A process that permits tobacco company involvement

The initiative is reported to have originated with a communications consultancy called ‘Sci-Com’ (http://www.sci-com.eu/main/index.php/about-us/who-we-are) which is ‘dedicated to delivering solutions to communicating the most challenging science’. From the outset, published documentation on the Sci-Com website has given prominence to ‘harm reduction’ and addiction to tobacco, alcohol and drugs.3 We can find no declarations of interest in any of its publications or reports.

The Declaration states that it was developed following a series of consultations with more than 300 stakeholders from 2012 onwards.2 One hundred and sixty-five are named in the report, 62 of whom, including the organisers, were present at the concluding event in 2016, and another 103 who attended earlier consultation events or contributed text.1 There is little information available on how participants were selected, or who was involved, apart from those named. The Declaration lists those participating, including distinguished public health scientists, science journalists, senior science advisors and other officials in governments, current and former members of parliaments and government ministers, and representatives of industry. This is not surprising; there are many fora where these groups come together to discuss science policy.

What is surprising, however, is the substantial presence of representatives from the tobacco industry and, especially, British American Tobacco (BAT). Indeed, with participation by its Group Scientific Director, Chief Scientific Officer, Head of Biosciences at Group R & D, International Scientific Affairs Manager, as well as the Chief Medical Officer, Head of R & D and International Public and Scientific Affairs Director of its wholly owned Nicoventures,1 BAT’s representation exceeds that of any other company. There are also two representatives each from the American Chamber of Commerce to the European Union and the European Smokeless Tobacco Council (BAT is a member of both, as are other tobacco companies), and from Swedish Match1 (which is also a member of the latter).

This was very different from the first meeting in 2012, from which the tobacco industry was excluded as a condition for the participation of the European Commission’s health directorate, DG Sanco.3 This was in compliance with Article 5.3 of the Framework Convention on Tobacco Control (FCTC)4 which precludes the tobacco industry and related vested interests from having any input into public health policy. We see no indications that the various named individuals linked to the European Commission were involved after the first meeting.

It is unclear why the situation changed subsequently, with BAT personnel becoming prominently involved, for example, through multiple authorships among 20 ‘thought leader’ essays.1 The content of the Declaration expects rather a lot of scientists and policymakers, and rather less of industry actors; it questions the integrity of scientists, calls on them to be less ‘aloof’ and demands accountability for their use of public funding. It also calls on policymakers to be more accountable. It proposes that they give much greater access to industry, the organisations that represent its interests and ‘the public’, which although unsaid is likely to include many ‘grassroots organisations’ created by industry, sometimes described as ‘astroturf’,5 as well as those who advocate on their behalf on social media, whether real people or ‘bots’.6

In this context, it is worth considering further the ‘we’ whose voices make this declaration. 

Richard Horton, the editor of the Lancet, who attended the first event and was unaware of any subsequent process involving tobacco companies, was quoted, as if as offering an endorsement, despite not having seen, or been aware of, the Declaration.7 This calls into question the process described to produce this statement of principles and recommendations, particularly as it declares that: ‘All too often, who those experts are, how they are chosen and how reliable their advice really is, is open to question.’1 We asked the founder and chief executive officer of Sci-Com by email whether they were; ‘willing to share details of all those who funded the work that led to the Declaration, and also advise whether you have any policy on taking funding from the tobacco industry.’ We received no reply. In these circumstances, a careful analysis of the wording of the Declaration seems justified.

Detailed content analysis

The Declaration advocates an alternative to ‘a continued dangerous slide into the realm of policy-biased evidence,’1 though what this alternative is or might be is unclear. The content is broad in scope, covering a wide range of issues at the science/policy interface. For example, there is much content on science advice, science policy and public health, and importantly in relation to prospects for future influence, the initiative secured the involvement of high-level scientific advisors to public institutions, as well as other senior officials.

