Big Tobacco and data nabbing: when freedom of information reduces transparency

A tobacco corporation is attempting to access confidential data on teenagers' smoking habits, obtained by university researchers. Just one case where Freedom of Information benefits companies against the public interest.

How could researchers possibly be expected to breach confidentiality commitments by passing sensitive data concerning children to a corporation that profits from selling a lethal and addictive product? Yet this is the predicament of researchers at the University of Stirling, who have been told that a tobacco corporation has the right to request their confidential data on teenagers’ smoking preferences.

The world’s largest transnational tobacco corporation, Philip Morris, is attempting to use freedom of information legislation to access data obtained from teenagers on the basis that it would remain confidential within the research team. The data are interviews and survey responses from teenagers concerning their smoking habits and perspectives on plain packs (a tobacco control measure that the Department of Health is currently considering to prevent children from being attracted to the glitzy designs on current packaging). In June this year, the Scottish Information Commissioner upheld Philip Morris’ claim that they, at the very least, had a legitimate right to request this information.

The University of Stirling will now have to provide a formal response. Whilst the University seems unlikely to hand over the data immediately, ultimately, they may be legally ordered to do so.  If this happens, the researchers will have to decide whether to abide by the law or by their own professional and ethical standards. The dilemma raises big questions around the politics of transparency and data access, and questions the assumption that freedom of information legislation will necessarily improve the democratic health of the UK.

It is not hard to imagine why Philip Morris wants the data.  In a Philip Morris employee’s own words: ‘Today’s teenager is tomorrow’s potential customer, and the overwhelming majority of smokers first begin to smoke while still in their teens… it is important to know as much as possible about teenage smoking patterns and attitudes’.  Whilst these comments were made back in 1981, the underlying point stands: tobacco company business depends on recruiting new smokers to replace those who die (and as cigarettes kill around half of all long-term users, this is a high turnover business). 

As the declining credibility of the tobacco industry in the UK makes it difficult for tobacco companies to undertake their own research with teenagers, the University of Stirling’s data potentially offers Philip Morris some tantalisingly elusive insights into teenagers’ smoking preferences.  The data would also allow tobacco company researchers to construct counter-arguments designed to undermine academics’ conclusions that plain packs would help reduce smoking amongst young people.  It should be remembered that Phillip Morris is a company that (along with its competitors) deliberately sought to cast doubt over the link between smoking and lung cancer, even when it was well aware that the link was unequivocal. 

As the University of Stirling and Philip Morris battle out this case, broader questions need to be asked about potential conflicts between ethical research commitments and freedom of information legislation.  Does the latter really trump the former? Are all researchers who accept public funding now under an obligation to share the resulting data if a request is made, no matter whom from?  If so, what are the implications for research ethics and for the assurances researchers give to potential participants? Academics in the UK could now find themselves in a similar quandary to journalists who are legally obliged to reveal sources who they have committed to protect. It is worth asking whether this kind of transparency will improve democratic accountability or whether it might, in fact, be damaging.

Writing for The Guardian’s ‘Comment is Free’ section and drawing on the University of Stirling case, Heather Brooke claims that  ‘if data is used by businesses to make money, everyone wins’. If data are available to us all, Brookes argues, ‘we can see for ourselves if any attempt is made to "spin" it.’  This implies both that the interpretation of data is an objective process and that it is relatively simple (something everyone can do). Yet, if this were true, there would be little call for researchers to do anything more than collate and report data.  In reality, interpretation and analysis are crucial parts of their jobs and are, necessarily, somewhat subjective. 

This is precisely why, in direct contradiction to Brooke’s claim that academics are ‘scared of revealing their methodologies’, most peer-reviewed articles require researchers to describe the chosen methodology and process of analysis and to support claims with specific references to the data.   Few people argue that universities should remain in their ‘ivory towers’, but reservation about handing over data provided on a confidential basis is very different from reservation about adequately explaining methodological and analytical processes. Whilst the latter should always be explained, if tobacco companies gain access to publicly funded data, there is every chance that the data will be manipulated, and there is no guarantee that they will make their own analysis transparent. 

This is crucial because, as tobacco companies know only too well, ‘what is truth to one is false to another’.  Indeed, the seemingly endless ingenuity of the tobacco industry’s underhand activities has enabled the sector to prevent a range of tobacco control measures in various contexts, despite their declining public and political credibility. Most recently, evidence has emerged that tobacco companies, including Philip Morris, have actively promoted legislation relating to data access and requirements for policy to be based on ‘sound science’, because they believe this will help them prevent or delay public health legislation. It is perhaps unsurprising to learn, then, that in the USA, businesses are one of the most common users of Freedom of Information requests.

So do transparency laws, including freedom of information legislation, necessarily improve the democratic health of Britain? Academics and journalists have certainly used Freedom of Information requests to aid accountability. But the current situation, in which private companies have access to publicly funded data but are under no equivalent obligation to share their own data, seems a long way from democratic. The inherent power imbalances in this legislation must be dealt with, or we may increasingly see research conducted with taxpayers’ money manipulated and turned against the public interest.