By Professor Julian Kinderlerer, CONSULTATION EVENT CO-CHAIR, former President of the European Group on Ethics in Science and New Technologies, European Commission, reporting to Presidents Barroso and Junker and the European Parliament; Prof. of Intellectual Law, Cape Town University; Adviser to the South African Science Ministry; Occasional Adviser to the UN Environment Programme; UN Industrial Development Organisation; World Intellectual Property Organisation; Former Director of Institutes on Biotech Law, Ethics & Society at Delft and Sheffield Universities.
By Dr Marina Murphy, Head of Scientific Media Relations, Research & Development, British American Tobacco
Scientific integrity is the cornerstone of all research, ensuring that the results are based on research that is repeatable, traceable and open for critique and comment by the scientific community. The worst examples of scientific misconduct are plagiarism, double publishing of results and falsifying of data. The choice of data that supports a particular interpretation poses a major threat to scientific integrity as well. There have been numerous examples over the last decade across both industry and academia of these practices. In most cases the peer review system has successfully rooted out the small (though often potentially extremely high impact cases) number of misconduct cases. Peer-review requires careful and honest assessment of the quality of the research and assurance that the data is not over-interpreted or that research that negates that suggested is not over-looked. This discussion has extended beyond the peer-reviewed literature to scientific conferences, the mass media, government advisory bodies, and more. While there is no doubt that there can be many possible sources of bias, the debate so far has focused almost exclusively on the concern about links between industry and science and how the demands of the former can undermine the public-interest mission of the latter.
But there is no getting away from the crucial role industry plays and will continue to play in the research process. Our society in general expects innovation to be tied to commercialism, and so much of the science supporting innovation, from green chemistry to new drugs will happen in or be commissioned by companies not universities. All of us depend on industry to put research results into practice; to feed us, clothe us, get us around, and provide medicines when we are ill. Industry scientists outnumber academic scientists and industry funds more research than any government or organisation. Science students are more likely to end up working for an industrial organisation than a government or university. Almost all industrial research is for positive ends, and the development of research ideas into products by industry is the cornerstone of our society. That is not to say that mistakes have not been made in the past. But we can learn from these mistakes and work together to engender a culture of transparency and collaboration; to fulfill our duty of care and to ensure that society benefits from the immeasurable good possible through scientific and technical innovation and expertise. It is incredibly important that society trusts those who do research and those who use the research for our (and their) benefit. This aim requires an approach to openness and transparency which industry has not displayed in the past, primarily due to a fear of losing a competitive advantage. The farming out of research effort to multiple agents who are then not permitted to evaluate the full details of the results of the combined efforts has exacerbated the perception that too little of the interpretation involves the scientists actually performing the research.
Industry and Integrity
A considerable and growing body of literature has evolved in recent times on the subject of conflicts of interest and the possibility that such conflicts of interest can influence the integrity of scientific research and subsequently adversely influence the scientific record. It is often taken as a given that conflicts of interest are a disqualifying factor in doing research and having that research published. A common view is that scientists with conflicts of interest are at least partially compromised when it comes to integrity and even when these conflicts are completely and fully disclosed, the scientists and the science are viewed to some degree at least to be of suspect credibility.
However, editorials such as that on the ‘Integrity of Science [Barr D.B (2007). Integrity of Science, Editorial, J of Exposure Science and Environmental Epidemiology, 17, 123] have openly reinforced that the existence of conflict does not translate into poor science and furthermore should be included in acknowledgements giving full transparency over the author’s potential perspectives.
Conflicts of interest are often cited as a reason not to trust any industry science, particularly in the areas of food and health. There is no denying that there have been problems. Open any newspaper to find reports questionable industry behaviour. Recent reports saw Coca-Cola’s top scientist, leaving the company over disagreements about research funded to counter claims that fizzy drinks can cause obesity1; car manufacturer Volkswagen is understood to have fitted around 11 million of its cars with software to dupe emissions tests2 and Exxon Mobil is said to be under investigation to determine to what extent it played down the threat of global warming3.