The view that ‘vested interests can be beneficial,’1 as long as they are declared has long been advocated by the tobacco industry, notwithstanding evidence of studies being designed to give the intended results, including the creation of doubt,8 and by meta-analyses showing that the funding source can be an important factor in the results obtained.9–12 Scientists are encouraged not to discount ‘citizen science’, despite extensive evidence of how corporations have supported social media campaigns to challenge accepted scientific wisdom13 and used front groups to shift the discourse and attack legitimate researchers.14 The tobacco industry’s long history of distorting scientific evidence means that many leading medical journals will not consider for publication any material resulting from tobacco industry funding.15 16 The challenging issues inevitably raised by any tobacco company involvement in any consideration of the relationships between science and policy are entirely ignored.

The Declaration champions views promulgated by organisations such as the Institute of Economic Affairs,17 an organisation endorsed by some British politicians,18 which is among the lowest scoring on an index of funding transparency19 but which is known to have received tobacco industry funding. It has a long history of opposing health organisations and measures they propose to reduce the harms of tobacco and alcohol.17 It regards public funding of science as a source of bias in policymaking,20 as part of its long-standing hostility to public institutions, deploying this argument to attack what is described in the Declaration as the ‘scientific-political establishment’.1 It seems unconcerned by any risk of bias from funding of research by those with vested commercial interests.

The scale on which tobacco companies are still seeking to undermine tobacco control measures has been recently documented by the WHO,21 and confirmed yet again by media exposés such as the acquisition, by Reuters, of a large repository of industry documents22 and a series of reports in the Guardian newspaper.23 The policies fiercely opposed by tobacco companies are crucial for achieving the Sustainable Development Goals and other global targets to reduce non-communicable diseases.24 Yet the Declaration fails to mention the FCTC or Article 5.3.4 On the contrary, although not identifying the tobacco industry as such, it states that: ‘Industry is not to be shunned when it comes to policy-making’1 and ‘industry should speak out more when denied access to important policy-making.’1

By promoting industry involvement in policymaking without any explicit safeguards for the management of vested interests, the Declaration is bound to raise concerns. These are exacerbated by its treatment of commercial conflicts of interest, which it portrays as straightforward to manage, compared with the non-financial conflicts of interest of scientists: ‘industry is too often perceived as suffering from fatal conflicts of interest and its views are therefore dismissed. In fact, commercial conflicts of interests are fairly easy to deal with if they are properly declared…Ideological, personal or academic conflicts of interest, on the other hand, are much harder to detect or deal with.’1 This ignores both the vast body of evidence about the difficulties in managing commercial conflicts, as well as the obligation on academics to declare any conflicts of interest, while implying, without evidence, that some participants in policy debates from the health arena base their concerns on ‘ideological, personal or academic conflicts’.1

Many other statements in the Brussels Declaration seem entirely reasonable (see box 1). Yet there are also some that, especially if they are taken in isolation, such as those discussed here, give considerable cause for concern.

The nexus of tobacco and alcohol industry strategies?

The content of the Declaration is reminiscent of the promotion of ‘sound science’ and ‘good epidemiology’ practice in previous decades, by tobacco companies such as Phillip Morris.25 In that instance, perfectly reasonable propositions were articulated, alongside the subtle inclusion of content that served key purposes for the company.25 For example, it was proposed that any findings where the relative risk was less than two could be discounted because of the risk of unrecorded confounding, which conveniently eliminated most of the evidence on the harms of passive smoking. This earlier initiative was successful in achieving the support of key scientific organisations such as the International Epidemiological Association, which was unaware of the involvement of Phillip Morris.25 There are also similarities with an approach previously used by BAT in relation to ‘better regulation’; promoted to enable the inclusion of mandatory consultation with all stakeholders affected by policy decision-making, including those otherwise excluded from policymaking.26

