Journal policy on research funded by the tobacco industryBMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f5193 (Published 15 October 2013) Cite this as: BMJ 2013;347:f5193
All rapid responses
I would like to suggest the book "Merchants of DOUBT" by Naomi Oreskes and Erik M. Conway. There are good chapters about the tobacco industry which are worth reading.
Competing interests: No competing interests
It has been apparent for a long time that every tobacco product is directly harmful, and research is about quantifying and minimising the harm caused. Even its worst critics cannot say the same about pharmaceutical products, and companies wish to provide useful and beneficial medicines or devices. There is no moral reason not to publish research, which is usually the basis of their product development.
Not surprisingly, they wish to present it in the best light, and prefer to present work that does so. I would strongly support the current moves toward ensuring that all trial data is made available, to counteract this bias. If that can be done, then it becomes the responsibility of the publisher to present the data. Of equal importance is the responsibility of the reader to interpret the data when deciding on the use of the product. Too few of us, I suspect, review the statistical treatment of the data in research articles, or the methodology of the research, but how else can we assess its value.
Banning research funded by 'Pharma" is unnecessary, provided that we understand and expect bias; a clear statement of conflicting interest is needed, of course. However, if we wait for entirely independent funding of all research into treatment, there may be precious little of it. Those who pay for health care (Government or insurers) may also be biased, toward cheaper or older treatments, and research they fund might be biased. It possibly would discourage new product development, too.
We should employ our critical faculties when reviewing research, not rely on a blanket ban to protect us from the possibility, or even likelihood, of bias.
Competing interests: I rarely prescribe any product other than vaccines. I have previously worked as an NHS prescribing adviser. I have no connection with any pharmaceutical company.
The BMJ has no plans to revisit its decision.
1 Godlee F, Malone R, Timmis A, Otto C, Bush A, Pavord I, et al. Journal policy on research funded by the tobacco industry. BMJ 2013;347:f5193. (15 October.)
Competing interests: I am Editor in chief of the BMJ and a director of BMJ, which is a wholly owned subsidiary of the BMA. BMJ receives revenues from pharmaceutical and device manufacturers in the form of advertising, reprint sales, and sponsorship. I receive a bonus based in part on the financial performance of BMJ. The BMJ is an open access journal that charges author fees for publication of research articles. I have campaigned on issues of research transparency and integrity for quite a few years. The BMJ is a founding member of Alltrials.
We are surprised and disappointed by your recent decision not to consider for publication in the BMJ Group journals any scientific studies funded by the tobacco industry. We have concerns on at least four counts:
1. The detrimental effect your decision could have on ensuring that the widest possible range of scientific research is funded and promulgated.
2. The general lack of consideration for the valuable contributions that industry can make to advance scientific research in full compliance both with governmental policies and increasingly stringent academic norms of transparency around the world.
3. The drastic choice of a censorship approach rather than a focus on improving and evolving the journal’s own critically important peer-review process.
4. The choice to isolate one industry from participating equally in the development of scientific knowledge based on prejudice originating from ideology rather than science while continuing to support and favour other industries that also have ideological detractors, as some critics have observed.
The BMJ’s new policy of banning consideration of scientific studies based on their source of funding is particularly disappointing in the light of the BMJ’s historical policy of encouraging robust scientific discourse independent of ideology. Ten years ago the BMJ was “passionately anti-tobacco” but also “passionately pro-debate and pro-science” and further commented that the type of ban recently instituted “would be anti-science” . Allegations of misconduct by the tobacco industry were cited at that time in support of a proposed ban that the BMJ rightly rejected.
It is ironic that the BMJ has now revised its prior view. There is a new commitment to transparency of funding and potential conflicts of interest throughout the scientific world. There is also a renewed interest in tobacco science. In 2009 the US Food & Drug Administration (FDA) was given jurisdiction to regulate tobacco products in the US. In order to create an evidence-based regulatory framework, the FDA is investing billions of research dollars to fill the scientific gaps in tobacco regulatory science (e.g. ). The FDA also recognises that the regulated industry has a role to play in this emerging community of tobacco regulatory scientists, alongside other stakeholders including the regulator, government and public health.
It has been argued that tobacco harm reduction is potentially the world’s greatest public health opportunity today . To have this kind of impact, those with an interest must find pragmatic ways to work together to find solutions based on sound science. For this research to be disseminated widely, it will be important for the science publishing industry to retain an independent, critical, yet open approach.