Reports like these are not new. In the 1990s, the pharmaceutical sector was rocked by a number of revelations. Drug giant Pfizer was accused of not following protocol when it used children in Nigeria to test the experimental antibiotic Trovan; eleven of the children died and others were disabled4. In 2004, Merck had to recall its painkiller Vioxx after a study showed that the drug more than doubled the risk of heart attacks, strokes and death. Merck was reported to have known about the dangers for years and was accused of fabricating and cherry-picking study results to mislead doctors and patients about the drug’s safety5. In September 2014, GSK was found guilty of systematically bribing doctors and hospitals in China to promote its products6.
Probably the most infamous and controversial industry of all is the tobacco industry for its historic lack of clarity over the addictive nature and massive public health impact of cigarette smoking, and its lack of progress in developing products that were truly less harmful than the traditional cigarette, cigar or pipe tobacco.
The tobacco industry now clearly accepts that cigarette smoking is addictive and extremely dangerous and is developing products such as e-cigarettes that could be substantially less harmful. In the pharmaceutical sector, for every story of potential misdoing there will be hundreds of products that help millions of people. The controversial cases should not overwhelm the good work done by the vast majority of scientists or prevent the efficient transfer of the inventions and innovations that will improve our lives.
Conflict of interest and bias
Regrettably, much of the literature and indeed much of the debate mixes ups bias and conflict - if all conflicts were banished forever, there would still be multiple sources of bias. But conflicts of interest are not, in themselves, determinants of bias.
Even the existence of multiple conflicts, in and of itself, carries with it no certainty of bias. A ‘conflict of interest’ is a situation in which a person has a private or personal interest that appears to influence the objective exercise of his or his duties, e.g. scientific research. A ‘bias’, on the other hand, is the tendency to make errors in judgement as a result of filtering information through one’s own likes and dislikes, one’s own personal opinion. For the purposes of scientific research, a conflict of interest may refer to a situation where the funder has a financial stake in the research and hopes for a particular research result that might benefit them. The result is not necessarily biased, depending on whether there is a deviation between the research results and what is perceived to be the true state of affairs.
You could argue that of course industry scientists are conflicted – of course they want the research to yield positive results, better products, advances in technology etc. But it is not necessarily biased. Conflicts of interest will not go away, they will always exist. But as long as they can be managed so that the risks are minimised and the benefits maximised, all should be well.
We need more transparency not less
Some scientific publishing houses, albeit a small number, have reacted to the ingrained belief that all industry science is biased by banning for example, all tobacco-industry-funded science from being published. But there have also been calls to ban publication of research funded by pharmaceutical companies. Richard Smith, former editor of the British Medical Journal, claimed that pharmaceutical companies and tobacco companies were equally untrustworthy and cannot be relied upon to provide research that isn't influenced by commercial aims, but argued that it was better to publish such research than ban it. (BMJ has not published research undertaken by tobacco companies since 2013).
We need to be able to trust the science and research that comes out of industry, and the regulators that oversee it. Doesn’t this mean that we need to foster a culture of collaboration, cooperation and greater—not less— transparency?
As Henk Jan Out, Professor of Pharmaceutical Medicine at Radboud University Medical Centre in The Netherlands and Global Clinical Head for Women's Health R&D at TEVA Pharmaceuticals, points out, pharma companies sponsor the majority of drug trials, and most are conducted after extensive consultation with regulators. Medical journals, he says, should publish all drug research, irrespective of the funding source, but be very critical on the research question, the methodology and the interpretation in the discussion7.
There are a number of organisations working to try to create the kind of transparent and collaborative culture that would help ensure that crucial scientific advances and evidence are not ignored. The American Chemical Society (ACS), for example, recently set about laying out guidelines for best practice in engaging with industry scientists and best use of scientific knowledge in informing good policy decision-making. Among its recommendations is that government agencies use unbiased science in a transparent way; encourage open scientific discourse; declare conflicts of interest; use all relevant peer-reviewed sources; document and, if possible, reconcile conflicting results; publish and present research or explain why not; and transparently select scientific review panellists.
Corporations and regulators alike would do well to apply these principles of scientific integrity to their everyday conduct, encouraging teamwork and frank dialogue.
