Smoking out the facts

On 5 July 2004, two leaders of the UK medical profession were seen hauling a giant cigarette packet up Downing Street.

They delivered letters from 4,500 doctors demanding a government ban on smoking in public places. I, and presumably most of the other 128,000 British Medical Association (BMA) members, had received emails asking us to sign such letters. Meanwhile, Vivienne Nathanson, head of science and ethics at the BMA, was in the news promoting a paper published online in the British Medical Journal (BMJ) that provides new evidence of a link between passive smoking and heart disease (1).

As the self-proclaimed ‘voice of doctors’, the BMA might argue that it is just trying to do the best for public health. But its campaign is built on fuzzy science.

The study Nathanson has trumpeted is, in the words of its authors themselves, ‘modest in size, with limited precision’. The research is based on evidence from a trial set up 20 years ago, between 1978 and 1980, when over 4,000 men gave blood samples, a medical history and a smoking history. The researchers have used that original blood sample to test for cotinine, a breakdown product of nicotine that is considered to be a ‘marker’ for passive smoking, that is, for nicotine exposure levels.

The researchers then correlated the men’s cotinine levels from 20 years ago with their health problems since then. They found an excess risk of coronary heart disease of about 50 per cent in non-smokers who were exposed to passive smoke, as compared with those who weren’t exposed. This is considerably greater than that found in previous trials.

Studies of passive smoking can be inaccurate because they rely on subjects to be honest about the amount they smoke. Using cotinine levels should iron out some of these problems. However, it has been suggested that cotinine levels don’t always accurately predict smoke inhalation, because other sources of ingested nicotine (such as potatoes) could also cause raised cotinine levels in the blood (2).

Another problem with the study is that a significant proportion of the non-smoking group studied were ex-smokers, who have higher risks than normal of developing coronary heart disease. Ex-smokers may also be more likely to start smoking again – the men were surveyed five and 12 years into the study to check that they were still non-smokers, but these responses were not backed up by cotinine testing.

There is also the fact that exposure to tobacco smoke in the UK today is much lower than it was in the late 1970s, as a result of measures such as the banning of smoking in the workplace and on public transport, as well as a decline in the numbers of people who smoke. This raises the question of whether the study’s results hold much relevance for today.

This is not to say the research is completely flawed and ‘bad science’, as one irate correspondent to the BMJ‘s website alleges (3). It would be more accurate to say that it’s not ideal science, and has rather unclear results – like most of the work on passive smoking.

A much bigger trial, published last year in the BMJ, analysed data from the American Cancer Society’s cancer prevention study, an enormous study following 118,000 Californian adults between 1959 and 1998 (4). The researchers looked in detail at the health outcomes of subjects who had never smoked but lived with a smoking spouse, compared to the ‘never-smokers’ who lived with a non-smoking spouse. They found no correlation between living with a spouse who smoked and rates of coronary heart disease and lung cancer. However, this study was heavily criticised on the basis that its authors had received funding from the tobacco industry.

The uncertain state of research on passive smoking contrasts with Richard Doll’s famous study in 1954 establishing the links between active smoking and lung cancer (5). While Doll’s simple study picked up clear indisputable effects, no such striking increased risk from passive smoking has yet been discovered, despite a lot of people looking for it over many years.

Given that the question of whether passive smoking is harmful has been highly controversial, it is disingenuous of the BMA to promote this fairly small study as the final piece in the jigsaw.

The BMA’s campaign around passive smoking is only the latest of its headline-grabbing public health promotions – which have included supporting a ‘fat tax’ on unhealthy foods, and demanding a ban on junk food advertising on TV before the 9pm watershed. In the organisation’s recent response to the government’s ‘Choosing health’ consultation, it reassures the government that it needn’t worry about accusations of nanny statism in health. ‘In relation to certain areas, a greater danger than nannyism is abdication and a failure to act – the Pontius Pilate approach…. The government has a moral duty to ban smoking in public places.’ (6)

It is obviously tempting to play to the gallery, trying to boost public respect for doctors by taking a radical campaigning stance. But for the BMA to throw its weight around, with disregard for the science, is a dangerous tactic. The legitimacy of the medical profession rests on our scientific expertise. In situations such as the MMR vaccine controversy, where it is vital that doctors ‘hold the line’, it is important that we are in a position to be believed. If we don’t always put science first, we may win some headlines but we are in danger of losing our right to be trusted in the long run.

Dr Liz Frayn is a surgical trainee in southwest England.

(1) ‘Passive smoking and risk of heart disease and stroke: prospective study with cotinine measurement’, PH Whincup et al, BMJ

(2) ‘The nicotine content of common vegetables’, EF Domino et al, New England Journal of Medicine, 1993; 329:437

(3) Rapid Responses on BMJ website

(4) ‘Enviromental tobacco smoke and tobacco related mortality in a prospective study of Californians, 1960-98’, Enstrom et al, BMJ, 2003;326:1057-61

(5) ‘The mortality of doctors in relation to their smoking habits: A preliminary report’, R Doll et al, BMJ 1954;ii;1451-5

(6) BMA response to the consultation: ‘Choosing health’, July 2004, available from the BMA website