Editorial I

Ethics in publishing; are we practising to the highest possible standards?

J.M. Hunter

For new editors of scientific journals, the ethical issues involved in publishing are the most significant of the many challenges which they will face. A new world opens as they become very aware, yet again, that university did not prepare them for this venture. How should they manage such problems as dual publication, redundant publication and plagiarism? How should they define publishing misconduct?

Three years ago, to help with these problems, the Editor-in-Chief of the British Medical Journal (BMJ), Richard Smith, invited the editors of several specialist journals associated with this publishing group (PSP) and the editor of the Lancet to found a new committee, the Committee on Publication Ethics (COPE). This is a voluntary body. It aims to act as a forum for discussion and to give advice to scientific editors on practical methods of dealing with difficult issues related to publishing. It also aims to improve standards in publishing practice. It meets regularly to consider particular problems (suitably anonymized) and to advise if appropriate. Editors are invited to submit any difficult problems to the committee in writing for discussion, and hopefully this results in constructive advice.

As the new Editor-in-Chief of the British Journal of Anaesthesia (BJA), I was invited to join this voluntary body, and found it a useful educational exercise. The group, which is chaired by Professor M. Farthing, the Editor of Gut, recognized the need to produce guidelines for setting standards in the ethical issues related to publishing. After a widely advertised and detailed discussion meeting attended by over 70, mostly European, editors of scientific journals in April 1999, an advisory document was released, which is available in print1 and through the World Wide Web (http://www.bmjpg.com/cope/contents/gpp.htm).

This document makes useful reading not only for editors but also for all potential authors interested in publishing their data. It discusses study design and ethical approval; data analysis; authorship; conflicts of interest (which the editors of the BJA are at present reviewing); peer review; redundant publication; plagiarism; the duties of editors; media relations; and advertising. It is also deals with publishing and editorial misconduct.

Guidelines on good publication practice are not new. The pharmaceutical industry has produced them for research in volunteers,2 3 as has the Medical Research Council on the policy and procedure for inquiring into allegations of scientific misconduct.4 But the COPE guidelines bring together all the various aspects of research ethics in a unique way. Their recommendations relate to the following eleven issues.

 1. Study design. All research must have the approval of the local ethics committee or an alternative body, and fully informed consent must be sought. The BJA has insisted on this requirement for many years, even for the retrospective analysis of patient records and for audit studies. Statistical issues should be considered early in the design of a study, including power calculations, to ensure that an appropriate number of patients are being studied. For animal experiments, full compliance with national legislation (the 1986 Scientific Procedures Act in the UK) is essential.

 2. Data analysis. Fabrication or falsification of data is, at all times, considered to be misconduct. Methods of analysis of whatever type must be given in sufficient detail to be reproducible. Post hoc analysis of subgroups which become apparent during a study is only acceptable if it is clearly disclosed. Any issues of bias developing from the methodology (for instance, a higher preponderance of female patients in a study group) must be mentioned in the discussion.

 3. Authorship. This has proved difficult to define and has been a topic of much discussion in the BMJ.5 All authors must be willing to take public responsibility for the contents of their paper. An intellectual contribution to the conception, design and writing up of the study must be balanced against the collection of data, such as analysis of plasma samples, and routine work such as the completion of patient questionnaires.

 4. Conflicts of interest. The BJA is devoting more attention to the detection of this problem. At present, authors are asked to inform the Editor-in-Chief of any conflict of interest when receipt of a manuscript is first acknowledged. The Guidelines to Authors published in each issue of the journal and the Extended Guidelines to Authors6 also stress the importance of such honesty. But this is inadequate, and more detailed questioning is being planned, in line with other international peer-reviewed anaesthetic journals. At present, most of the admissions of conflict I receive are of little consequence and thus are easy to admit. Do the real offenders ever confess them voluntarily?