Resources are also available for ethical investors looking to screen company behaviour, such as the Sustainability Accounting Standards Board, an independent non-profit aimed at developing and disseminating sustainability accounting standards; the FTSE 4 Good Index, designed to measure the performance of companies demonstrating strong environmental, social and governance practices; and the Ethisphere Institute’s list of the world’s most ethical companies, based on corporate responsibility, the culture of ethics at the firm, corporate governance and oversight, transparency, and legal track record.
Research and innovation—whether or not conducted or funded by industry—have the potential to immeasurably improve people’s lives. It is vitally important that everyone involved in industry science is candid and accountable for their assertions, recognising their duty of care to consumers. Those in industry must show that they conduct research to the highest standard, demonstrate integrity and best practice, avoid bias and cherry picking data, and openly and transparently engage with external scientists and policy makers. But the same standards should also apply to interest groups, the media, and regulators.
Regulators in particular have a difficult job. They have to consider, not just the science, but also economic, legal and political factors and of course, public opinion when making decisions. Of course decisions made now will impact not just the people of today, but generations to come. Take climate change, for example. Politicians argue over details, while some use fierce rhetoric to drive home the reality of a future that is inevitable if we don’t reduce greenhouse gas emissions. But few are unaware of the impact the decisions made today will have on the world of tomorrow. At a recent summit in France (December 2015), US president Barak Obama called climate change an escalating crisis that ‘will condemn our children to a planet beyond their capacity to repair.’
Another ‘billion-people’ issue is that of smoking-related death and disease and the potential positive public health impact the advent of e-cigarettes might have. The World Health Organisation says that in this century, about one billion people will die from smoking-related disease, compared to one hundred million in the last century. Many believe that the popularity of e-cigarettes and the fact that they are substantially reduced risk compared to cigarettes could be the answer to this public health crisis. Public Health England, an executive body of the UK Department of Health, recently published a report saying that e-cigarettes are 95% safer than cigarettes. But the WHO has failed to support e-cigarettes and some regulators have responded by banning them.
Co-author of this note, Julian Kinderlerer, then President of the European Group on Ethics in Science and New Technologies (EGE), told a delegation at the Annual Meeting of the American Association for the Advancement of Science in February 2015 that: ‘In the case of cigarettes and e-cigarettes, history will judge us harshly as to how we answer this billion-person question. It may also look back in anger at policy-making amounting to institutionalized manslaughter’.8
One regulator that recognises the need for interaction and collaboration with external partners including the regulated industry is the US Food and Drug Administration. The FDA favours an interactive culture with ‘successful engagement of other scientific communities in academia, industry, and government’ 9. The FDA’s strategic plan for regulatory science says that ‘FDA scientists must work in an environment that fosters creative thinking, promotes scientific multidisciplinary interaction and collaboration, offers opportunities for learning, and ensures both scientific excellence and integrity…’
Advances in science and technology over the last millennia have realized benefits in every facet of mankind’s existence, from shelter and food, to health and mobility. No scientific bodies have driven this progress more so than industrial scientists and their collaborators in research institutions as they have applied the fundamental science discoveries to deliver products. It is imperative that industry science continues to be seen as a flagship for scientific integrity and to this end all science deserves the opportunity to be commented upon and critiqued objectively by the scientific community through continued engagement and effective and honest peer review.
References in this article
1) http://well.blogs.nytimes.com/2015/11/24/cokes-chief-scientist-who-orchestrated-obesity-research-is-leaving/?_r=0
2) http://www.bloomberg.com/news/articles/2015-09-22/vw-s-u-s-chief-apologizes-for-diesel-cheating-we-screwed-up-
3) http://www.nytimes.com/2015/11/06/science/exxon-mobil-under-investigation-in-new-york-over-climate-statements.html
4) http://www.bbc.co.uk/news/world-africa-14493277
5) http://www.nytimes.com/2011/11/23/business/merck-agrees-to-pay-950-million-in-vioxx-case.html?_r=0
6) http://www.bbc.co.uk/news/business-29274822
7) http://www.bmj.com/content/348/bmj.g171/rapid-responses
8) http://www.science20.com/news_articles/ban_them_or_embrace_them_ecigarettes_face_a_fork_in_the_road-153217
9) http://www.fda.gov/ScienceResearch/SpecialTopics/RegulatoryScience/ucm268152.htm