In recognition of the critical role industry can play in advancing science, governments around the world are looking to industry to help fund collaborative academic research projects to help improve the competitiveness of their nations (e.g. ).
Professor Anne Glover, Chief Science Advisor to president of the European Commission, pointed out recently in a Commentary in the Financial Times , that although “it has been research policy for many years” there is also “…a fundamental mistrust of industry and industrial R&D by society – and this mistrust is increasingly hampering our [Europe’s] ability to innovate”. Glover goes on to say that a large part of the problem is the way that industry has behaved when doing business and talking to public. She says that industry must “…engage with its critics and examine its practices”. She stresses the need for transparency on all sides, and points out that although there can be actual or perceived conflicts of interest, there are clear ways of handling these through the peer-review process.
In recent years British American Tobacco has also been evolving its approach to become more transparent in our research efforts, as well as embedding rigorous quality standards. We encourage our scientists to publish (we have published over 100 manuscripts in peer-reviewed journals over the past 5 years), attend conferences and participate in scientific debate. We share scientific conference presentations through the Library of www.bat-science.com. We register every clinical trial in advance, and commit to publishing the results. Every academic who receives funding for fundamental research from BAT today is encouraged to publish any and all results that arise from the project, irrespective of the findings, as well as to acknowledge the funding source.
Should you be interested to learn more about our science, and how our scientists work on tobacco harm reduction, please feel free to come and visit us. Our R&D team in Southampton would be happy to host you and address any questions you may have in an open and transparent fashion, as we do with all who come to visit us.
Science should be judged on its merits as ascertained through peer-review, and good science should be published. The source of funding should be fully and openly disclosed, be it industry, the public health sector, government or any other source. Most journals today require detailed disclosure of any Competing Financial Interests, which we fully support.
We have also noted statements you have made in relation to the peer-review process currently followed at the BMJ. It appears that the BMJ Editors are concerned with the effectiveness of their peer-review system. This is a serious issue that needs to be addressed. Studies approved through peer review have long been the currency of scientific credibility. The reputation of the BMJ is clearly dependent on this capability functioning properly. Applying selective censorship does not aid in addressing fundamental problem, and separates the BMJ from journals of unassailable scientific integrity and relegates it to the class of journals that politicise science.
We respectfully invite you to revisit your decision not to consider for publication any scientific studies funded by the tobacco industry on the basis that such a policy is, in the BMJ’s own words, “anti-science”.
1. Smith, R. (2003). Comment from the Editor on Passive smoking. BMJ 327:505.2
2. Announcement of Awards for Tobacco Centers of Regulatory Science (TCORS). http://www.fda.gov/TobaccoProducts/PublicHealthScienceResearch/ucm369005...
3. Sweanor D. et al. (2007).Tobacco harm reduction: how rational public policy could transform a pandemic. Int J Drug Policy doi: 10.1016/j.drugpo.2006.11.013
4. Wilson, T. February 2012. A Review of Business-University Collaboration.
5. Glover A. Question of trust depends on openness. Financial Times, 17 October 2013.
Competing interests: I am employed as Chief Scientific Officer at British American Tobacco.
As a long-term consultant to the tobacco industry, and an occasional author in, and reviewer for, the BMJ, I was saddened to see the journal’s recent decision to no longer consider for publication any study partly or wholly funded by the industry. When I left the Tobacco Research Council in 1979 and became a consultant in epidemiology and medical statistics to various tobacco companies, it is interesting to note that the senior research scientist of one of the companies told me that they wanted to hear my opinion on what the available evidence actually shows, and not what I thought that they wanted to hear. Indeed I would not have wanted an alternative arrangement. Since that time I have published widely on the epidemiology of tobacco and health, as well as providing the industry with commentary on many publications and reports, and conducting analyses of relevant data. I have always attempted to give an unbiased opinion.
My concern is that a number of messages coming out on tobacco and health put out by the tobacco control community do not present a balanced view of the evidence, and that stopping the industry or its consultants having their say will inevitably increase this bias. While some industry statements in the past have been biased in the opposite direction, suppression of all their views, regardless of their validity, can hardly advance science.