     Assessors also need to be asked in more detail to declare conflicts of interest. Conflicts may be personal (I do not like the guy); commercial (I work for the pharmaceutical industry); political (I am against the policy of a particular political party); academic (I despise these authors from another academic institution); or financial (I have been paid to write this paper). Financial interests also include travel expenses or payment of lecture fees. The BJA is intending to improve its attention to such details. However, such reporting can be taken to inane extremes, and thus become a subject of ridicule. This is apparent in reading some of the weekly medical press in Great Britain, and I wish to avoid such silliness. Nevertheless, I appreciate that, on occasions, editors themselves should declare conflicts of interest to their readers and if necessary should withdraw from the review process in relation to relevant manuscripts. This is not difficult with a journal such as the BJA, which is run by a team of editors on a daily basis, and it is practised regularly.

 5. Peer review. The quality of a scientific journal is directly related to the quality of its peer review. I am fully aware that this is a limiting factor in the value of the manuscripts published in the BJA. Assessors must not be overburdened (and I admit to making this error on occasion), nor must they be asked to do that of which they are not capable. Assessors should be constructive in their approach, and not insulting to authors. They should aim to improve the quality of the manuscript they are reviewing and maintain confidentiality at all times. I would hope that in the near future an increasing number of BJA assessors would agree to sign their reports, as occurs in the review process for some other peer-reviewed anaesthetic journals. A few already do so. I believe, however, that this act should be voluntary.

     The COPE guidelines recommend that journals publish accurate descriptions of the peer review, selection and appeals process. The BJA editors must consider these recommendations in more detail. Regular audit of acceptance rates and publication times is also recommended. At present, these are detailed annually by the Editor-in-Chief to the BJA Editorial Board. The acceptance-to-publication interval of this journal has been 4–5 months for many years. The rejection rate is over 57% and rising with the increasing number of manuscripts submitted. If the acceptance rate were allowed to rise, each issue of the journal would increase in size and thus cost more to print and post. Hence, readers would have to be willing to pay the increased costs.

     It is entirely appropriate for authors to recommend the names of suitable assessors for their manuscripts. The editors of the BJA will often take up these suggestions, especially if they are appropriate, but will also freely use other assessors of their own choice.

 6. Redundant publication. The publication of two or more papers from a single study without cross-reference, from the same or different departments, is very difficult to detect. It is often found only by chance and expert assessment is the major route of detection. All editors of scientific journals work constantly to detect and eradicate this problem. Full reporting of such instances in a prominent section of the journal (e.g. immediately after the editorials) is encouraged by all editors. Republication of a paper in another language, especially in the Third World, may be considered acceptable but only with prior editorial approval and with full acknowledgement at the beginning of the translated text. Authors should always disclose details of their related studies on first submission of a manuscript, including those which are in press.

    Having considered many of these details, the editors of the BJA have recently updated the Extended Guidelines to Authors.6

 7. Plagiarism. This may occur at any stage of the research process: during the planning of a study; during data collection; on writing the manuscript; or after publication. It applies to printed or electronic reports of an investigation. Plagiarism relates to the unreferenced use of published or unpublished work belonging to other workers, including refereed grant applications. It is also relevant to work published in another language; indeed, this may be the most difficult to detect. Again, competent assessors are often responsible for detecting such fraud and editors rely on them to report it. Detection is often purely due to chance. Plagiarism is a daily concern for editors of international anaesthetic journals and is deplored by them.7 If it is necessary to publish large amounts of any other research worker’s material, permission should be sought both from author and publisher.

 8. Editorial responsibilities. In the COPE Report, editors are noted as being stewards of their journal. This is an onerous but highly privileged position. Working under constant pressure, all editors would admit that they occasionally make incorrect decisions on whether or not to publish a manuscript. Decisions on publication should be taken only in relation to a paper’s scientific contribution, originality and clarity. Particular consideration must be given to manuscripts which challenge accepted concepts; negative findings may also have significant importance. The paper should also be in the scope of the journal, a very relevant factor for the BJA. All manuscripts are considered confidential, and any attempt to break this confidentiality should be wholly resisted by editors. Indeed, I have had to exercise such a prerogative on a few occasions. Ultimately, editors must take responsibility for correcting any flaws which are subsequently realized in a published manuscript. Corrections must be recorded prominently and as rapidly as possible.

