Ethical Issues in Medical Journal Editing

From Ethics in Academia, ed by SK Majumdar, HS Pitkow, LP Bird, et al, pp 208-220. © 2000 The Pennsylvania Academy of Science. Reprinted by permission.

Introduction

The Editor of a medical journal serves many masters. First and foremost is the readership. When viewed in this light, the responsibility of the Editor is enormous. He (I will use the pronoun “He” throughout this chapter. This is in no way meant as a slight to the many female journal editors. It will, however, save the necessity for the clumsy application of expressions such as “he/she,” and I will be reflecting on my personal experiences.) is the final arbiter of what is published, thereby deciding upon to what new scientific knowledge the audience will be exposed. These decisions can have an enormous impact on everything from the future direction of an entire profession to implications for individual patient care.

The Editor is also responsible to the authors. He must assure that they receive a fair, unbiased and timely review free of potential conflicts of interest and prejudice. Furthermore, since the continuous flow of manuscripts is the lifeblood of any journal, the Editor needs to cajole established authors, respect junior authors and encourage future authors, all the while maintaining editorial credibility. Where this may not be as important to a major publication receiving far more manuscripts than they can ever hope to publish, it becomes vital to journals in smaller specialties or where there is intense competition for premier manuscripts.

The Editor must efficiently manage the Editorial Advisors (or equivalent) and the peer review process. Members of advisory boards and expert peer reviewers are almost exclusively volunteers. Other than a line on a curriculum vitae, a mention in a yearly acknowledgment page and professional satisfaction, they receive little for the tremendous contributions of time and effort expended on behalf of the journal. The Editor needs to respect the fact that these integral members of the journal team have a life and responsibilities outside of the publication. They should not be burdened with excessive numbers of papers to review, papers outside of their specialty or an unrealistic time constraint in returning their reviews. Advisory Board meetings should be called only when there is important business to discuss. They need to be held at a time convenient for the majority of members, realizing that there will never be complete agreement on time, place or agenda.

To complement the Editor in day-to-day business activities, most journals hire a professional staff of managing, associate and assistant editors, not to mention secretaries and the occasional free-lance medical copy editor. The Editor may or may not have direct line-order authority over this staff. Regardless, it is the responsibility of the Editor to interface with this staff to assure the smooth operation of the publication. Manuscript progress should be tracked properly to assure that no bottleneck exists, unnecessarily delaying a paper. Delinquent reviewers and authors need to be contacted with the appropriate “tone” applied to the communication. Inquiries from all interested parties need to be handled properly. Deadlines for publication must be met, and even the most assiduous Editor needs occasional prompting.

Finally, the Editor must be responsible to the organizational management that has hired him. Although he must demand and expect complete editorial freedom as to the scientific content of the publication, the Editor is responsible for following all organizational policies, procedures and guidelines. Although it may be anathema to an Editor that considers himself a “scientist,” the Editor needs to be somewhat politically savvy. Medical and scientific organizations tend to be political circuses. There are both hired management staff, such as executive directors, and elected officials and boards. Organizational presidencies frequently change yearly. Board members with varying agendas are also frequently voted in and out of office by the membership. In an ideal world, in order to maintain editorial purity the position of the Editor should not be affected by these yearly changes. Unfortunately, real life may interfere.

Given the above responsibilities faced by the Editor, it is not surprising that in the daily grind of publishing a monthly, peer-reviewed, indexed medical journal the editorial staff is confronted with situations that must be resolved quickly, decisively, and ethically while trying to assuage the multitude of involved egos. It is the purpose of this chapter to examine some of the more common dilemmas faced by journal Editors. Where applicable I will insert personal experiences or policies of the Journal of the American Podiatric Medical Association (JAPMA), the largest circulation refereed, indexed journal in the profession of podiatric medicine and the publication of which I have been the Editor for over seven years.

Topics to be Covered

• Authorship issues
  • Criteria for authorship
  • Order of authorship
  • Ghost authorship

• Conflicts of interest
  • Author, reviewer, Editor
  • Sponsored supplements/special issues

• Plagiarism (misappropriation)
• Editorial freedom and the role of the Editor—Organizational politics

Authorship

Criteria for Authorship

The question of who should receive credit, and take responsibility, for a published manuscript is probably the single most problematic issue that is faced by a medical editor. The number of authors on a single manuscript, the order in which they are listed and the significance of that order are all of concern.

