AMA Manual of Style - Stacy L. Christiansen, Cheryl Iverson 2020

Authorship Responsibility
Ethical and Legal Considerations

Some judge of authors’ names, not works, and then

Nor praise nor blame the writings, but the men.

Alexander Pope¹

Nearly 70 years ago, Richard M. Hewitt, MD, then head of the Section of Publications at the Mayo Clinic, described the ethics of authorship in a JAMA article entitled “Exposition as Applied to Medicine: A Glance at the Ethics of It.”² The following excerpts from Hewitt’s article demonstrate an appreciation of the basic ethical responsibilities and obligations of authorship:

Authorship cannot be conferred; it may be undertaken by one who will shoulder the responsibility that goes with it.

The reader of a report issued by two or more authors has a right to assume that each author has some authoritative knowledge of the subject, that each contributed to the investigation, and that each labored on the report to the extent of weighing every word and quantity in it.

If we would define publication of unoriginal, repetitious medical material as a violation of medical ethics, and would officially reprove it as such, the tawdry author would be silenced and the genuine one helped.

The by-line, then, is not merely a credit-line. He who took some part in the investigation, be it ever so minor, is entitled to credit for what he did. . . . Further, the generous chap who would bestow authorship on another, perhaps without even submitting the manuscript to him, may do his colleague no favor. For the investigation is one thing, the report of it another, and, sad the day that this must be admitted: The investigation may have been excellent but the report, bad.

Since all of us necessarily adopt and absorb the ideas of others, we must be scrupulous in maintaining the spirit of acknowledgment to others. Fundamentally, your integrity is at stake. Unless you make specific acknowledgment, you claim the credit for yourself for anything that you write. In general, it is better to say too much about your sources than too little.

The author who paraphrases or refers to an article should have read it.

5.1.1 Authorship: Definition, Criteria, Contributions, and Requirements.

Authorship offers significant professional and personal rewards, but these rewards are accompanied by substantial responsibility. During the 1980s, biomedical editors began requiring contributors to meet specific criteria for authorship. These criteria were first developed for medical journals under the initiative of Edward J. Huth, MD,³ then editor of the Annals of Internal Medicine, who cited Hewitt’s work² during discussions at the 1984 meeting of the International Committee of Medical Journal Editors (ICMJE). The ICMJE guidelines were first published in 1985⁴ and are now part of the Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals⁵ (see 2.0, Manuscript Preparation for Submission and Publication). These guidelines are reviewed, revised, and updated regularly, and numerous biomedical journals, the US National Library of Medicine,⁶ and the Council of Science Editors^7,8 use them as the foundation for policies and procedures on authorship.

5.1.1.1 Authorship Definition and Criteria.

According to the ICMJE guidelines, all authors should have participated sufficiently in the work to take public responsibility for the content, either all the work or an important part of it. To take public responsibility, an author must be able to defend the content (all or an important part) and conclusions of the article if publicly challenged. Sufficient participation means that substantial contributions have been made in each of the following areas⁵:

1.Conception and design of the work; or acquisition, analysis, or interpretation of the data for the work; and

2.Drafting the work or revising it critically for important intellectual content; and

3.Approval of the version to be published; and

4.Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

In 2013, the ICMJE added the fourth criterion to address concerns that all authors be accountable for the work if questions arise about its accuracy or integrity. According to the ICMJE recommendations, “an author should be able to identify which co-authors are responsible for specific other parts of the work. In addition, authors should have confidence in the integrity of the contributions of their co-authors.”⁵

To justify authorship, an author must meet each of the 4 criteria. However, the term substantial contribution has not been adequately defined (perhaps to allow for wider application of the ICMJE criteria for authorship). As a result, the first criterion, “conception and design; or acquisition, analysis, or interpretation of the data,” may be interpreted broadly. For example, an author of a nonresearch manuscript may not have analyzed data per se but may have analyzed literature, events, theories, arguments, or opinions. A substantial contribution may be interpreted as an important intellectual contribution, without which the work, or an important part of the work, could not have been completed or the manuscript could not have been written and submitted for publication.

The ICMJE also notes that the following contributions, alone, are not sufficient to justify authorship⁵: “acquisition of funding; general supervision of a research group or general administrative support; and writing assistance, technical editing, language editing, and proofreading” (see also 5.1.2, Ethical and Legal Considerations, Authorship Responsibility, Guest and Ghost Authors and Other Contributors). Those who do not meet the 4 authorship criteria should be acknowledged (see 5.2.1, Acknowledging Support, Assistance, and Contributions of Those Who Are Not Authors). The ICMJE notes that these criteria are “intended to reserve the status of authorship for those who deserve credit and can take responsibility for the work” and that these criteria are not intended to “disqualify colleagues from authorship who otherwise meet authorship criteria by denying them the opportunity” to participate in writing or reviewing and approving the manuscript. The ICMJE states that “all individuals who meet the first criterion should have the opportunity to participate in the review, drafting, and final approval of the manuscript.”⁵

See Box 5.1-1 for a list of common terms used for contributor, author, and collaborator.⁹

Box 5.1-1. Common Terms: Contributor, Author, and Collaborator

Contributor: Anyone—an author, a collaborator, or others—who has assisted or contributed in a meaningful way to a work.