 9. Media relations. Although the COPE guidelines provide detailed guidance to editors on this matter, the problem is perhaps not as great for editors of anaesthetic journals as for those of weekly general medical journals. For instance, the BJA does not produce regular press releases. It must be remembered, however, that journalists may attend national meetings at which early reports of ongoing research are presented and accounts may be published without permission or clarification. Organizers of such meetings should advise speakers if they expect journalists to attend.

10. Advertising. This used to be a significant source of revenue for all scientific journals but in recent years it has tended to decline, especially in specialist journals. Some income is also gained from large reprint orders. It is essential that editors are not influenced in their publishing decisions by advertising revenue or reprint potential. COPE strongly advises that advertising and editorial administration are kept separate, and this has always been BJA policy. But it has been routine for the Editor-in-Chief to approve all advertisements before they are published.

11. Misconduct. The COPE Report provides guidelines for editorial teams on dealing with misconduct. Misconduct is defined as ‘intention to cause others to regard as true that which is not true’. Deception may be intentional, or due to recklessness or negligence. Good practice at all times requires honesty and full disclosure. Editors have a responsibility always to investigate any form of misconduct. The response to such problems can at times be difficult to decide and COPE can be of particular use in this field.

     Editors do not have the legal legitimacy to conduct a thorough investigation into serious cases of misconduct. They must decide whether it is appropriate to alert the employers of an author to the misconduct, and evidence will be required. The editor must at the same time inform the authors of his/her actions. If there is no obvious employer and the authors are members of a profession, the appropriate professional body (e.g. the General Medical Council) has to be informed. An appropriate notice may also need to be inserted in the journal, especially in relation to published material.7 The authors must always be given the opportunity to respond to such allegations. If the authors are proved guilty of wrongdoing, then editors are advised by COPE to consider taking sanctions against them. For instance, COPE advises that an editor should refuse to consider any further manuscripts for publication from the authors or their department(s) for a stated time period. For lesser offences, sanctions may be limited to a warning or may involve a formal letter to the relevant funding institution or employer detailing the problem and the action taken or proposed. An editorial is also an appropriate method of detailing any misconduct.7 In serious cases, the paper should be withdrawn from the scientific literature by informing other editors and indexing authorities, and by placing an announcement regarding the misconduct in a prominent position in the journal.

Ethical standards in publishing continue to rise. Many of these guidelines would have been considered excessive even 10 years ago, not because they had no practical significance but because many medical researchers did not consider such activity as dual publication to be unacceptable. But attitudes have changed significantly, and for the better, as much more thought and debate have been concentrated on these issues. As Editor-in-Chief of the BJA, I am very aware that we must at all times strive to maintain high standards in anaesthesia. An advisory yet voluntary body such as COPE is a most useful reference group for any harassed, overburdened editor who aims, at times fruitlessly, to attain such high ideals.

J. M. Hunter

University Department of Anaesthesia

University Clinical Department

Duncan Building

Daulby Street

Liverpool L69 3GA

UK

References

1 Committee on Publication Ethics. Guidelines on Good Publication Practice. The COPE Report 1999. London: BMJ Publishing Group, 1999

2 Association of the British Pharmaceutical Industry. Facilities for non-patient volunteer studies. London: ABPI, 1989

3 Association of the British Pharmaceutical Industry. Guidelines for medical experiments in non-patient human volunteers. London: ABPI, 1990

4 Medical Research Council. Policy and procedure for inquiring into allegations of scientific misconduct. London: MRC, 1997

5 Horton R, Smith R. Time to redefine authorship [editorial]. Br Med J 1996; 312: 723–4[Free Full Text]

6 Extended guide to contributors. Br J Anaesth 2000; 84: 131–7

7 Todd M. Plagiarism. Editorial view. Anesthesiology 1998; 89: 1307–8[Medline]





This Article
Full Text (PDF)
E-Letters: Submit a response to the article
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (1)
Disclaimer
Request Permissions
Google Scholar
Articles by Hunter, J.M.
PubMed
PubMed Citation
Articles by Hunter, J.M.