Of course cigarette smoking is extremely harmful, as my colleagues and I have made clear to the industry from the start, and are still making clear in our publications (see for example my recent meta-analyses on smoking and lung cancer1-3 and on COPD4 and on quitting smoking5,6). Indeed I remember that, when still working with the Tobacco Research Council about 1970, I had extended interchanges with the late Philip Burch arguing against his claim (e.g. 7) that the smoking/lung cancer relationship was essentially artefactual.
However, there are a number of areas where the situation is not so clear cut, and where the evidence has been misrepresented by tobacco control advocates.
One concerns the relationship of secondhand smoke (passive smoking) to lung cancer. Although an association is quite consistently seen in many epidemiological studies, I have presented detailed arguments that most if not all of this association can be explained by bias and confounding8. My arguments have never been refuted, only ignored.
Another is the effect of tar/nicotine reduction, where the message being put over is that this provides no health advantage whatsoever, and that the whole idea is some sort of con trick put over by the industry. While it is clear that, on switching to lower tar/nicotine cigarettes, smokers “compensate” by increased inhalation and/or puff volume, such compensation is only partial9, with uptake of smoke constituents lower following the switch. Consistent with this is the clear evidence that there is a reduced risk of lung cancer associated with the reduction10, this increase not being explained, as suggested by the Institute of Medicine11 by inappropriate adjustment for amount smoked12. Though the proportional reduction in the smoking-related risk of lung cancer is certainly less than the proportional reduction in yields, it certainly exists.
While switching to a lower tar and nicotine yield cigarette may only have a modest effect on risk, other switches may be much more beneficial to those unable or unwilling to forgo their nicotine. Swedish moist snuff, or “snus”, is the alternative product with by far the most evidence, and a series of industry-funded papers13-16 have made it clear that any health effects of snus are vastly less than those of cigarettes, and hardly detectable at all. Some reviews by those not funded by industry have not made this clear, which may explain why, somewhat amazingly, it remains legal to sell cigarettes in all European Community countries, but snus only in Sweden.
When it comes down to it, evidence is evidence, and suppressing valid data simply because of who financially supported the research cannot be right. Many leading figures in smoking epidemiology have been willing to have discussions with industry. Why should not the BMJ also be willing to do so?
PS This letter was not financially supported by the industry.
References Style BMJ
1. Lee PN, Forey BA, Coombs KJ. Systematic review with meta-analysis of the epidemiological evidence in the 1900s relating smoking to lung cancer. BMC Cancer 2012;12:385.
2. Lee PN, Forey BA. Indirectly estimated absolute lung cancer mortality rates by smoking status and histological type based on a systematic review. BMC Cancer 2013;13:189-224.
3. Fry JS, Lee PN, Forey BA, Coombs KJ. Dose-response relationship of lung cancer to amount smoked, duration and age starting. World Journal of Meta-Analysis 2013;1:57-77.
4. Forey BA, Thornton AJ, Lee PN. Systematic review with meta-analysis of the epidemiological evidence relating smoking to COPD, chronic bronchitis and emphysema. BMC Pulm Med 2011;11:36.
5. Lee PN, Fry JS, Hamling JS. Using the negative exponential distribution to quantitatively review the evidence on how rapidly the excess risk of ischaemic heart disease declines following quitting smoking. Regul Toxicol Pharmacol 2012;64:51-67.
6. Fry JS, Lee PN, Forey BA, Coombs KJ. How rapidly does the excess risk of lung cancer decline following quitting smoking? A quantitative review using the negative exponential model. Regul Toxicol Pharmacol 2013;67:13-26.
7. Burch PRJ. The biology of cancer: a new approach. Lancaster: MTP Press Limited; 1976.
8. Lee PN, Fry JS, Forey BA. Revisiting the association between environmental tobacco smoke exposure and lung cancer risk. V. Overall conclusions. Indoor Built Environ 2002;11:59-82.
9. Scherer G. Smoking behaviour and compensation: a review of the literature. Psychopharmacology 1999;145:1-20.
10. Lee PN. Lung cancer and type of cigarette smoked. Inhal Toxicol 2001;13:951-76.
11. Stratton K, Shetty P, Wallace R, Bondurant S, editors. Clearing the smoke. Assessing the science base for tobacco harm reduction. Washington, DC: National Academy Press; 2001.
12. Lee PN, Sanders E. Does increased cigarette consumption nullify any reduction in lung cancer risk associated with low-tar filter cigarettes? Inhal Toxicol 2004;16:817-33.