Author: A contributor who has participated sufficiently in the work to take public responsibility for the content, either all the work or an important part of it, and meets defined criteria for authorship. Identification of authorship in a manuscript/article can appear in 2 ways.

Byline author: Author name in byline

Nonbyline author: Author name not in byline—listed elsewhere, typically in an Acknowledgment or Article Information section.

Group author: A group of individuals, usually involving multicenter study investigators, members of working groups, and official or self-appointed expert boards, panels, or committees, who wish to display a group name to indicate authorship. (Also known as corporate authorship.)

Collaborator: A nonauthor member of a formal group who contributes significantly to the work; often also called investigator. In these cases, a group name must be in the byline.

Other contributors: Anyone else who contributed in some meaningful way who is not an author and is not a nonauthor collaborator. They can be listed under Additional Contributions in an Acknowledgement or Article Information section (see 5.2, Ethical and Legal Considerations, Acknowledgments).

5.1.1.2 Author Contributions.

Authors may not be aware of the ICMJE authorship criteria. To inform or remind authors of these responsibilities and to encourage appropriate authorship, many journals require authors to attest in writing how they qualify for authorship and to indicate their specific contributions to the work.^5,10,11,12 The ICMJE guidelines state, “Editors are strongly encouraged to develop and implement a contributorship policy, as well as a policy on identifying who is responsible for the integrity of the work as a whole.”⁵ Some journals ask authors to describe their specific contributions in an open-ended narrative format, some describe examples of various types of author contributions, and some journals provide a list of specific contributions in the form of a checklist. For example, the JAMA Network journals require all authors to complete a statement of authorship responsibility based on the ICMJE guidelines and to indicate their specific contributions from a checklist based on the ICMJE authorship criteria and empiric data from studies of authorship and author contributions. This statement is required for authors of all types of manuscripts, including editorials and letters to the editor¹² (see Box 5.1-2). The JAMA Network journals use an authorship form for authors to indicate information about authorship responsibility, criteria, and contributions, as well as information about conflicts of interest and funding, publishing agreement and copyright license options, and an acknowledgment statement. An updated example of this authorship form is available online in the JAMA Instructions for Authors.¹²

Box 5.1-2. The JAMA Network Journals Authorship Responsibility, Criteria, and Contributions Statement

Each author should meet all criteria below (A, B, C, and D) and should indicate general and specific contributions by reading criteria A, B, C, and D and checking the appropriate boxes.

□ A.I certify that

✵The manuscript represents original and valid work and that neither this manuscript nor one with substantially similar content under my authorship has been published or is being considered for publication elsewhere, except as described in an attachment, and copies of closely related manuscripts have been provided; and

✵I agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; and

✵If requested, I will provide the data or will cooperate fully in obtaining and providing the data on which the manuscript is based for examination by the editor or the editor’s assignees; and

✵For papers with more than 1 author, I agree to allow the corresponding author to serve as the primary correspondent with the editorial office, to review the edited manuscript and proof, and to make decisions regarding release of information in the manuscript to the media, federal agencies, or both; or, if I am the only author, I will be the corresponding author and agree to serve in the roles described above.

□ B.I have given final approval of the submitted manuscript.

C.I have participated sufficiently in the work to take public responsibility for (check 1 of 2 below)

□part of the content.

□the whole content.

D.To qualify for authorship, you must check at least 1 box for each of the 3 categories of contributions listed below.

I have made substantial contributions to the intellectual content of the paper as described below:

1.Check at least 1 of the 2 below

□concept and design

□acquisition, analysis, or interpretation of data

2.Check at least 1 of the 2 below

□drafting of the manuscript

□critical revision of the manuscript for important intellectual content

3.Check at least 1 below

□statistical analysis

□obtained funding

□administrative, technical, or material support

□supervision

□no additional contributions

□other contributions (specify)________________________________

Some journals publish author contributions. This practice, first suggested by Rennie et al in 1997^10,11 and endorsed by the ICMJE⁵ and the Council of Science Editors,^7,8 makes the specific contributions of authors transparent to editors and readers. For example, the JAMA Network journals publish the specific contributions of each author for articles that report original data (eg, research and systematic reviews) in the Acknowledgment section at the end of the article (see 5.2, Acknowledgments).

