Ethical publishing in Orthopaedics

As surgeons we rely in part on textbooks and learned articles to inform our practice, to keep up to date and to ensure we are able to deliver the best care to our patients. We also undertake research, review scientific articles and present our findings at meetings and in print. Underlying all we do should be the principles of honesty and transparency. This article considers these principles in publishing. While authors may be the most important in these issues, editors, peer reviewers and journal proprietors play essential roles. The discussion which follows is based in part on the paper ‘Ethical Standards for Orthopaedic Surgeons’ produced with co-authors for EFORT.¹

In his provocative book The Trouble with Medical Journals,² R. Smith noted that medical journals are a force for good, but need considerable reform. He asked, amongst other things, why we should bother with them at all, especially if their honesty is questionable. He suggested that peer review is a flawed process at the heart of science and journals, and raised concerns about research misconduct. He drew attention to publishing too much with too little content or relevance. He recognised that money clouds objectivity, and that editors were not immune to misconduct and their freedom should be closely linked with accountability. There have been many attempts to address the issues raised by Smith, and it is to be hoped that our criteria for research, article assessment, editorial responsibility and publication will continue steadily to improve.

While most scientists conduct their research properly and report their results honestly and without bias, there are lamentable lapses from these high standards. There have been examples of data invention, manipulation or omission which have allowed fraudulent conclusions to be drawn. Another issue of concern is failure to obtain both ethical study approval and fully disclosed informed consent. Only too often articles are quoted and referenced which have not been read and from which incorrect deductions are made. Authors should only use citations that are relevant to their projects. The addition of references irrelevant to the study should be discouraged as they are often simply inserted to increase citation rates. Authors, typically senior members of a department, are often listed who have played no significant part in the research, while others who have contributed substantially are not credited. There are in addition matters of breached confidentiality, plagiarism and so-called ‘salami-slicing’ (where a paper is divided into several sections for publication although all relate to the same topic) and dual publication.

When the words or ideas of another researcher are used without acknowledgement, this counts as intellectual theft or plagiarism. While electronic searching and systematic reviews have made this more readily identifiable, the ‘cut and paste’ so widely practised make the risk of inadvertent failure to quote sources all too easy. We should remember that we may use the phraseology of others provided we properly reference the original article. In any published review article a careful study of other work is an essential pre-requisite. As this article attests, it is only too easy in writing a review to re-use previously published personal work as well.

An editorial in Nature³ drew attention to ethical misconduct in scientific research and publication. It noted that, although deplorable, action was often not taken: “This is odd, as simple steps can help put right plagiarism.” At its simplest, editors may simply attach a note to the online version of the paper, marking it as a plagiarism and directing readers to the original work. An erratum may be included in the next issue of the journal. The paper may be withdrawn from the literature. At its worst, plagiarism is a serious offence, especially when an author falsely claims that an idea is original. The issue may be made more difficult in large institutions where ghost writers may report the study findings in a standardised format. The consequences of plagiarism have been widely reported in the academic world. Allegations of plagiarism may ruin a reputation and academic career as publishing rights are withdrawn. There may also be legal repercussions as the originator may sue in the event of a lost reputation or earnings.

While we all have a responsibility to report on suspected fraud and plagiarism, we should, however, try to distinguish between honest error and deliberate falsification and recognise that there may be differences in the analysis and interpretation of data. Not all error is deliberate fraud.

Authors should not submit a manuscript to more than one journal at the same time. Similar publication may be acceptable later if the original article is cited and the circulation predicted is different. Authors and editors should be aware that duplicate submission may include the submission of similar articles in different languages to different journals. If after discussion, however, the parallel publications are accepted, appropriate recognition should be made in the second journal.

Editors act as gatekeepers to ensure the validity of published material and to deliver unbiased accurate reports to journal readers, as stressed by JBJS Emeritus Editor Henry R. Cowell.⁴ The Committee on Publication Ethics (COPE) Code of Conduct for Journal Editors⁵ provides a set of minimum standards to which all COPE members are expected to adhere. It outlines the responsibilities that editors face and adds aspirational best practice guidelines for wide-ranging, increasingly complex ethical issues. In general, editors make the final decision about publishing; when they receive papers to consider for publication, COPE advises that four key questions should be asked:

• How important is the study?

• How reliable is the study?

• Have the authors paid attention to patient protection?

• What do the regulations state in the country where the study was done?

There may be problems with this process: once a paper is submitted, bias may become evident. Editors might select as peer reviewers those known to favour a particular view. Many editors have other jobs, and the sheer weight of submitted material may make in-depth analysis of papers impossible. National conventions and regulations concerning, for example, animal experimentation, vary widely as do the standards for patient recruitment and investigation. These issues must be balanced against the value of the information and ideally should be addressed in the advice given to peer reviewers. We should not forget that falsified information may lead directly to patient harm: our treatments may be changed wrongly as a consequence of fraudulent data. On the other hand, valuable information may be lost if our criteria become too stringent. The International Committee of Medical Journal Editors (ICMJE) also provides useful recommendations for the conduct, reporting, editing, and publication of scholarly work in medical journals.⁶

Peer reviewers should bring confidentiality and impartiality to their opinions. If peer review is to remain a cornerstone of high-quality scientific publication, this must be a rigorous process: reviewers should be charged with careful data analysis and be aware of the risks of fraud. They should be alert to the possibilities of duplicate publication and plagiarism. They must preserve confidentiality before publication and critically must not allow professional or personal rivalry to influence their advice to editors. All peer reviewers should analyse and report promptly to avoid delay in publication. There have sadly been examples of reviewers who have rejected a scientific paper and then used this as a basis for their own studies.

Publishers, too, have a role to play in ethical publishing: they may be asked to evaluate complaints against editors or editorial boards and take appropriate action. However, they should not interfere with editorial freedom unless there is clear evidence of unethical editorial abuse of the privilege of independence.

It is difficult to be sure how much published data is scientifically unjustified and how often articles are cited and reproduced erroneously. It is to be hoped that newer analysis techniques will continue to reduce fraudulent publication and ensure that we may believe more of what we read in our scientific journals.