Epidemiology-Biostatistics Research Support
Guidelines for Authorship and Acknowledgments
Editorial: The New England Journal
of Medicine
November 21, 1991 -- Vol. 325, No. 21
On Authorship and Acknowledgments
Large clinical trials from multiple
institutions now involve dozens and sometimes hundreds of people in
their conception, design, implementation, analysis, and preparation of
reports for publication. Increasingly, we have become concerned about
two interrelated aspects of many reports of multicenter studies:
ambiguous authorship and lengthy acknowledgments.
Clear specification of authors is
essential so that any substantive questions about a submitted or
published study can be resolved. Authorship has been defined in many
ways, but most of the definitions have in common a requirement that
authors have sufficient intellectual involvement with the overall study
to be able to take responsibility for it 1,2,3. Obviously, a clinician
whose only contribution is to enter patients into a multicenter trial
does not qualify for authorship, nor does a secretary in the trial
office. Yet many such people may appear in a long list of members of a
study group designated as the collective author of a study - as a
hypothetical example, the Boston Porphyria Study Group. In such a case,
the study has at once too many authors, because not all of them could
take comprehensive responsibility for the study, and too few, because it
is not clear who is accountable. The problem is not limited to
multicenter clinical trials, of course, but it is more likely in such
studies because the cooperation of so many people is required and there
is a tendency to offer authorship to obtain it.
A second problem is the growing length
and detail of the acknowledgments. Traditionally, authors use
acknowledgments to identify those who made special intellectual or
technical contributions to a study that were not sufficient to qualify
them for authorship. In reports of multicenter clinical trials, however,
acknowledgments are often made to everyone who had anything to do with
the study, including those who were merely carrying out their jobs, such
as technicians. Sometimes principal investigators from each
participating institution are acknowledged, even though they are also
identified as authors. Many acknowledgments list committees, and the
same person may be acknowledged several times on different committees.
We recently accepted a manuscript with
an acknowledgment section that listed 63 institutions and 155
physicians, the number of patients each institution had contributed
(some as few as one), the 51 members of seven different committees,
their institutions and their specialties, and the secretaries in the
trial office. Many persons were named on more than one committee. The
paper was 12 pages long; the acknowledgments took up 5 pages. We do not
consider an extensive and repetitious list of participants in a clinical
trial, complete with committee assignments and other details of the
trial's internal organization, a good use of Journal space; it cannot be
of much interest to our readers. Furthermore, it tends to blur the
distinction between authors and those who merely warrant acknowledgment.
In the above example, all those who were acknowledged were also
considered authors, because they were members of the group to which
authorship was attributed.
The Journal subscribes to the criteria
for authorship formulated by the International Committee of Medical
Journal Editors and published in its Uniform Requirements for
Manuscripts Submitted to Biomedical Journals 4. According to these
criteria,
Each author should have participated
sufficiently in the work to take public responsibility for the content.
Authorship credit should be based only on substantial contributions to
(a) conception and design, or analysis and interpretation of data; and
to (b) drafting the article or revising it critically for important
intellectual content; and on (c) final approval of the version to be
published. Conditions (a), (b), and (c) must all be met.
These guidelines were developed in part
as a response to episodes of research fraud in which coauthors were too
remote from the work or had too limited a role to exercise the
responsibility that authorship implied 5. Defining the criteria for an
acknowledgment is, of course, far less important. We believe, however,
that acknowledgments become less meaningful if they include people who
were simply doing their jobs and who offered no unusual intellectual
contribution or technical expertise to the endeavor.
In the past we dealt with concern about
inappropriate authorship and acknowledgments on a case-by-case basis
during the review process or after we accepted a manuscript for
publication. But given the magnitude and frequency of the problems we
are encountering and our concern that ad hoc decisions may appear
arbitrary, we think it is appropriate to set out specific guidelines
about authorship and acknowledgments. It is difficult to arrive at such
guidelines because we do not wish to discourage cooperative research
enterprises and we are aware that there are pressures on principal
investigators to offer authorship as an incentive for enthusiastic
cooperation. Yet it is not in our readers' interests to permit unlimited
lists of authors and acknowledgments, and it undermines the meaning of
authorship and the value of an acknowledgment.
Accordingly, we shall institute the
following, fairly liberal guidelines for authorship and acknowledgments:
(1) Authorship attributed only to a group (the Boston Porphyria Study
Group) will not be acceptable. At least one person's name must accompany
the group name. The group name should appear after the authors' names,
as follows: "Thelma J. Smith, Louise J. Jones, and Duane J. Brown, for
the Boston Porphyria Study Group." (2) If more than 12 authors are
listed for a multicenter trial, or more than 8 for a study from a single
institution, we shall require that each author sign a statement
attesting that he or she fulfills the criteria for authorship of the
Uniform Requirements. The reason for selecting these maximums, which are
admittedly arbitrary, is that it is difficult to imagine that more than
12 people can have the comprehensive intellectual involvement necessary
to fulfill the criteria for authorship. (3) We shall leave to the
authors the choice of those acknowledged, but limit the space devoted to
acknowledgments. To conserve space, those acknowledged will be listed
only once, along with their institutions (one each). Committee names,
numbers of patients contributed, and other details about the process of
the trial will not be included. (4) If acknowledgments fill more than a
column of Journal space (about 600 words of small type), we shall
deposit them with the National Auxiliary Publications Service. At the
authors' request we shall consider publishing fuller acknowledgments,
including committee assignments, in reprints of the paper.
To remind authors of these guidelines,
they will be included in our Information for Authors, which appears in
the first issue of each month. We are aware that there may be individual
circumstances in which these policies need to be modified. In
particular, modifications may be required for reports of trials already
in progress in which commitments have been made about the way
participants are to be designated. Even in these cases, however, we
would expect those named as authors to have fulfilled the criteria for
authorship. We welcome your comments about these policies.
Jerome P. Kassirer, M.D
Marcia Angell, M.D.
References
- Huth EJ. Editors and the problems of
authorship: rulemakers or gatekeepers? In: Council of Biology Editors
Editorial Policy Committee. Ethics and policy in scientific
publication. Bethesda, Md.: Council of Biology Editors,
1990:175-80.Return to Text
- Huth EJ. Authorship from the
reader's side. Ann Intern Med 1982;97:613-4. Return to Text
- Relman AS. Responsibilities of
authorship: where does the buck stop? N Engl J Med
1984;310:1048-9.Return to Text
- International Committee of Medical
Journal Editors. Uniform requirements for manuscripts submitted to
biomedical journals. N Engl J Med 1991;324:424-8.Return to Text
5. Relman AS. Lessons from the Darsee affair. N Engl J Med
1983;308:1415-7.
Top |