Another initiative, following the rationale of identifying contributions that many biomedical journals used for many years, is called the Contributor Roles Taxonomy (CRediT).¹³ This taxonomy “provides a high-level classification of the diverse roles performed in the work leading to a published research output in the sciences” and includes the following: conceptualization, methods, software, validation, formal analysis, investigation, resources, data curation, writing−original draft preparation, writing−review and editing, visualization, supervision, project administration, and funding acquisition.¹⁴ The purpose of this taxonomy system is “to provide transparency in contributions to scholarly published work, to enable improved systems of attribution, credit, and accountability.”¹⁴ According to the CRediT organizers, the implementation of this taxonomy is not intended to replace authorship or “define what constitutes authorship. Rather, the roles are intended to apply to all those who contribute to research that results in scholarly published works, and it is recommended that all tagged contributors be listed, whether they are formally listed as authors or named in acknowledgements”¹⁴ (see 5.2.3, Author Contributions).

5.1.1.3 Additional Author Requirements.

Depending on the journal, all authors may also be required to complete a publishing agreement (transferring copyright or a publication license or opting for an open access license), identify relevant conflicts of interest or to declare no such interests, identify funding and the role of the sponsors in the work to be published, attest that they had access to the data for reports of original research, and indicate a plan for data sharing (required for reports of clinical trials; optional for reports of other study types) (see 5.6.5, Copyright Assignment or License, and 5.5.2, Requirements for Authors). Many journals send emails to all authors at various stages of the manuscript submission and evaluation processes to confirm that all are indeed authors or to have them complete authorship and disclosure forms and publishing agreements. Some journals have begun to request or require authors to provide unique researcher identifiers in an attempt to disambiguate or distinguish individual authors from other authors and improve authentication and the linking of outputs (eg, funding, published journal articles) with individual authors.¹⁵ A common identifier is ORCID, (Open Researcher and Contributor ID), which is a universal resource identifier with a 16-digit number that is compatible with the International Organization for Standardization’s Standard International Standard Name Identifier (see https://orcid.org/).

5.1.1.4 Access to Data and Data Sharing Statement Requirements.

The ICMJE recommends that journals ask authors to indicate whether they “had access to the study data, with an explanation of the nature and extent of access, including whether access is on-going.”⁵ Consistent with this recommendation, the JAMA Network journals require at least 1 named author (eg, the principal investigator), and preferably no more than 2 authors, to indicate that she or he had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis for all reports that contain original data (eg, research articles, systematic reviews, and meta-analyses)¹² (see 5.2.5, Access to Data Statement, and 5.5.5, Access to Data Requirement).

For reports of randomized clinical trials, the ICMJE recommends that authors of clinical trials provide a data sharing statement with submitted manuscripts and indicate if data, including individual patient data, a data dictionary that defines each field in the data set, and supporting documentation, will be made available to others; when, where, and how the data will be available; types of analyses that are permitted; and if there will be any restrictions on the use of the data.^4,16 This follows a previous ICMJE requirement for authors of clinical trials to also include trial protocols with submitted manuscripts and published articles. Data sharing statements may also be submitted for reports of other types of studies, but the ICMJE member journals only require this for reports of clinical trials.

The data sharing statements should address the following items:

Data

Will the data collected for your study, including individual patient data and a data dictionary defining each field in the data set, be made available to others?

Yes or No (if No, authors may explain why data are not available)

List all data that will be made available.

Deidentified participant data

Participant data with identifiers

Data dictionary

Other (please specify)

List where to access these data. Provide complete URL if data will be available in a repository or website, or provide complete email address if request for data must be sent to an individual.

List the beginning date and end date (if applicable) when these data will be available. If the beginning date of data availability will be when the article is published, please indicate “with publication.”

With publication

At a date different from publication

Beginning date

End date (if applicable)

Supporting Documents

If your manuscript is accepted for publication, the journal will publish your trial protocol, including the statistical analysis plan, and any amendments as online supplements. Please list any other supporting documents that you wish to make available (eg, statistical/analytic code, informed consent form).

Statistical/analytic code

Informed consent form

None

Other (please specify)

List where to access these documents. Provide complete URL if the documents will be available in a repository or website, or provide complete email address if request for documents must be sent to an individual.

List the beginning date and end date (if applicable) when the documents will be available.

With publication

At a date different from publication

Beginning date

End date (if applicable)

Additional Information

Indicate the types of analyses for which the data will be made available (eg, for any purpose or for a specified purpose).

Indicate the mechanisms by which the data will be made available (eg, with investigator support, without investigator support, after approval of a proposal, or with a signed data access agreement).

List any additional restrictions on the use of the data or any additional information.

If you would like to offer context for your decision not to make the data available, please enter it below (optional).

