To ban or not to ban (industry-funded publication): That should not be the question

In this article Blair Hesp explores industry-sponsored publications and question whether it is time to re-think the culture surrounding scientific publications.

A former editor of the British Medical Journal recently suggested that papers relating to pharmaceutical industry-funded science should not be published in the journal, questioning the validity and accuracy of pharmaceutical company-funded research, amongst other oft-repeated allegations of scientific misconduct that are levelled at the pharmaceutical industry in general.1 Leaving the validity of this reasoning behind, would it in the best interests of science for journals to implement such policies, how would it be done and are there other alternatives?

Why only ban papers that are linked to pharmaceutical industry funding?

Unfortunately, you do not have to travel far to uncover unethical publication practices in scientific publications. There is an obvious incentive for any scientific industry, whether it be the pharmaceutical, medical device or nutritional supplement industry, to ensure that their research is published, as it helps build credibility for their product and can support marketing initiatives. However, peer-reviewed publication is also the lifeblood of academic careers, creating the same incentives and temptations to abuse the peer-review publication system when publishing non-industry-funded research. Therefore, it is naïve, at best, to lay the blame for unethical publication practices, including honorary authorship and undeclared conflicts of interest, solely at the doorstep of the pharmaceutical industry and to exclude a subset of authors from publishing on the basis of their funding source alone, not the quality and relevance of their science.

How do you define industry-funded research?

Furthermore, much academic research is funded by, or otherwise associated with, industry. Therefore, making decisions on what constitutes "industry-funded research" is a potential minefield. Key questions could include: Who will make this decision? What does and does not constitute a pharmaceutical company? Does this include industry-funded research performed by an independent institution at arm's length? Likewise, many academic institutions publish research to validate concepts that will ultimately be commercialised via spin-out companies, so attempting to define this concept has all the hallmarks of a very slippery slope. Accordingly, the logical conclusion would be that the resources required to enforce a ban on industry-related publications could be better used to improve the peer-review and editorial selection process.

Is post-acceptance review the answer?

The patent system offers an interesting case study to compare and contrast with the publication process used by scientific journals. Patents are centrally examined for novelty, industrial applicability and inventiveness prior to acceptance. Much like the peer-review process for scientific journals, some are rejected, but most are eventually accepted once pertinent amendments are incorporated. However, a key principle of accepting a patent application is that it is done so with a presumption of validity. Accordingly, the option remains open to correct errors and oversights within accepted patents that were not identified during the examination process. Notably, a request to amend a patent may result from the patent owner becoming aware of a potential issue and wishing to correct it or as the result of a challenge from an independent, but interested, party.

Concepts such as PubMed Commons and Publons are now making crowd-sourced post-publication review a reality, so the question then becomes how will journals respond? In the pre-internet era, publications containing errors and oversights that slipped past the peer-review goalie were effectively set in printed stone, with both the cost and effectiveness of printing errata in later journal issues being of questionable value.

However, given the modern reliance on electronic publications and searchable journal databases, it is not unreasonable to expect more errata to be published given that both the logistical and expense barriers associated with print-only journals have been removed. Likewise, errata can be easily appended to electronic articles and effectively communicated via updated database entries. This could arguably have a much greater impact on the quality and accuracy of published research compared with other proposed actions as it would bring a level of ongoing accountability and responsibility for both authors and publishers that has not previously been seen in scientific publications.

The industry publication ban / transparency contradiction

The idea that experiments should be replicated, theories tested and ideas debated and modified is a key principle of science, and constitutes one of the reasons behind the high-profile push for all clinical trial data to be made publically available for independent analysis and interpretation. Of course, a frequent response to these calls is that some data is submitted for publication, but never published following (presumably multiple) rejections from journals, particularly in instances where neutral or negative outcomes are observed. While increased transparency and refusing to publish industry-funded research do not have to be mutually exclusive concepts, it would be difficult to reconcile how a publication ban, and the potential absence of peer review, supports initiatives to improve transparency for clinical trial data.

Is it time to re-think the culture surrounding scientific publications?

Science is a democracy where all are welcome and free to submit their views, theories and research data, wherein objective self-correction and continuing, incremental improvements are encouraged. Within this environment, editorial teams and peer reviewers may consider themselves to be the gatekeepers for the scientific establishment, whereas in reality they act as a valve – publications may pass, but once they have passed they will not return. A narrower valve, for example one that blocks industry-funded publications, will not improve the flow of information. Likewise, pressure building on the peer-review valve from post-acceptance review has the potential to lead to its failure if the current publication system does not adapt to meet modern expectations.

Reference

1. Smith R, Gøtsche GC, Groves T. Should journals stop publishing research funded by the drug industry? BMJ 2014;348:g171.

About the author:

Blair Hesp is the Managing Director of Kainic Medical Communications Ltd., a New Zealand-based agency that specialises in providing on-demand professional medical writing support to overseas medical communications agencies and New Zealand-based biotechnology companies. Blair has a PhD in Pharmacology from the University of Otago and a New Zealand Diploma in Business, and has been recognised as one of New Zealand's most innovative marketing and communications professionals. In addition to several years' experience working in the Global and European medical communications industry in the United Kingdom, he has also spent several years working in the international intellectual property industry.

Closing thought: How do you define industry-funded research?