13. Lee PN. Summary of the epidemiological evidence relating snus to health. Regul Toxicol Pharmacol 2011;59:197-214.
14. Lee PN. The effect on health of switching from cigarettes to snus - a review. Regul Toxicol Pharmacol 2013;66:1-5.
15. Lee PN, Hamling JS. The relation between smokeless tobacco and cancer in Northern Europe and North America. A commentary on differences between the conclusions reached by two recent reviews. BMC Cancer 2009;9:256.
16. Lee PN, Hamling JS. Systematic review of the relation between smokeless tobacco and cancer in Europe and North America. BMC Med 2009;7:36.
Competing interests: I am a long-term consultant to the tobacco industry
Richard Smith makes a good point.(1) Like the tobacco industry the pharmaceutical industry has been found repeatedly to have manipulated and suppressed research data in order to make its products appear better than they are. Non-industry funded research also suffers from such misconduct but the scale and influence of pharma funded research makes its current state of particular concern.
I am firmly of the view that clinical practice should be based on independent research. Drug and device companies should not be evaluating their own products. All phase III trials should be designed, analysed, and reported independent of the manufacturers. Clinical practice guidelines should be free of industry bias.
The BMJ is working on its own (2) (3) and in support of others (4) (5) to increase transparency and accountability in clinical trials, to reduce the influence of industry in medical education and clinical guidelines,(6) and to unpick the legacy of research bias and misconduct. (7) (8) If these efforts do not soon bring about a necessary sea change in the way industry funded trials are performed, the BMJ may well decide to stop publishing them. Whether an editor would survive such a decision is a question I may have to test.
I would like to hear readers’ views on what more we and others should be doing and especially on whether journals should continue to publish research funded by drug and device manufacturers.
1. Smith R. The same arguments mean that journals should stop publishing research funded by the pharmaceutical industry. http://www.bmj.com/content/347/bmj.f5193/rr/668130
2. Godlee F, Groves T. The new BMJ policy on sharing data from drug and device trials. BMJ 2012;345:e7888.
3. Open data campaign http://www.bmj.com/open-data
4. Alltrials http://www.alltrials.net/
5. Kmietowicz Z. European regulator appeals against order stopping it from releasing clinical study documents. BMJ 2013;347:f4728. (24 July.)
6. Lenzer J. Why we can’t trust clinical guidelines. BMJ 2013;346:f3830 (14 June.)
7. Abramson JD, Rosenberg HG, Jewell NJ, Wright JM. Should people at low risk of cardiovascular disease take a statin? BMJ 2013;347:f6123. (22 October.)
8. Doshi P, Dickersin K, Healy D, Vedula SW, Jefferson T. Restoring invisible and abandoned trials: a call for people to publish the findings. BMJ 2013;346:f2865. (13 June.)
Competing interests: I am editor in chief and a director of the BMJ, which is a wholly owned subsidiary of the BMA. BMJ receives revenues from drug and device manufacturers in the form of advertising, reprint sales, and sponsorship. I receive a bonus based in part on the financial performance of the BMJ. The BMJ is an open access journal that charges author fees for publication of research articles. I have campaigned on issues of research transparency and integrity for quite a few years. The BMJ is a founding member of Alltrials.
One would expect that Public Health departments, the WHO, Unicef, etc would make it their priority to check on harm from their nationwide programs with longterm cohorts of vaccinated and unvaccinated children, in particular in order to check Non Communicable Diseases (NCD).
As standard mass intervention programs should have surveillance studies on harm. Or could it be that public health research has suffered from non-publication bias?
Is an "all trials" register available for Public Health research? http://www.alltrials.net/supporters
It doesn't look proper that over the decades, new vaccines have been introduced to billions of children and that there is a clear increase in NCD, which is explained in medical journals as "due to environmental factors". At present there is a push to vaccinate a further quarter of a billion children in developing countries. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(13)62127-3/fulltext Why is this done without properly designed local ongoing surveillance cohorts to check consistently for NCD?
Vaccine safety surveillance studies seems to get away with self serving checks for vaccine coverage, vaccine effectiveness or days spent in hospital due to infection morbidity or mortality. There appears to be a void in cohorts designed to check over the following years, on harms like NCDs such as asthma, pervasive developmental disorders and ADHD.