The International Committee of Medical Journal Editors (ICMJE), in their “Uniform Requirements for Manuscripts Submitted to Biomedical Journals” [1], begins the section on authorship by stating that “Each author should have participated sufficiently in the work to take public responsibility for the content.” Knowing full well that this general statement, although certainly sounding impressive, was not going to be concise enough, they go on to become much more specific. In order for an author to take credit, “substantial contributions” must have been made to (quoting directly):

• Conception and design, or analysis and interpretation of data. AND
• Drafting the article or revising it critically for important intellectual content. AND
• Final approval of the version to be published.

All three of these conditions must be met to justify authorship. Of course, the most vague aspect of these requirements is the definition of “substantial contributions.” This key phrase is not defined in the ICMJE publication. In order to clarify this definition, the American Medical Association (AMA), in their Manual of Style [2], defines this term as follows:

A substantial contribution is an important intellectual contribution, without which the work could not have been completed or the manuscript could not have been written and submitted for publication.

The ICMJE is very clear on what does NOT constitute grounds for authorship and the authority of the Editor in these matters [1].

Participation solely in the acquisition of funding or the collection of data does not justify authorship. General supervision of the research group is not sufficient for authorship. Editors may ask authors to describe what each contributed; this information may be published.

Despite the best efforts of the ICMJE and AMA since they first published their guidelines in 1985, the number of authors attached to scientific papers has actually been rising [3]. Along with this increase in the number of authors per paper has been a concern that the level of contribution of each author has decreased. Years after the guidelines first appeared, Shapiro in 1994 [4] surveyed 184 first authors from 200 papers with four or more authors published in 10 leading biomedical journals. Although he did not directly use the ICMJE guidelines, his questionnaire asked for similar participation. He concluded that authorship conveyed little about the contributions made by any other than the first author. (Of course, since the first author was the one queried, it may be difficult to even draw a conclusion about those individuals’ self-reported contribution!) Shapiro further notes that authors of clinical papers tended to make even fewer contributions than did the authors of basic science papers. Also, second and last authors appeared to have generally contributed more than other middle authors.

Directly related to these issues is the question of total number of authors permitted on any given paper. How many authors are too many? Multiple journal pages have been utilized to list authors on some large multi-centered trials. Various attempts have been made to limit the number through individual journal guidelines or by limiting the value of the citation. The ICMJE and the National Library of Medicine have agreed on a policy of only listing the first 24 and the last 1 author in the citations found on MEDLINE. The AMA journals have taken this decision to the next logical step by only allowing 25 authors in any given reference citation [2]. At JAPMA, a past managing editor, concerned about the “author inflation” she was finding, arbitrarily and absolutely limited the number of authors listed on any paper at 6. Although we have loosened our policy somewhat, to this day any paper submitted with more than the prescribed number of authors must be accompanied by justification of the contributions for any author above the sixth. Although the use of these “sworn statements” has been advocated by many journals, with varying degrees of success, there seems to be no stemming of this tide.

One reason for purposely limiting the number of authors that can be listed is to prevent the naming of “honorary authors.” At our journal this began to reach epidemic proportions. On short clinical case reports it was not uncommon for the paper to be written by a resident and contain the name of the intern or sub-intern that helped with the library research, the attending under whose service the case was seen, the residency director, and even the chief of service. I remember one instance where a paper was submitted to JAPMA with a noted clinician and residency director, well known to me, listed as the senior author. Upon reading the paper I was quite disappointed at the obvious poor quality. A few weeks later I ran into the author at a scientific meeting. I felt that I needed to pull him aside and discuss my concerns with him. After hearing me out he responded, “Oh, I never saw that paper, the residents just put my name on it and sent it in.” As an Editor I was left with very mixed emotions. On the one hand, I was relieved that my estimation of this physician’s abilities was intact. However, I was most disappointed that he would have let an academic breach like this occur in the first place. I explained this to him, received assurances that he would take an active role in the revision of the manuscript and was rewarded with a significantly improved, publishable piece of work.

