Publishing research supported by the tobacco industry

Journals should reverse ban on industry sponsored research

The tobacco industry's behaviour is as noxious as its products. During the past year, tobacco executives have denied that nicotine is addictive, spent heavily to stop antismoking laws in the American Congress and states, prosecuted their own former executives who have tried to speak out, and even threatened the media's attempts at investigations.1 So it is not surprising that the American Thoracic Society, the scientific arm of the American Lung Association, decided last month that it would no longer accept any medical research that is funded by the tobacco industry in its two peer reviewed journals (6 January, p 11.)2 The decision was a step further in the medical society's laudable fight against tobacco, but it was a misguided one.

Tobacco accounts for a third of deaths in the developed world3 and about 10% of medical costs in the United States result directly from tobacco use. More perniciously tobacco advertising has been shown to be aimed successfully at young people.4 In Washington, the industry spends millions to maintain its own livelihood by being, according to one congressman, “the most pervasive lobbyist in politics today.”5 The tobacco industry also promotes itself through research grants. In 1994 in the United States it distributed $19.5 million for research, which resulted in 375 scientific papers. Nearly 1100 investigators at more than 300 institutions, including medical schools, have accepted grants since funding began in 1954.6

We praise the American Lung Association and the American Thoracic Society for their firm stand against tobacco. Indeed, the society's directive that its members should not accept funding from the tobacco industry is a good step, although enforcement will be difficult. But the extension of the rule into the pages of its scientific journals, the American Journal of Respiratory and Critical Care Medicine and the American Journal of Respiratory Cell and Molecular Biology, is a threat to medical science, to journalism, and ultimately to a free society.

The editors' decision was made after much thought, and the leaders of the American Thoracic Society said it was essentially a moral decision. Before the decision was made the issue was debated by two medical ethicists in the American Journal of Respiratory and Critical Care Medicine. Arthur Caplan of the University of Pennsylvania wrote: “Any organisation committed to the goal of preventing respiratory illness and disability and to working with government agencies who seek to do everything in their power to reduce the use of tobacco products among children and adults cannot remain credible if it permits research sponsored by the tobacco industry in its publications.”7 But H Tristram Engelhardt Jr of Baylor University used the “slippery slope” moral argument. “If receiving tobacco money is unacceptable, why is it acceptable to take governmental money if acquired through unjust taxation policies? Or if the party in power does not support health reform in accord with ATS and ALA views? … Or if the funds come from corporations that sell tobacco products?”8

But the issue is more than a moral one, as Engelhardt implies. Scientific inquiry rests on investigation that is presented to other scientists for their review and judgement. Every sponsor of every study has an agenda to promote, but all studies undertaken must be made available in some form: if some studies are systematically suppressed then we will reach false and biased conclusions when reviewing a body of research. The peer review system is designed to try and ensure that the conclusions of studies are supported by the evidence they contain, but peer review cannot guarantee the validity of studies. Studies must be published so that other researchers can make up their own minds on their quality. Because peer review cannot guarantee the validity of a study and because bias operates very subtly, many journals, including this one, print authors' funding sources alongside papers. By doing so, the journals ensure that the ultimate peer reviewers, practising doctors, can use that information to make up their own minds on the validity and usefulness of a piece of research. (Some readers may consider drug company sponsored research to be suspect, never mind the tobacco industry.) By impeding the process of scientific publishing the American Thoracic Society is being antiscientific.

The leadership of the society says that taking tobacco money represents a conflict of interest for researchers. Engelhardt wonders where it will all end: Will smokers be banned from the pages of the journals? He also raises an issue that is too seldom discussed: the conflicts of interests beyond money. They can be personal, political, academic, and religious,9 and Engelhardt adds personal habits. Conflicts of interest cannot be eradicated from medical journals, and the American Thoracic Society is deceiving itself to think it can.

Most journals subscribe to the International Committee of Medical Journal Editors' (Vancouver Group) position on conflict of interests by requiring authors to disclose their own perceived conflicts.10 Journals have more to do in this difficult area, because what authors perceive and what readers might perceive as conflicts of interest can be very different. But for the time being, many journals, including the BMJ, tell readers the conflicts that authors feel they have.

A third strand of this argument is one that is most critical to Americans: freedom of the press through prohibition on prior restraint of information. It is a centrepiece of the US Constitution. This fundamental freedom, which dates to the Enlightenment, is based on the idea that a free and open marketplace of ideas fosters truth. “That the best test of truth is the power of the thought to get itself accepted in the competition of the market …. That, at any rate, is the theory of our Constitution,” wrote the jurist son of the great American physician, Oliver Wendell Holmes. To ban scientists who take tobacco money is to restrict the freedom—and so the effectiveness—of the press.

The ban of tobacco funded research by the two American journals turns two respected scientific journals into publications with political agendas. In the end, this will make them little more than house organs for one group. And no matter how praiseworthy the goals of that group, it and its publications will be diminished. In the nasty debate over tobacco, the industry may well welcome the journals' censorship. It may try to say, “See, you are no better than we are.” We urge the American Thoracic Society and its journals to reverse their decision and fight the tobacco industry not with censorship but with the abundant evidence on the serious harm that its product inflicts.