5.1.1.5 Corresponding Author.

Every manuscript and published article should have at least 1 author who will serve as the primary contact and correspondent for all communications about the submitted work and, if it is accepted for publication, the published article. It is not efficient for editorial offices to have more than 1 formal corresponding author. However, it is helpful to provide the editorial office with contact information for coauthors in case the corresponding author becomes unavailable during the editorial and publication processes. For example, the JAMA Network journals require a corresponding author for each submitted manuscript to serve as the primary correspondent with the editorial office and, if the manuscript is accepted, to review an edited manuscript and proof, to make decisions regarding release of information in the manuscript to the news media and/or federal agencies, and to have his or her name published as corresponding author in the article.¹² Corresponding authors for the JAMA Network journals also complete a statement that they have identified all persons who have made substantial contributions to the work but who are not authors¹²:

I certify that all persons who have made substantial contributions to the work reported in this manuscript (eg, data collection, analysis, or writing or editing assistance) but who do not fulfill the authorship criteria are named with their specific contributions in an Acknowledgment in the manuscript.

I certify that all persons named in the Acknowledgment have provided me with written permission to be named.

I certify that if an Acknowledgment section is not included, no other persons have made substantial contributions to this manuscript.

(See 5.1.2, Guest and Ghost Authors and Other Contributors, and 5.2, Acknowledgments.)

5.1.1.6 Co—Corresponding Authors.

There have been increasing requests for multiple corresponding authors, which risks diluting the meaning of corresponding author and confusion for journal editors about which author is primarily responsible for communications before and after publication.⁹ The JAMA Network journals consider requests for co—corresponding authors on a limited basis if justified but permit no more than 2 co—corresponding authors. In such cases, the journals require that a primary corresponding author be designated as the point of contact responsible for all communication about the manuscript and article, manage the tasks described above, and be listed first in the corresponding author section.^9,12

5.1.2 Guest and Ghost Authors and Other Contributors.

At least 1 author must be responsible for any part of an article crucial to its main conclusions, and everyone listed as an author must have made a substantial contribution to that specific article.⁵ As described in 5.1.1, Authorship: Definition, Criteria, Contributions, and Requirements, many journals require authors to complete statements of authorship responsibility and to indicate specific contributions of all authors. In addition to improving the transparency of author responsibility, accountability, and credit, these policies may help eliminate guest authors and identify ghost authors.^{17,18,19,20,21,22,23,24}

5.1.2.1 Guest (Honorary) Authors.

A guest author or honorary author is a person who has not contributed substantially or at all to the published work but is listed as an author. Traditionally, supervisors, department chairs, and mentors have been given guest, or honorary, places in the byline even though they have not met all the criteria for authorship. However, this custom is not acceptable because it devalues the meaning of authorship.^17,19,22,24 The ICMJE guidelines state specifically that acquisition of funding and general supervision of the research, alone without other contributions, are not sufficient for authorship.⁵ Such supervision and participation should be noted in the Acknowledgment (see 5.2, Acknowledgments). Guest authors have also included well-known persons in a particular field who have accepted money or other compensation to have their names attached to a manuscript that has already been researched and prepared by a ghost writer for an organization with a commercial interest in the subject of the paper.^17,25,26,27 Such practice clearly is deceitful.^17,28 Several studies^20,21,29 have documented the prevalence of guest authors in biomedical journals, ranging from 10% of research articles to 39% of review articles, in journals that were not requesting authors to disclose their specific contributions.

5.1.2.2 Ghost Authors and Ghost Writers.

A ghost author is someone who has participated sufficiently in the research or analysis and writing of a manuscript to take public responsibility for the work but is not named as an author in the byline or Acknowledgment section of the published work. Studies have found prevalence rates of journal articles with ghost authors that range from 8% to 26%.^20,21,29 In biomedical publication, ghost authors have included employees of pharmaceutical companies (eg, researchers, managers, statisticians, epidemiologists), medical writers, marketing and public relations writers, and junior staff writing for elected or appointed officials.¹⁷ As described elsewhere, ghost writers have been hired by firms with commercial interests to write reviews of specific subjects, and their authorship is not disclosed.^{17,18,25,26,27,28,30}

Ghost writers are not necessarily ghost authors. For example, a writer may not have participated in the research or analysis of a study but may have been given the data and asked to draft a report for publication. If participants in the project do not meet all the criteria for authorship but have made substantial contributions to the research, writing, or editing of the manuscript, those persons should be named, with their permission, in the Acknowledgment along with their contributions and institutional affiliations, if relevant^12,17 (see 5.2, Acknowledgments). Editors and authors should not permit anyone who has participated sufficiently to meet authorship criteria or any nonauthor who has made other important contributions to not be appropriately identified in the byline or Acknowledgment, respectively (see 5.2, Acknowledgments, and 5.1.6, Changes in Authorship).