These issues will prove difficult if not impossible to resolve. While Editors continually try to limit the number of authors and to confirm each individual’s participation, authors have incentives to continually place their names on manuscripts. For those authors in academia the axiom of “publish or perish” continually haunts their thoughts. Publication means a better chance of receiving grant monies, academic promotion, tenure and ultimate security. For those already having reached that vaunted goal, their name is often placed on the manuscript by underlings anxious to appease their mentor or to assure eventual publication by impressed editors. The listing of “honorary authors” is discouraged, as it “devalues the meaning of authorship” [2]. The academic ego also plays a major role in determining authorship. Often, those in positions of authority will dictate that their name is included in the list of authors. There is a saying, I apologize that the source from which I first heard it is long forgotten, that goes something like this: An Assistant Professor reads an entire article. An Associate Professor only takes time to read the abstract. A full Professor just checks the reference list for his name.

Order of Authorship

The question as to the order of authorship and who should be rewarded the exalted position of “first author” has never satisfactorily been answered. Even the ICMJE has not taken a stand on this thorny issue and instead takes the neutral position that “The order of authorship should be a joint decision of the co-authors” [1]. Numerous methods have been proposed, including everything from alphabetizing to mathematical formulae. In 1990, the AMA Editorial Board and staff published their guidelines for determining order of authorship, which are abstracted here. They are quick to point out that even this is not the result of a consensus [5].

• It is first essential to decide who is an author and who is not.
• Being a laboratory or departmental sponsor and “last” author are not mutually exclusive but should depend on contribution to the work being reported. The rewarding of honorary authorship is intellectually dishonest.
• The first author is that person who contributed most to the work including writing the manuscript. The sequence of author listing is determined by the relative contributions to the work.
• Decisions about authors and the order should be made as early as possible.
• Disagreements about these matters should be resolved by the principles, not by the editor. Editors may request justification.

The importance of proper positioning cannot be underestimated. The AMA journals relegate any author above the sixth position to the ubiquitous “et al” category [2]). Landmark studies are often generally known by the name of the first author [5]. The rank of first author may be used in appointment and promotion decisions.

At one academic institution in my profession the requirements for promotion to the rank of full professor clearly stated that the applicant needed to fulfill a publishing requirement of x number of papers as “first author.” This physician, having published many papers with residents and students, frequently put his name in the last position, claiming “senior authorship” while allowing the more junior authors to be listed first. The dean overturned the recommendation for his promotion, as put forth by the promotions committee of the institution. The dean reasoned that the promotion was not deserved since the requirements for “first authorship” had not been satisfied. The denied physician then contacted me to write a letter to the dean clarifying the position of JAPMA on this issue. Since our journal has wrestled with this issue, as have so many others, I wrote that we did not have an official policy on the order of authorship. Furthermore, it is not uncommon for a faculty member supervising students and residents, despite playing a major role in the conception and writing of the paper, to place their name last, allowing the other authors to embellish their curriculum vitae for future employment or residency and fellowship placement. I cannot vouch for the outcome of this communication. Two lessons can be learned from this experience. First, it is necessary for any journal to prepare a written policy in this regard, and second, sometimes generosity in the academic community can backfire.

Ghost Authorship

The opposite of listing too many authors is the failure to list authors on the byline by design. This practice, known as “ghostwriting,” is most common on papers written about commercial products such as drugs and devices. Frequently, in order to get a favorable review or study published quickly, a company will have a professional medical writer actually write the paper. A well-known, authoritative physician is then contracted to put his name on the finished manuscript as the author. The name of the actual ghostwriter never appears on the byline or acknowledgments. The listed physician may or may not have had any input into the content of the paper. If the listing of an excessive number of names could be considered intellectually dishonest, then the practice of ghostwriting may be even more dishonest.

The problems this raises for the journal editor are obvious. Frequently, the editorial staff has no way of knowing that an article is ghostwritten or sponsored. They must assume that the listed author is the responsible party. The Editor contacts the named author with editorial comments and problems with the manuscript. The author disavows any knowledge of the content and refuses to accept responsibility. The blame is put on the ghostwriter, who is not even listed. With whom does the Editor work?