It is noteworthy that tobacco industry actors would not have been permitted to be involved overtly in discussions of tobacco-specific scientific or policy issues in line with Article 5.3 of the FCTC.4 This means that broader focused initiatives such as the Brussels Declaration, and the examples discussed above, may be strategically important to tobacco companies. The specific tactics used to advance tobacco industry interests in science and/or policy warrant careful examination; for example, a National Dialogue on Cancer in the USA, was later shown to have been orchestrated by the tobacco industry.27

While we have focused so far on the tobacco industry, there are also seven different alcohol industry organisations named.1 Although there are, for example, pharmaceutical actors involved including four from Johnson and Johnson, no other corporate sector matches the tobacco and alcohol industries in extent of involvement, perhaps because of the chosen prominence of addiction and public health issues. Subtle reformulations of key research and policy ideas are also characteristic of alcohol industry actors,28 though we know far less about their methods than we do about tobacco industry actors, due to lack of access to internal company documents and the underdevelopment of research using other data sources.

The involvement of the Brewers of Europe is, however, noteworthy. In 2006 that organisation convened a group of researchers who concluded that: ‘violence is a subjective term which is fairly nebulous and elastic,’ which could undermine research on alcohol-related violence.29 In 2014 Spirits Europe, another alcohol industry trade association involved, criticised individual scientists, research projects and research agendas that it regarded as biased, proposing an alternative public health research agenda.30 The Portman Group has previously paid large sums to researchers who wrote anonymous critiques of a WHO sponsored alcohol policy evidence review.31 The Scotch Whisky Association led the campaign to delay the Scottish Government’s policy on minimum unit pricing32 despite compelling evidence33 in its favour.

Further, there are similarities with earlier alcohol industry approaches, notably the 1997 Dublin ‘Principles of Cooperation Among the Beverage Alcohol Industry, Governments, Scientific Researchers, and the Public Health Community’ that sought to promote partnerships between the research community and industry actors.34 This initiative was sponsored by the International Centre for Alcohol Policies (ICAP), a body presented as a reliable and authoritative source of public health and policy guidance, but formed and funded by global alcohol producer companies, and in part by Phillip Morris through their ownership for decades of the Miller Brewing Company.35 (The successor organisation to ICAP, the International Alliance for Responsible Drinking now claims the ‘Dublin Principles’ as their own.36) The Dublin statement drew the attention of other tobacco companies, most notably BAT. The most senior BAT figure involved in the Brussels Declaration said of the Dublin Principles in 2000 in an internal company email to addressees including another senior figure named in the Brussels Declaration that: ‘They make interesting reading. Something to aspire to?’35 37 Thus, BAT has a long-standing interest in the Brussels Declaration type of initiative and, despite much rhetorical distancing from tobacco companies by alcohol industry actors, it is clear that they are willing to work closely together on influencing the science/policy interface.38

We identified 20 of 165 names listed in the Declaration as directly representing tobacco or alcohol industry organisations; others also have such associations. For example, a Nicoventures-funded scientific paper on tobacco harm reduction was authored with a company employee involved in the Declaration39 and another individual’s companies have previously consulted for BAT.40

And on to Africa

As smoking rates in traditional markets decline, low-income countries, including those in Africa, are seen as important growth areas for tobacco companies, while increasing disposable incomes make these countries attractive to other global sectors, including alcohol. The Brussels Declaration process was used as the model for a further initiative convened by Sci-Com; in June 2017 participants from nine African countries met in Cape Town with some of the same ‘thought leaders’ from five other countries ‘to put an African-specific process on the rails’.41 This Sci-Com initiative again broadly addressed many aspects of the relationships between science and policy, this time in the context of Africa, and again recruited senior scientific figures. Similarly, there was a prominent focus on ‘harm reduction’ for tobacco and alcohol, and although there was no BAT presence, among those involved were the Head of Communications for sub-Saharan Africa for Phillip Morris International, and a senior alcohol industry trade association representative, as well as political, pharmaceutical company, and governmental figures.41 A 4-year process leading to a further declaration statement on ethics and principles of science and policy is intended. It seeks to address similar issues to those covered in the Brussels Declaration in an African context.41

Issues and implications

A range of issues are raised by the Brussels Declaration. We emphasise five key points. First there are questions to be answered by Sci-Com about the Brussels Declaration process, including how participants were selected, who actually attended the 2013–2015 events, how much involvement the named participants had in the final Declaration and how the costs were met. Similar questions should be asked about the African extension to this project.

