British Report Says Risk from Passive Smoking Overstated

By Patrick Goodenough | July 7, 2008 | 8:07 PM EDT

London ( - The extent of the link between passive smoking and lung cancer has been questioned by British statisticians who say medical journals tend to ignore studies that find no link, while focusing on those that do.

Statisticians from the University of Warwick reviewed 37 previous studies, which appear to indicate that the risk of lung cancer in those exposed to other people's smoke is increased by 24 percent.

In fact, because of what Professor John Copas and Dr. Jian Qing Shi call "publication bias," the real risk is considerably lower - around 15 percent.

The findings, published in Friday's edition of the British Medical Journal, were welcomed by the British tobacco industry.

John Carlisle, director of public affairs at the Tobacco Manufacturer's Association, told that the industry had long argued there was "no strong statistical evidence" that passive smoking was a serious risk to health.

"Certainly passive smoking is a nuisance to some people and unpleasant to some people ... it seems that this report to a certain extent would confirm our opinion and lessen the alarmist impact which previous reports from the World Health Organization and others have produced, which frankly were taken out of context and did not take the statistics as a whole."

Carlisle recalled that the WHO was criticized in 1998 for allegedly suppressing the results of tests carried out in Europe, which did not deliver the desired results.

After examining lung cancer patients in seven European countries over a seven-year period, the WHO found that exposure to passive smoking at home or in the workplace did not increase risk in a statistically significant way, according to reports from the time.

"We think there's been some very alarmist talk from so-called responsible health bodies. Perhaps now they should tone down their language and look at the statistics as we have done in the past, and draw more sensible conclusions than the rather dramatic ones [drawn in the past]."

He said there had been about 40 tests on passive smoking - also called environmental tobacco smoke (ETS) - over the past 20 years.

The Warwick University specialists estimated that some 23 earlier studies that did not link passive smoking with lung cancer may have been ignored. Researchers often average all previous findings, but the overall picture may be distorted because some reports with negative findings have not been written up in journals.

Because of this "publication bias," they suggest previous levels of risk should be interpreted with caution.

Prof. Copas warned, however, that it was wrong to suggest that researchers were ignoring results that did not match their "predetermined views."

"It is just the tendency, always present, that studies which come to interesting and positive findings are more likely to be accepted for publication in the journals than studies which are inconclusive," he told Friday.

Last month researchers at the U.S. Department of Energy's Oak Ridge National Laboratory in Tennessee found that exposure to ETS could be lower than indicated in earlier studies for one group assumed to be at risk - bartenders, waiters and waitresses.

In March 1998, the WHO issued a statement defending itself against charges that it had suppressed the results of its European study, and urged people not to be fooled into thinking passive smoking did not cause lung cancer.

"The results of this study, which have been completely misrepresented in recent news reports, are very much in line with the results of similar studies both in Europe and elsewhere: passive smoking causes lung cancer in non-smokers," WHO said.

"The study found that there was an estimated 16 percent increased risk of lung cancer among non-smoking spouses of smokers. For workplace exposure the estimated increase in risk was 17 percent. However, due to small sample size, neither increased risk was statistically significant."

WHO's press office in Geneva did not respond Friday to inquiries from

A British appeals court last year upheld the legality of plans by the Labor government to outlaw cigarette advertising. Tobacco manufacturers will take their case for further appeal to the House of Lords in May, and to the European Court of Justice later in the year.

The companies concerned, Imperial, Gallaher, Rothmans UK and British American Tobacco, reject claims that advertising encourages people to start smoking, saying it only affects consumer choice regarding which brands to use.

Patrick Goodenough
Patrick Goodenough
Spencer Journalism Fellow