Intended for healthcare professionals

CCBYNC Open access
Editorials

Open science prevents mindless science

BMJ 2018; 363 doi: https://doi.org/10.1136/bmj.k4309 (Published 15 October 2018) Cite this as: BMJ 2018;363:k4309

Research Integrity, Open Science, and Health Policy

Munafò, Hollands, and Marteau point to the serious damage caused by academic misconduct to the trust in science. The authors call for ‘open science’ based on a fundamental cultural shift in the science ecosystem where research workflows are made transparent and subject to greater scrutiny from the early stages of protocols and planning through publication and data sharing. As they indicate, this is increasingly the case in clinical trials while more and more being considered for all sciences.

The need for research integrity becomes more acute in a world strongly influenced by science and increasingly reliant on scientific innovation to address challenges in all sectors across society, not just health. The collecting and reporting of scientific evidence is, as the authors point out, not only a matter for scientists among scientists, but even more importantly for how that evidence is used to create and shape public policies in health and, indeed, in all areas of our societies today.

Science is only possible as an investment. The return on that investment is provided, not only in knowledge, technologies, and products, but also in public policies (governmental, legal, financial) that affect how we structure our communities, our interactions with one another, and even our collective goals.

The retraction of scientific articles based on fraudulent representations of science is damaging to science. It is additionally damaging to the scientists, their institutions, and the publishers. Trust is lost and embarrassment ensues, with the accompanying professional and financial repercussions. But the greatest potential damage is most likely that to a society reliant on scientific evidence to inform important policy making. Thus, ‘open science’ is often resisted by its very gatekeepers, especially when fraud is alleged, due to their own financial interests or concern with embarrassment [1]. This can even lead to whistle blowers being made suspect and, at times, more endangered than those who should be investigated [2].

We not only need to examine the failures on the part of those submitting erroneous or fraudulent science, but also the failures on the part of those publishing science that is not supported by clear evidence or is fraudulently presented. The International Committee of Medical Journal Editors’ ‘Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals’ shows clearly how medical journal editors have avoided taking their own proper responsibility for addressing weak science, misconduct, and fraud in scientific publications: authors are assigned ‘responsibilities’; journal editors are assigned ‘freedoms’ [3], even considering that many of these editors are employees of a for-profit industry whose interest is protecting its own investment and reputation.

The fact that journal editors could write ‘Recommendations’ giving responsibilities to others (authors), but no responsibilities to themselves and their own profession shows the weakness of the publication system in maintaining its own quality and integrity. Similarly, the Committee on Publication Ethics (COPE) also holds authors accountable but provides little guidance on what the responsibility of editors is when erroneous science or fraudulent research conduct is demonstrated [4].

As Munafò, Hollands, and Marteau indicate, health policy may be directly affected by the evidence scientists produce. Where the evidence is lacking, corrupted, or misrepresented, serious threats may arise to health policy and, thus, public health. Science is never only for the sake of science. Science is always for the sake of good societies. Scientists should not lose sight of the fact that it is public policy that makes their science possible, while also bearing in mind their responsibility to ensure their reported outcomes are of value and benefit to public policy.

Achieving ‘open science’ requires more than just the development of technologies that can manage the exponentially increasing number of scientific publications and the huge data sets that we increasingly expect to be shared. ‘Open science’ will also need to assign clear responsibilities to authors, editors, reviewers, and institutions. However, achieving real ‘open science’ will require fundamental changes in science education, in how we structure and reward research, and ultimately in the creation of a symbiosis in science and in society between ethics and learning, ethics and knowledge, ethics and policy.

[1] Williams, P., and Wager, E. (2013). Exploring why and how journal editors retract articles: findings from a qualitative study. Science and engineering ethics, 19(1), 1-11. https://doi.org/10.1007/s11948-011-9292-0
[2] Hawkes, Nigel. ‘Academics who raised concerns about research misconduct are threatened with lawsuit’. BMJ 2018; 362 doi: https://doi.org/10.1136/bmj.k3100 (Published 12 July 2018). Cite this as: BMJ 2018;362:k3100.
[3] International Committee of Medical Journal Editors (ICMJE). ‘Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals.’ December, 2017. http://www.icmje.org/recommendations/
[4] Committee on Publication Ethics (COPE). ‘What constitutes authorship? COPE discussion document.’ 9 June 2014. https://publicationethics.org/files/u7141/Authorship_DiscussionDocument_...

Competing interests: Francis P. Crawley has been a member and past vice-chairman of the Ethics & Standards Committee of the International Society of Medical Publication Professionals (ISMPP). He has taught good research conduct at all levels of the university and outside the university for more than 25 years.

06 November 2018
Francis P. Crawley
Executive Director
Good Clinical Practice Alliance - Europe (GCPA) & Strategic Initiative for Developing Capacity in Ethical Review (SIDCER)
Brussels/Leuven, Belgium