Fortunately, an experienced Editor can frequently spot a ghostwritten manuscript before it goes into the review process. Papers that smack of overt commercialism are immediately suspect. Any review or study that discusses a particularly “hot” or exciting new drug or device should receive extra scrutiny. This is especially true if the product is painted in a positive light. In a relatively small specialty, knowledge of colleagues’ abilities may be useful. As an example: I received a paper sent to JAPMA that was “authored” by a private practitioner with no particular history of scientific research or writing. The paper was a well-written, well-researched, in-depth review of a wound-healing agent that included some impressive study data. In my opinion, there was no way that this physician would have been capable of writing this manuscript. On confronting him with my concerns, he admitted that the paper had been ghostwritten and that he had been paid to submit it for publication. As an interesting endnote to this story, a few months later, while attending a conference, I saw this practitioner in the exhibit hall, behind the booth of the manufacturer of the product, acting as a salesman.

Ghostwritten papers are not necessarily all bad. There are many well-written, well-balanced, acceptable papers. The ethical issue that arises is that of disclosure and acknowledgment of the true authors. Flanagin and Rennie [6] have spelled out the AMA policy on ghostwriting. In order to assure that all parties are disclosed, they have added the following two statements to the Journal of the American Medical Association (JAMA) author statements:

All persons who meet JAMA criteria for authorship are listed as authors.

All persons who have substantial contributions to the work reported in the manuscript (including writing and editing assistance), but are not authors, are named in the acknowledgment.

Conflicts of Interest

Conflict of interest for a given manuscript exists when a participant in the peer review and publication process—author, reviewer, editor—has ties to activities that could inappropriately influence his or her judgment, regardless of whether judgment is, in fact, affected. [7]

With this rather broad definition, the ICMJE in 1993 began its statement on conflict of interest in peer review and medical publication. On first glance this definition seems sound. Who could argue with the need to disclose any factors that might influence the validity of the publication process? But even a subject as superficially evident as this raises profound ethical questions. Rothman [8], in his insightful commentary “Conflict of Interest, the New McCarthyism in Science,” argues that

Although intended to raise scientific discourse to a higher standard of ethics, these new disclosure policies themselves contribute new ethical problems by abridging the right of honest scientists to an impartial hearing of their work.

• He goes on to equate some journals’ requirements for disclosure as a form of censorship.

Whereas policies at JAMA and The New England Journal of Medicine emphasize financial conflicts, Science asks authors to divulge “any relationship that they believe could be construed as causing a conflict.…”

• Rothman interprets this to mean that some journals may actually require authors to disclose religious or sexual orientations (i.e. a homosexual studying homosexuality).

How much disclosure is necessary and what should an Editor do with this information? From whom should an Editor ask disclosure? The most obvious answer to the latter question is that, most importantly, the author should be asked to disclose potential conflicts of interest. At JAPMA we do not ask for the broad range of information requested by Science or recommended by the ICMJE. Our copyright transfer form has the single statement:

The author(s) has no financial interest in the drugs, devices, or procedures described herein. In addition the author(s) has no financial relationship that might lead to a conflict of interest.

• It is my opinion, echoing that of Rothman, that limiting this disclosure to financial interest only is a reasonable compromise. This gives the reader the ability to judge the work in light of knowing there may be an incentive to the writing. This information should not be used to prejudge the work as being flawed, although even this potential supposition of bias is controversial. Again quoting Rothman [8],

Nevertheless, the label of conflict of interest is so commonly used with the intent to discredit a person or work that it is disingenuous for anyone to claim that no accusation is intended when describing conflicts of interest.

Even reviewing the JAPMA form in my preparation of this chapter raises another concern. The above statement has been written with a definite negative connotation. We are specifically asking authors to disclose that they DO NOT have a relationship. By so doing we have fallen into the trap of using this information to discredit a work before it is even judged. The form should simply ask for disclosure. Each individual manuscript should be judged by the Editor, reviewers and the audience on its own merits.

Besides asking for disclosure from the author, the Editor need also determine whether or not there is potential conflict of interest with the reviewer. Although this may also be financial, such as when the reviewer is a consultant for a particular company, there may also be less well defined academic conflicts. Peer reviewers possessing expert knowledge in a field are frequently familiar with others working in the same field. Although a review may be blinded in both directions, a savvy reviewer may be able to determine the identity of the author by subject matter or writing style. Professional jealousy may lead to discounting a manuscript that is of value. Likewise, personal or professional friendships or academic and research collaboration can lead to an inappropriately positive review. For these reasons, some journals, such as JAMA, request that peer reviewers also sign a conflict of interest statement [2]. JAPMA does not request that reviewers sign such a statement, since I try to make the determination in the reviewer selection process. Neither method is without fault.

