 studies and found a highly significant 24% excess risk of lung cancer in non-smokers living with smokers (95% confidence interval 1.13 to 1.36).
BAT had known about the IARC study for some time, with the timing of the release calculated to pour cold water on the report of the Scientific Committee on Tobacco and Health, which the tobacco industry correctly expected would be released on 11 March. In late January this year, public affairs staff at BAT’s Australian subsidiary WD&HO Wills began inviting Australian journalists to individual briefings on passive smoking. Wills’ “manager of scientific issues,” Dr Linda Rudge, wanted to give the journalists the benefit of her impartial insights into the thorny topic. At the briefings, Dr Rudge told the journalists that the IARC report would be forthcoming and that they should feel free to contact her for advice on its meaning. On 9 March, Wills issued a statement welcoming the release of the IARC study, adding that the “vast majority” of all studies on passive smoking and lung cancer had “failed to establish such a risk.” In fact 31 of 39 studies reported relative risks greater than unity (see www.bmj.com).