To give proper credit to medical writers and authors’ editors, journal editors should require authors to identify all persons who have participated substantially in the writing or editing of the manuscript. Substantial editing or writing assistance should be disclosed to the editor at the time of manuscript submission and included in the Acknowledgment, along with the person’s name, academic degrees, affiliation, type of assistance, and whether compensation for this assistance was provided^12,17 (see 5.2, Acknowledgments). The American Medical Writers Association supports this transparency in a position statement that recommends that “biomedical communicators who contribute substantially to the writing or editing of a manuscript should be acknowledged with their permission and with disclosure of any pertinent professional or financial relationships.”³¹

Some journals specifically prohibit the submission of manuscripts that have been ghostwritten.^32,33 For example, the journal Neurology requires that “professional writers employed by pharmaceutical companies or other academic, governmental, or commercial entities who have drafted or revised the intellectual content of the paper must be included as authors.”³³ Other journals require declaration and explanation of substantive writing and editing assistance. For example, corresponding authors of the JAMA Network journals attest to a statement that all persons who have made substantial contributions to the work (eg, data collection, analysis, or writing or editing assistance) but who do not fulfill the authorship criteria are named with their specific contributions in an acknowledgment in the manuscript.¹² The Lancet requires an additional signed statement from any medical writers or editors who provided substantial writing or editing assistance.³⁴

Journal editors and manuscript editors who substantially edit a manuscript to be published in a journal generally are not specifically acknowledged if their names appear in the journal’s masthead or elsewhere in the journal.

5.1.3 Unsigned Editorials, Anonymous Authors, Pseudonymous Authors.

The practice of publishing unsigned or anonymous editorials provides “vituperative editorialists” protection from adversaries or competitors when taking unpopular stands in the pages of their journals.³⁵ However, without named authors and affiliations, readers lack information to judge the objectivity and credibility of such articles. Although this practice is the norm for newspaper editorial pages, it has fallen out of use in most peer-reviewed journals. One rationale for anonymity has been that editorials, signed or not, represent the official opinion of the publication or the owner of the publication. However, such anonymity distances the real author(s) from accountability. For many years, JAMA published unsigned editorials. However, beginning in 1960 JAMA began to publish signed or initialed editorials, and since 1970 all JAMA editorials have been signed by their authors, including editorials written by the journal’s editors. The BMJ began publishing signed editorials in 1981.³⁶ As of this writing, The Lancet continues to publish unsigned editorials that reflect an unstated consensus among the editors^34,37 (see 1.4, Opinion Articles).

Journals that publish unattributed editorials and unattributed scientific articles may give contradictory messages to their readers about the merits and responsibility of authorship. Authors who submit scientific papers must publicly stand by what they write, whereas unsigned editorialists can remain anonymous or hide behind a journal’s masthead. Unattributed editorials may also allow the publisher or owner of the journal and influential organizations to compromise the journal’s editorial independence (see 5.10, Editorial Freedom and Integrity). In addition, PubMed indicates “no authors listed” for anonymous and pseudonymous articles. Therefore, authors’, names are published with all articles, including editorials in the JAMA Network journals.

Occasionally, an author may request that his or her name not be used in publication. If the reason for this request is judged to be important (such as concern for personal safety or fear of political reprisal, public humiliation, or job loss), the article could be published without that author’s name. However, justification for such publication is very rare and should include careful consideration of the value of the information to be published as well as the potential risks to the author. In such rare cases, the phrase “Name withheld on request” could be used in place of the author’s name, perhaps with an explanation in the Acknowledgment (see 2.2, Author Bylines and End-of-Text Signatures).

If anonymity is to be used, the author must still complete statements of authorship responsibility, conflicts of interest disclosure, and a publishing agreement (using his or her actual name), and those records must be kept confidential as part of the manuscript file (see 5.7.1, Confidentiality During Editorial Evaluation and Peer Review and After Publication). For the rare case in which withholding of an author’s name is justified, the author’s name should be withheld from peer reviewers and readers. However, reviewers and readers should be informed that the author has requested anonymity. Citations to such articles in PubMed will note “no authors listed” in the author field.

Pseudonyms are inappropriate in bylines of scientific reports because they are misleading and cause problems for literature citations.

