Dr Horton is Editor of The Lancet , London, England.
In 1994, Doug Altman invited the International Committee of Medical
Journal Editors to consider how it might help authors who had complaints about
editors. The concern that editors abuse the power and trust invested in them
by authors, readers, and publishers had grown with the careful documentation
of instances of unambiguous editorial misconduct. One obstacle to their proposal
was the logistic complexity of coordinating, at the international level, an
appeals procedure through a single body. In an effort to open a wider discussion
about editorial accountability, The Lancet established
an ombudsperson in July 1996. Clear criteria, based on the UK Parliamentary
Commissioner for Administration, were drawn up. Our ombudsman could investigate
delays in handling manuscripts and letters; editorial discourtesy; failure
to follow stated editorial procedures; failure to take reasonable account
of representations by authors and readers; and challenges to the publishing
ethics of the journal. Complaints about the substance of editorial decisions
were ruled out of the ombudsman's remit. Twenty complaints were recorded in
the first 18 months, 11 of which were upheld. The appointment of an ombudsperson
should be considered by editors of all scientific journals. Benefits go well
beyond immediate complaints, drawing an editor's attention, for example, to
the importance of efficient and courteous journal processes. No discernible
harm was discovered. A case still remains for wider oversight of the biomedical
AT THE 1994 Peer Review Congress in Chicago, Ill, Altman and colleagues1 presented a bleak and disturbing assessment of editorial
practice and called on medical journal editors to "explore possible procedures
for allowing authors' grievances to be heard and for possible sanctions if
complaints are upheld." They cited 3 instances in which editors had clearly
breached accepted standards of behavior. In one case, an editor reprinted
a complete article in his journal without seeking the author's permission.
In a second example, an editor commissioned a review and sent galley proofs,
but then published a similar review, with identical passages although different
authors, several months later. Finally, an editor sent a paper reporting adverse
events of a drug to a reviewer at the pharmaceutical company that manufactured
the drug. The paper was rejected.
In turning their attention away from researchers, who usually receive
much the worse blame for episodes of scientific misconduct, Altman et al focused
on "actual dishonesty by editors towards authors." They identified "the secrecy
inherent in the editorial aspects of scientific publishing" as a critical
reason to explain why so few misdeeds by editors come to light. Altman et
al proposed that an international scientific press council might enable authors
to seek redress in an independent setting.
In July 1996, The Lancet appointed an ombudsman—Professor
Thomas Sherwood, retiring dean of the medical school at the University of
Cambridge and a former member of the journal's international advisory board—"to
record and, where necessary, to investigate episodes of alleged editorial
maladministration when a complainant remains dissatisfied with the journal's
first response to criticism."2 The remit of
the ombudsperson is shown in Table 1.
We guaranteed that we would publish his reports annually.
We modelled our ombudsperson on the UK Parliamentary Commissioner for
Administration. The job description of this individual is well defined. He
investigates complaints about government departments; he is independent; his
investigations are confidential; he provides his services free; and he recommends
redress if he finds a complaint justified. His concerns are "maladministration
[that] has led to injustice."3 Maladministration
is defined as avoidable delay, bias or unfairness, failure to give appropriate
advice when asked, discourtesy or harassment, failure to follow proper procedures,
failure to take account of representations, and mistakes in the handling of
claims. In 1996 and 1997, for example, the UK ombudsman received 2219 complaints
about the National Health Service.
Complaints to our ombudsman were made directly to him in writing without
being seen by editors. His investigations involved reading correspondence
and filing. When necessary, he interviewed the relevant editor, assistant,
or other staff members concerned. Our ombudsman's final judgment was delivered
to that editor, to me, and to the complainant in writing.
In contrast to the National Health Service, our ombudsman received 11
complaints during his first year (July 1996-June 1997), increasing to 20 by
the end of 18 months.4 To develop mechanisms
to handle these inquiries, he visited the journal's office regularly to study
our editorial processes. He interviewed editors, talked to assistants, attended
all meetings, and dug through deep filing cabinets. His effect on editors
was startling. At first, most were either opposed to or profoundly skeptical
about an "outsider" having total freedom to roam through the editorial department,
ask questions, and read any document relevant to his investigations. Many
felt that this maneuver was an exercise in "big brother" oversight that compromised
their independence. Slowly, a phase of grudging acceptance was entered, and
now we have begun a period marked by polite approval, if not joyous enthusiasm,
at his presence. Our ombudsman seems to have shouldered his burden well; I
can detect no long-lasting adverse changes to his personality despite his
exposure to these difficult environmental pressures.