Is in public health a similar tendency prevailing to the pharmaceutical industry's, to publish studies that serve their interest? Are only cohort study results published for 'non-specific vaccine side effects' which are negative on harm, eg http://www.bmj.com/content/318/7192/1169 as there seems to be a scarcity in surveillance cohorts reporting on possible harm in vaccinated compared to unvaccinated children?
An intervention program by the WHO in Egypt illustrates the importance of surveillance programs on harm. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(99)06527-7/fulltext
However, in the UK, where for economic reasons, a Rotavirus vaccine and flu vaccine for infants were added this year, this again has not resulted in initiating ongoing surveillance cohorts to check on NCDs, such as inflammatory bowel diseases, asthma, pervasive developmental disorders and ADHD?
Somehow, with reference to vaccines, there are no red flags and everyone is comfortable?
Competing interests: No competing interests
It is easy to understand what motivates the BMJ's policy on refusing research funded by the tobacco industry. There is no doubt that the tobacco industry does immense harm, and that they have been responsible for promulgating a great deal of misleading information about the harms of tobacco.
However, a blanket policy of automatically rejecting all research from the tobacco industry makes me uncomfortable. It is a policy based on nothing other than an ad hominem fallacy. Much research funded by the tobacco industry is indeed flawed and unworthy of publication in as prestigious a journal as the BMJ, but surely such research should be rejected because of its flaws, rather than who funded it?
It is important to consider what incentives this gives to the tobacco industry. If we imagine that tobacco industry researchers would like to have research published in the BMJ, then a policy of accepting good quality research and rejecting flawed research gives them an incentive to up their game. If you are going to reject their research anyway, you remove that incentive.
The recent rise of e-cigarettes is an important development in reducing harms from tobacco, but there is much we have yet to learn about them. Can e-cigarettes help smokers to quit? If smokers switch to e-cigarettes on a long-term basis instead of quitting, what are the harms (almost certainly less than smoking, but how much less)? Do e-cigarettes act as a "gateway" to smoking, whereby young people who might not have taken up smoking start e-cigarettes and later switch to smoking? Do some young people who might have taken up smoking never actually smoke but simply take up e-cigarettes instead? Those are important questions, and it is not inconceivable that some of the research that helps to answer them will be funded by the tobacco industry. If that research is of good quality, it would be a shame if the BMJ refused to published it simply on ad hominem grounds.
In his "Critique of Judgement", Kant argued that it is important to think for oneself. A blanket policy of rejecting tobacco industry research removes the need for BMJ editors to think for themselves, and to rely instead on what Kant called prejudice. Prejudice does not seem to me to be a good way to make decisions about scientific publishing.
Competing interests: No competing interests
I was the editor of the BMJ when we decided after much discussion that we would publish research funded by tobacco companies, but I fully respect the decision of the current editors of BMJ, BMJ Open, Heart, and Thorax to decide not to do so.
I do, however, have what I think is difficult question for the editors, and I’d appreciate it if each of them could answer.
The two arguments for stopping publishing research funded by the tobacco industry are that the research is corrupted and that the companies are publishing research in journals primarily to advance their commercial aims oblivious of the harm they do.
I suggest that exactly the same is true of the pharmaceutical industry and that we probably have even more evidence on the misconduct of pharmaceutical companies than of tobacco companies. Both Ben Goldacre and Peter Gøtzsche have gathered together this evidence in important books. (1 2)
So will the editors stop publishing research funded by the pharmaceutical industry, and if not why not? Knowing the heavy financial dependence of journals on the pharmaceutical industry, I shall be looking for sophistry in the explanations.
1 Goldacre B. Bad Pharma. London: Fourth Estate, 2012. ISBN 978-0-00-735074-2
2 Gøtzsche P. Deadly Medicines and Organised Crime: How Big Pharma has Corrupted Healthcare. Oxford: Radcliffe 2013. ISBN 9781846198847.
Competing interests: Competing interest: RS was the editor of the BMJ and chief executive of the BMJ Publishing Group, in which role he was responsible for Thorax and Heart. He was a member of the board of the Public Library of Science from 2004 to 2011. He works for a for profit company, UnitedHealth Group, and is the chair of the board of a for profit company, Patients Know Best, in which he has equity. He receives a pension from the BMA, the owners of the BMJ Group.
So if, hypothetically, a trial partially funded by the tobacco industry showed that e-cigarettes were harmful to health you would refuse to publish it? Not a likely prospect, but not inconceivable. What then?
Competing interests: No competing interests