5.1.4 Number of Authors.

The number of authors whose names appear in the byline of scientific papers increased steadily during the second half of the 20th century.³⁸ Data from MEDLINE/PubMed of more than 29 million indexed articles indicate an average increase in the number of authors per article from 1.9 before 1975 to 5.81 in 2015-2019.³⁹ Beginning in the 1990s, articles with large numbers of authors, usually as part of a collaborative group, began to be published. According to an analysis by Aboukhalil,⁴⁰ in 1998, the first articles with more than 500 authors were published, and in 2010, articles with more than 2000 authors were published (many of these in high-energy particle physics journals). This increase occurred because of specialization, multidisciplinary and multinational collaboration, the advent of large multicenter studies, and an increase in team science. However, authorship inflation has diluted the meaning of authorship. For example, which authors in a byline that contains more than 100 names can state that they actually wrote the paper or that they participated sufficiently to take public responsibility for the work? In response to this problem, suggestions were made in the 1980s and 1990s to limit the number of authors listed in the byline and database citations.^41,42 However, such limitations may be considered as arbitrary limits and may interfere with policies to encourage transparency of author contributions. The US National Library of Medicine no longer limits the number of individual authors’ names listed in an article’s citation in MEDLINE/PubMed.^39,43

For major articles (eg, research and review articles), the JAMA Network journals do not set limits on the number of authors that can be listed, as long as each author meets the journal’s criteria for authorship and each author completes an authorship form that indicates specific contributions. However, the JAMA Network journals limit the number of authors for an opinion article or letter. For practical reasons, it might not be possible to list a large number of authors because of limited space available on the first page of a print/PDF article or HTML design limitations. In such cases, the names of all authors in an article with a large number of authors may be listed at the end of the article instead of in the byline at the beginning of the article (see 5.1.9, Group and Collaborative Authorship).⁹ In addition, many online displays of articles present truncated lists of authors on article pages (eg, 2-3 authors and “et al”) by default with the option for a reader to expand the display to the full list of authors.

Also for practical reasons, many journals limit the number of authors listed in reference list citations (see 3.7, Authors, and 5.1.9, Group and Collaborative Authorship). However, the online versions of many journal articles contain reference lists with links to original articles and to MEDLINE records in PubMed, both of which list all authors for articles.

5.1.5 Order of Authorship and Shared Positions.

Before proposals for identifying authors’ contributions began to be implemented, proposed guides for determining order of authorship ranged from simple alphabetical listings to mathematical formulas for assessing specific levels of individual contributions.^44,45,46 However, even the most systematic calculations of contribution levels will require some measure of subjective judgment, and determination of order of authors is best done by the authors’ collective assessment of each author’s level of contribution. Moreover, as Rennie et al¹⁰ have argued, attempts to provide information to readers by ordering authors in particular ways are not meaningful, especially if each author’s contributions are not made public. The following may help determine order of authorship⁴⁷:

1.Only those individuals who meet the criteria for authorship may be listed as authors (see 5.1.1, Authorship: Definition, Criteria, Contributions, and Requirements).

2.The first author has contributed the most to the work, with other authors listed in descending order according to their levels of contribution. Note: Some groups of authors choose to list the most senior author(s) last, irrespective of the relative amount of their contributions.

3.Decisions about the order of authors should be made as early as possible (eg, before the manuscript is written) and reevaluated later as often as needed by consensus (see 5.1.6, Changes in Authorship).

4.Disagreement about order should be resolved by the authors, not the editor (see guidance in 5.1.8, Author Disputes).

5.Authors may provide a publishable footnote explaining the order of authorship, if there is a compelling reason.

6.Editors may request documentation of authors’ specific contributions.

It has become increasingly common for authors to request “co—first authorship,” “co—senior authorship,” or some other indication of equal contribution.⁹ Journals will accept indication of co—first authorship, but one person’s name will need to go first in the byline or author list. Requests for “co—first authorship” or “co—senior authorship” beyond 3 or 4 named authors may not be justifiable. This information can be displayed in the Acknowledgment just before the list of author contributions, such as

Drs Brown and Jones served as co—first authors and contributed equally to the work.

5.1.6 Changes in Authorship.

Changes made in authorship (ie, order, addition, and deletion of authors) should be discussed and approved by all authors.^9,12 Any requests for changes in authorship after initial manuscript submission and before publication should be explained in writing to the editor in a letter signed by all authors, or if sent by email, all authors should be copied (ie, included as recipients of the email). BMJ’s policy for alterations in authorship of manuscripts under consideration is a useful guide for other journals: “Any change in authors and/or contributors after initial submission must be approved by all authors. This applies to additions, deletions, change of order to the authors, or contributions being attributed differently. Any alterations must be explained to the editor. The editor may contact any of the authors and/or contributors to ascertain whether they have agreed to any alteration.”⁴⁸ The Committee on Publication Ethics (COPE) also has useful guidance and a flowchart for addressing changes in authorship.⁴⁹

5.1.7 Deceased or Incapacitated Authors.

In the case of death or incapacitation of an author during the manuscript submission and review or publication process, a family member, an individual with power of attorney, or the corresponding author can confirm that the deceased or incapacitated person should be listed as an author. In this event, the corresponding author can forward correspondence from the individual representing the deceased author and can provide information on the deceased or incapacitated author’s contributions. Designation that an author is deceased can be made in the Acknowledgment or Article Information section of the manuscript/article (see 2.3.2, Death of Author[s]).