Appeals to our ombudsman concerned matters that went beyond the editorial
department (Table 2). Although
these complaints were not originally intended to be part of our ombudsman's
remit, he decided that it was a matter of courtesy to offer a reply. Inferior
standards were detected in the journal's decisions about overseas advertising
compared with more local procedures, and a subscriber complained about the
efficiency of our subscription service. Appeals against editorial decisions
about publication were made, despite clear instructions ruling out these complaints,
and were rejected. In one instance, the standard of our peer review process
was called into question and, after further peer review, editors agreed that
their earlier decision to reject a letter had been incorrect.
Povl Riis summed up the difficulties inherent in the editor-author relationship
very well when he wrote the following: ". . . the editorial system is generally
without appeal mechanisms. This makes justice a central ethical principle
of the editorial process. Justice means ensuring fairness in evaluation [of]
The ombudsperson offers one means of improving justice at the hands
of editors. However, the published reaction to our initiative has been largely
critical. George Davey Smith argued that "the real interest is in papers The Lancet rejects."6 The
effect of studying the outcome of rejection "would be considerably greater
than the flattery of authors through the correction of discourtesy. . . ."
Tim Albert, a former editor of a medical magazine himself and now a trainer
of editors, was irked. Why appoint "an ombudsman who spends most of his time
soothing the egos of disappointed authors"?7
"If editors spend most of their time worrying about whether they are being
fair to authors, they are not communicators; they have become bureaucrats,"
A former editor of The New England Journal of Medicine (NEJM) was even more caustic. In response
to a call for the NEJM to appoint an ombudsperson,
Arnold Relman pleaded, ". . . I can hope the NEJM's
present editors will resist that suggestion and wait for the results of your
initiative before deciding whether to open such a Pandora's box. It will be
interesting to see how many complaints . . . your ombudsman will be able to
satisfy—and at what cost."8
Despite the criticism heaped on us, our ombudsman has proven helpful.
His presence has led us to devote far more attention to our editorial process
than we might have done otherwise. Editors tend to be preoccupied by the content
of their journals, by the new data and ideas that they are dealing with, perhaps
at the expense of the seemingly more mundane parts of their editorial operation.
At The Lancet, we have started to set time
limits for handling all manuscripts, introduced a fast-track publication procedure
for articles and research letters, and appointed an editor with specific responsibility
for editorial process. Our instructions for authors have been clarified and
we have conducted our first author survey. An additional benefit is that we
now have an experienced and independent clinician who rules on matters of
editorial and commercial dispute. This assistance could be of decisive help
in maintaining the journal's integrity and independence.
One aspect of our work has been deliberately protected from the grasp
of our ombudsman—namely, decisions about publication or rejection of
submitted material. We judged that this activity is the proper—and final—responsibility
of the editor. An editor's job is to balance the content of a journal to create
a mix that will interest the reader. Appeals are welcome but an appeals procedure
outside the editor's control seems to run contrary to the role of editor.
If the appeal is handled badly—delays, rudeness, and the like—then
the ombudsperson has a part to play, but not otherwise.
Finally, the idea for an ombudsperson arose out of a call for an international
scientific press council. Is there still a case for another body to which
aggrieved authors could go? Clearly, yes. At the time of this writing (April
1998), The Lancet remains the only scientific journal
to have appointed an ombudsperson. If more journals were to appoint ombudspeople,
a press council might be a good place for them to share their experiences
and develop guidance for editors about how to improve their journals. Indeed,
it might provide a valuable means of supporting their decisions: for ombudspeople
surely have rights too.
Horton R. The Journal OmbudspersonA Step Toward Scientific Press Oversight. JAMA. 1998;280(3):298-299. doi:10.1001/jama.280.3.298