A less often considered area of potential conflict of interest is with the Editor himself. As with what I assume is the case with many other biomedical editors of journals in smaller specialties, I do not have the luxury of being a full-time Editor. Along with my parttime editorial responsibilities I maintain a full-time academic appointment, clinical, teaching and hospital responsibilities, along with being a consultant and speakers’ bureau member to various companies. As the final arbiter of what is and is not published in JAPMA, it is important to keep all of these responsibilities and potential conflicts in mind. This variety of activities raises some important issues. I am, of course, biased in the opinion that the above collection of job descriptions improves my ability to edit a medical journal. Since I see patients on a daily basis, I am anxious to read about new findings and ideas. My teaching of residents and students is improved since I can honestly say that I must be “up on the literature.” My personal ability to write and publish is enhanced by reading so many well-written, and poorly written, manuscripts. Finally, being involved with the pharmaceutical industry gives me insight into how they work in the funding of research and publications. The flip side of this coin is that all of these activities could lead to a litany of potential conflicts. My academic position may cause me to favor authors from my institution. Being a clinician, I may be set in my ways and not open to new information that may be contrary to what I have seen in practice as successful. Finally, and most obviously, I may be biased toward manuscripts favorable to a commercial product.

As a matter of course, I am contractually bound to scrupulously avoid any potential conflict. Specifically, I am not supposed to make the final decision on any manuscript generated from my institution. Anecdotally, I had understood that there might have been a problem with this situation concerning a past Editor. This caveat in my contract is a mixed blessing. There is no question that the ability to ward off questions from colleagues of mine at work about why their manuscript was rejected on review has saved me from some very uncomfortable situations. A dismissive “Sorry, I had nothing to do with the decision” can be a powerful defense. It is also a bit of a copout. If my charge as Editor is to determine the scientific direction of the journal, how can I deny responsibility for what could amount to a significant number of manuscripts? Ideally, an Editor needs to be able to judge any and all papers submitted to a journal in an unbiased manner. It should be left up to the discretion of the individual Editor to determine if this is feasible in a given situation.

Sponsored Issues

Another form of potential conflict of interest is the publication of supplements or special issues. Both serve the same purpose, to provide concentrated information on a single topic. The difference is that a supplement is an extra issue that is separate from the actual journal whereas a special issue is a regular issue of the journal turned over to a single topic. The concern is that these special publications are frequently underwritten by a grant from a corporation. Of course, there are pros and cons to these special issues. On the positive side, they can quickly disseminate useful clinical information in a concentrated format. Readers do not have to search through various different journals over a number of years to find material about the subject. Furthermore, they can be a financial blessing for a journal working under a relatively tight budget. Usually, the funding received for the publication more than exceeds the cost of the actual editing, printing and mailing. On the flip side of the coin, these issues are quite frequently biased toward the product of the sponsoring company and do not undergo the same rigorous peer review of a nonsolicited issue. Bero et al [9] studied 625 published supplements and found that 42% had a single pharmaceutical sponsor. The sponsored publications were more likely to have misleading titles and use brand names and less likely to be peer reviewed than other articles in the parent journal. They concluded that the financial relationships among participants, sponsors and the journals should be clearly identified and that the journal editors needed to maintain control over the content.

The conclusions of Bero et al were echoed by the ICMJE in their recommendations on the use of supplements [10]. I have abstracted a number of their most germane recommendations here.

• The journal editor must take full responsibility for policies, practices and content of supplements. The journal editor must approve the appointment of the supplement editor and retain authority to reject papers.
• The sources of funding should be clearly stated and prominently located in the supplement, preferably on each page. Whenever possible funding should come from more than one sponsor.
• Editing by the funding organization should not be permitted.
• Journal editors and supplement editors should not accept personal favors or excessive compensation from sponsors.

Plagiarism

It is not the purpose of this chapter to define plagiarism or discuss other aspects of scientific and research misconduct. Rather, this brief discussion will focus on the role of the journal editor in the detection and handling of cases of plagiarism.