5.1.8 Author Disputes.

Authorship disputes sometimes occur. For example, 10% of researchers who have received a grant from the US National Institutes of Health admitted to assigning authorship “inappropriately.”⁵⁰ In surveys of plastic surgeon authors, 29% reported being involved in a dispute with a colleague over authorship issues in 2003, and 22% reported being involved in such disputes in 2011.⁵¹ According to Karen Peterson, scientific ombudsman at the Fred Hutchinson Cancer Research Center in Seattle, Washington, 20% of the disputes adjudicated in the ombudsman office were about authorship.⁵² A 2011 systematic review of 118 studies about authorship issues in various scholarly disciplines identified 4 common themes: “authorship perceptions, definitions and practices, defining order of authors on the byline, ethical and unethical authorship practices, and authorship issues related to student/non-research personnel-supervisor collaboration.”⁵³ In meta-analysis of 14 of these studies, 29% (95% CI, 24%-35%) of researchers reported their own or others’ experience with misuse of authorship.⁵³

COPE lists numerous case examples of authorship disputes with relevant guidance and information on resolutions.⁴⁹ Authorship disputes are best resolved among the authors or within the institutions of the author. Useful tips for authors involved in such disputes include the following⁵²:

First try to discuss the issue amicably and try to understand the other person’s point of view. For example, discuss how the idea for the work was first conceived.

If a junior author has a dispute with a senior author, who is that person’s supervisor, keep the tone inquisitive, not accusatory. Consider asking a question intended to understand how authorship was decided.

If an individual contributor’s authorship merit is in question, consider what the manuscript would have looked like without that person’s contribution, and whether another author could have made or did make the same contribution.

In addition, authors can consult an intended journal’s Instructions for Authors and the ICMJE criteria for authorship⁵ and discuss how each author meets the criteria to help resolve any disputes.

5.1.9 Group and Collaborative Authorship.

Group, collaborative, or corporate authorship usually involves multicenter study investigators, members of working groups, and official or self-appointed expert boards, panels, or committees (see Box 5.1-1). These groups can comprise hundreds of participants and often represent complex, multidisciplinary and multinational collaborations; therefore, decisions about listing group authorship pose several problems and dilemmas for authors, editors, journals, librarians, and bibliographic databases^{6,9,54,55,56,57} (see 13.9, Collaborative Groups, and 2.2.4, Multiple Authors, Group Authors).

Some large clinical trials and observational studies are often best known and frequently referred to by their study name (eg, Women’s Health Initiative or the Thrombolysis in Myocardial Infarction Study Group) or by their abbreviation (eg, WHI or TIMI Study Group). As a result, these groups often include the official name of the study group in an article’s byline (ie, the position on an article’s title page where authors are listed). However, not all members of a study group may meet authorship criteria (see 5.1.1, Authorship: Definition, Criteria, Contributions, and Requirements), and having the group name in the byline does not distinguish those members of the group who qualify for authorship from those who do not. In addition, without a single person named as author, no individual person can take responsibility and be held accountable for the work. For this reason, at least 1 individual (eg, the corresponding author or the principal investigator) should be named as primary corresponding author or guarantor (see 5.1.1.5, Corresponding Author, and 2.0, Manuscript Preparation for Submission and Publication). To address these concerns, members of a writing team or a subgroup are often identified as the authors for large groups.

For group author articles, providing appropriate credit and accountability for the many individuals involved—authors and nonauthor collaborators—and ensuring proper citation to enable online searching and retrieval of the articles are important considerations.^9,54,58 The guidelines that follow may help authors and editors determine who should be listed and where.

One or more authors may take responsibility for a group (as the authors or writing team). In this case, the names of individual authors are listed in the byline with a designation that these authors are writing on behalf of or for the group. Those members of the group who do not qualify for authorship would not be listed in the byline but may be listed in the Acknowledgment at the end of the article. In this case, the byline might read as follows:

Jacques E. Rossouw, MBChB, MD; Garnet L. Anderson, PhD; for the Women’s Health Initiative

or

Writing Group for the Women’s Health Initiative

In the latter example, the writing group members are the authors for the group, and their names should be listed in the author affiliation or Acknowledgment section (with their specific contributions identified). In these cases, the formal group author name (eg, Women’s Health Initiative) should be coded in the journal’s online version and in bibliographic databases so that the results of online searches for articles from this group will include articles that combine individual names or a subgroup name (eg, the Writing Group) with the formal group name in the byline.