Second, there are issues relating to how to protect from vested interests those networked forms of governance, in which consultancies, and other proliferating non-traditional organisations such as think tanks42 are participants. In particular, the exclusion of tobacco companies from involvement in public health policymaking required by Article 5.3 of the FCTC may be undermined by their activities in broader domains of policymaking. Tobacco companies are adept at subtle steering of others’ actions, including putting their own arguments in the voices of others.43 Consequently, there is a well-founded need for monitoring of their activities to incorporate attention to third parties, in order to prevent the subversion of public health policies.21 Consideration should be given to extending the governance model of Article 5.3 to identify those in receipt of any tobacco company funding for any purpose, notwithstanding the complex pathways through which industry funding may be channelled. Although there is no equivalent international treaty for alcohol, similar issues have been raised about the alcohol industry,44 45 particularly as its representatives are extensively involved in alcohol policymaking38 and there is no governance model operating to address these concerns. If evidence is to inform policymaking, there must be full transparency about funding and other conflicts of interest from all involved to protect against distortions of the evidence. Norms concerning the integrity of the process need to be articulated clearly, in the public interest, and there is a clear need to build on what has been learnt from research on the tobacco industry to see how it can be applied to other sectors, increasing knowledge about what has been termed the corporate determinants of health.46

Third, there is a question of how to manage potential direct or indirect interactions with tobacco company personnel and others acting on their behalf. While governmental officials should be cognisant of their duties to avoid tobacco company actors in respect of public health, vigilance is also warranted in relation to science and other areas of public policy. Scientists and policymakers accepting invitations to meetings need to undertake due diligence about the credentials of the organisers and ensure that their names and reputations will not be inappropriately used. It is also important that scientific and health-related organisations take care to avoid other ways in which their names and reputations can later be associated inadvertently with industry-supported initiatives, such as the commercial hire of meeting rooms.47

Fourth, the Brussels Declaration is far from the only recent development in which tobacco and alcohol industry actors have been operating in the research community on science and policy. The recent announcement of a substantial investment by Phillip Morris International in a ‘Foundation for a Smoke-Free World’48 has been widely criticised by leading scientists and health authorities, including in this journal,49 and by the WHO, which noted: ‘that research and advocacy funded by tobacco companies and their front groups cannot be accepted at face value.’50 Such activities require dedicated in-depth studies of whether the claims made are consistent with the activities of the actors involved,51 and the development of new research agendas. From using research funding to bias the evidence base as a principal strategy in earlier decades, to subtle alterations to scientific norms, it now appears that science policy and the institutional governance of research have been identified by tobacco and alcohol industry actors as new targets for their attention in the contexts of developing global political strategies.

The final issue centres on the uses to which the Brussels Declaration may be put, and more broadly, how vested interests may use this and other scientific issues to influence policy.52 53 It is premature to assess the extent to which this Declaration has been used successfully as an instrument of influence. We do already know that it has been used to gain access to key figures in policymaking on public health, science and universities in the Brussels Declaration, and now the subsequent African process. We cannot at this stage assess whether it has influenced policymakers’ views on conflicts of interest or on the evaluation of scientific evidence, or whether there are any discernible impacts on public health or science policies. However, it does seem likely that it can be used in this way by vested interests. It is thus of particular concern that the continent of Africa, which is clearly of major strategic importance to the tobacco and alcohol industries,21 54–56 is now being targeted.