The greatest difficulty in dealing with this issue is identifying potential plagiarism while the manuscript is still early in the review process. Needless to say, it is much easier to deal with a charge of plagiarism before the paper is published. Unfortunately, this early detection mostly occurs by happenstance. It is impossible for an individual Editor to be familiar enough with all of the published literature that the detection is made on the first reading by the Editor. More likely, the expert peer reviewer, by definition an individual with more complete knowledge of the subject, will be the level at which the plagiarism is identified. On a number of occasions I have sent a paper to a reviewer only to be contacted by that reviewer and informed of word-for-word lifting of another’s thoughts inserted into the manuscript. Frequently, the reviewer recognizes his own words in the manuscript in question. At this point in the process the matter is relatively easy, albeit sensitive, to handle. The author is sent a copy of the written concerns of the reviewer and asked to either explain or correct the situation. If the author complies in a manner satisfactory to the reviewer and the Editor, the process of peer review continues. The matter is kept private between the Editor, the reviewer and the author.

At JAPMA this type of readily detected blatant plagiarism seems to be most evident in papers written by students and residents. This problem could be minimized if institutions of medical education would educate their students as to what does and does not constitute plagiarism. Although it could be argued that students should have received this information long before entering medical school or residency training, it may be idealistic to believe that this material is stressed at the undergraduate level. Another way to minimize this type of plagiarism would be to have the senior author more directly involved with the actual writing of the paper. Of course, if this were as easily said as done, the entire above section on authorship would have been unnecessary!

Once a paper is published, an allegation of plagiarism takes on a much more critical nature. The Editor should review the charge in depth before making any decisions. The ICMJE states that the Editor should be responsible for contacting the person against whom the allegation is made but not responsible for conducting an investigation [2]. It is up to the individual author’s institution to proceed with any formal investigation and disciplinary action. Despite this opinion, I will conduct a preliminary investigation by requesting that the accuser provide hard evidence of the alleged plagiarism. This could be as simple as an appropriate reference to the plagiarized material, to a line-by-line recitation of the copied literature. Once I have this information in hand, I will contact the author and request a response to the charge. The response will then be considered, and if the charge is valid, a proper retraction published. There was one recent incident at JAPMA where the author from whom the material was lifted identified the situation and asked for an investigation. The author committing the plagiarism admitted his mistake and wrote a formal apology letter to the first author. At the request of the plagiarized author, no published retraction was deemed necessary, and we honored that wish.

Editorial Freedom

Finally, in any discussion on ethical issues in medical journal publication, it is important to include a short discussion on the importance of editorial freedom. This issue became most topical the very evening that this is being written. George Lundberg, MD, the Editor-in-Chief of JAMA, has been fired after 17 years for deciding to publish a “politically incorrect” study on the definition of sex. How the internationally esteemed editor of one of the most prestigious medical journals in the world can be fired on the basis of one decision should be of concern to anyone involved in medical publishing. Although I do not have all of the facts, there may be extenuating circumstances of which I am not aware, and I have not seen the manuscript in question as I write this, the firing appears to be an issue of poor timing and political sensitivity.

Ironically, the AMA’s own Manual of Style presents two full pages dealing with the issues of editorial freedom [2]. To quote directly from that publication:

In scientific publishing, editors must have complete authority for determining the editorial content of their publications. Editors must be free from restraint or interference from the publication’s owner, publisher, advertisers, sponsors, subscribers, authors, editorial board or publication committee members, reviewers, and readers. Owners, publishers, boards, and publication committees may have the right to select and dismiss the editor, but they should not interfere with editorial decisions and policies.

As with many of the issues discussed in this chapter, the ICMJE echoes the AMA in their beliefs [1]. In order to assure this editorial freedom, the ICMJE has published the following recommendations.

• The editor should have direct access to the highest level of ownership, not only to a designated manager.
• Editors of medical journals should have a contract that clearly states the editor’s rights and duties.
• An independent editorial advisory board may be useful in helping the editor establish and maintain editorial policy.
• All editors and editors’ organizations have the obligation to support the concept of editorial freedom and to draw major transgressions of such freedom to the attention of the international medical community.

It will be enlightening over the next number of months to see how the JAMA situation develops and whether or not there will be any uproar in the editorial community. The Editor of a medical journal must be able to exercise complete freedom in selection of manuscripts for publication. It is certainly within the right of the owner of the publication to disagree with those decisions and to replace the editor, but, as with any academic dispute, due process should be required to avoid arbitrary and capricious decisions.