|
Ethics explained
Being on the front lines of scientific publishing, all editors have an obligation to understand issues of publication ethics, to develop appropriate policies for their journals, to be able to explain these to authors, reviewers and other editors, and to ensure implementation of their policies. Faith McLellan, North American Senior Editor for The Lancet, delivered the presentation 
Ethics, Ethics Everywhere: Ethical Issues in Medical and Scientific Publishing at a recent editor’s conference in Edinburgh, Scotland, and she has agreed to deliver another such presentation at the editor’s conference in New Orleans in October 2005. In addition, she has just completed a year as President of the Council of Science Editors. We spoke to her about ethics.
McLellan studied medical humanities and ethics and is co-editor of,
Ethical Issues in Biomedical Publication (Johns Hopkins University Press, 2000) with Anne Hudson Jones so she is well-versed in the issues surrounding ethics in scientific publishing. “These issues are of great concern to editors, as editors are directly responsible for selecting the articles that will be published in their journals,” she explains.
Information for authors
The Lancet, like many journals, provides
information for authors spelling out its policies on issues such as authorship and conflict of interest. All journals need clear policies that clarify their positions on various ethical matters and that can serve also to educate authors and reviewers. “Serious misconduct is probably rare,” says McLellan. “Most ethical lapses, I would say, often stem from ignorance.” The guesswork, slippage, corner cutting, or other minor sin committed out of ignorance may be something that an author does not realize falls within the definition of ethical misconduct.
Conflict of interest
Entanglements in scientific research can easily occur because of competing interests, often financial ones. Where a researcher is funded by a body that will either benefit from or be disadvantaged by an outcome of the research, most would agree there is, at least, the potential for conflict of interest to exist. But McLellan refers to
an example concerning research into passive smoking that shows the effects conflict of interest can have. “Of 106 review articles identified in this study, more than one third concluded that passive smoking was not harmful to health; almost three-quarters of those articles were written by someone affiliated with the tobacco industry,” McLellan explains. “An analysis that controlled for several factors showed that only one thing was associated with that conclusion: an author was affiliated with the tobacco industry.” If authors’ affiliations are fully disclosed as a matter of journal policy, these conflicts of interest can be revealed.
Repetitive publication
Repetitive publishing is another important ethical issue. At the most basic level, publishing the same material more than once may violate copyright law. It also wastes editorial and journal resources.
More disturbing, however, is that the same data sets reported in several places can confound systematic reviews and meta-analyses. The publication of overlapping data sets, when not clearly identified as belonging to the same study, can, for example, skew results that may have important implications for public health.
Differences between disciplines
Fortunately, not all misconduct has the same impact as misconduct that occurs in medical research. “Editors of medical journals are always aware that people’s health could be at stake,” says McLellan. This focuses misconduct issues more acutely in the health sciences than in other disciplines but ethics is fundamental to research whether or not human lives are involved. “Perhaps people are unlikely to die if
physics research is falsified but the blow to the discipline is still terrible.”
Authorship and contributorship
Authorship is one of the most difficult issues in publication ethics. What one must do to qualify for authorship, whose names can go on the byline of a paper, and in what order these names appear are all matters of debate. Further, authorship practices may vary by discipline, by institution, and by individual laboratory or research group. One definition of authorship has been spelled out by the
International Committee of Medical Journal Editors. Other
ways of giving credit and recognizing responsibility, such as contributorship, have been proposed and adopted by some journals.
Reviewer misconduct
Reviewers are another possible source of misconduct. Professional rivalry can result in a reviewer delaying his review for his own competitive advantage, or a reviewer may breach the confidentiality of the review process by sharing the article with colleagues. And reviewers may have conflicts of interest that make them ineligible to review a paper. Journals should, as they do with authors, provide clear instructions to reviewers to minimize such problems.
Editor bias
Authors and reviewers are usually the primary objects of concern when issues of publication ethics are discussed. But editors are not immune from misconduct. “It is important to remember,” says McLellan, “that editors can, and sometimes do, breach ethical standards.” Therefore journal policies should also address ethics for editors.
Publication bias
There is a greater likelihood that
studies with positive results will be submitted to journals (a bias attributed to authors), and that
positive studies will be published (a bias attributed to editors) “It has been shown” explains McLellan, “that authors are less likely to submit trials with negative results, and that journals are more likely to accept positive studies and to publish them more quickly.”
Help for editors
When editors have questions about ethical matters, other editors may be the best source of information and guidance. Likewise, professional organizations for editors often provide a wealth of information, including policy statements, case studies, and sometimes advice and dispute resolution.
Some resources for editors are:
Ethics is meant to help further the highest ideals and best practices of scientific research. Although ethical issues can be quite complex, clear definitions and sound policies can contribute to strengthening and empowering the scientific community, rather than restrict it.
****
Please send responses to this article to EditorsUpdate@elsevier.com
|