The other nonauthor group members and their contributions may then be listed separately in the Acknowledgment section (see Box 5.1-1, Common Terms: Contributor, Author, and Collaborator and 5.2, Acknowledgments). PubMed lists these individuals in the article records as “collaborators.”⁶

To ensure that authors are cited appropriately in bibliographic databases, explicit use of the term authors or members of the writing group or writing committee for the article is preferred. According to the National Library of Medicine, the group name must be in the byline to be reflected as an author in the MEDLINE citation.⁶

Authorship can be attributed to an entire group, although this practice may be less common than the examples given above. However, as with all articles, clear justification for all members of the group meeting all criteria and requirements for authorship must be made, and for journals that publish authors’ individual contributions, all members of the group must identify their specific contributions (see 5.1.1, Authorship: Definition, Criteria, Contributions, and Requirements). In this case, the byline might read as follows:

Clinical Outcomes Trial Investigators

In cases in which every member of a large group qualifies for authorship and the group name appears in the byline, the individual members of the study group should be listed separately in the Acknowledgment section or in a clearly identified position within the article, such as a box set off by rules (as described in 2.2.4, Multiple Authors, Group Authors).

If the group name appears in the byline, it is recommended that at least 1 person serve as the corresponding author and be named as an individual who will coordinate questions about the article. This person can be named in the affiliation footnote as corresponding author. In this case, the byline might read

Clinical Outcomes Trial Investigators

and the affiliation footnote might read

Author Affiliations: A complete list of the authors in the Clinical Outcomes Trial Investigators group appears at the end of this article.

Corresponding Author: James S. Smith, MD, Department of Neurology, University of Chicago Medical School, 555 S Main St, Chicago, IL 60615 (smithjs@umc.edu).

(see 2.3.3, Author Affiliations, and 2.10.7, Corresponding Author Contact Information).

Publishing the names of all authors and their specific contributions, no matter how many, with the specific article is preferred. However, a long list of investigators and their affiliations could occupy a lot of article “space” (eg, several journal pages of print journals or PDFs or screen space online). Nevertheless, it is important to publish the name of each author with the article, for reasons of accountability and credit and to allow proper searching and retrieval of articles by individual author names in bibliographic databases. Best practice is to publish all author names in the article, even if at the end of the article in an Acknowledgment or Article Information section. Nonauthor collaborators who contribute substantially to studies (eg, investigators) should be listed in the Acknowledgment or Article Information section. A less effective option is to publish the list of nonauthor collaborators in an online-only supplement to the article, as long as this is made clear to readers. It is important to include and tag all authors and nonauthor collaborators in the article XML, otherwise, there is a risk that lists of authors or collaborators not included in the main articles may not be properly included in bibliographic databases (see 5.1.10, Standards for Electronic Editing and Tagging of Names of Authors, Collaborators, and Group Authors).

Study or other group participants should not be promised authorship status and a place in the byline merely for performing activities that alone do not qualify for authorship (eg, cooperating in a study, collecting data, attending a working conference, lending technical assistance). However, performing any of those activities in addition to writing or critically revising the manuscript and approving the version to be published would be sufficient to merit authorship (see 5.1.1, Authorship: Definition, Criteria, Contributions, and Requirements). Editors and authors should assess the need to publish lengthy lists of authors and other group participants on an individual basis, and journals should publish their policies about group authorship in their instructions for authors.⁹

5.1.9.1 Citation of Articles With Group Authors.

Articles with authors from a large group have been difficult to locate in bibliographic databases and have resulted in citation errors and miscalculated citation statistics.^54,55,56 To help resolve these problems, the following has been recommended^9,58:

Group author articles should identify named individual authors who accept responsibility for specific articles.

Each group author article should clearly identify all individual authors (preferably full names, but last names and initials are acceptable) as well as the complete name of the group, whether they appear in the byline, Acknowledgement, or Article Information section.

Individual authors should be distinguished from other contributors (eg, collaborators) and participants who are not authors.

The names of individual authors and the group name should be formatted and coded for easy identifiability, searching, and retrieval of the article in print and online and in bibliographic databases (see Box 5.2-1 in 5.2, Acknowledgments).

Each group author article should clearly indicate a preferred citation (eg, in the article or in an online link from the article to a reference manager).

Search results on journal websites should clearly indicate group names as published in article bylines in addition to relevant author information.

Citation standards for group author papers should continue to be developed and followed by journals, bibliographic databases, and authors.

5.1.10 Standards for Electronic Editing and Tagging of Names of Authors, Collaborators, and Group Authors.

To ensure appropriate display in print and online journals as well as indexing, search, and retrieval in bibliometric databases, journals should follow standards for tagging (coding) the names of authors, nonauthor collaborators, and group authors. The National Information Standards Organization and the National Library of Medicine have produced the Journal Article Tag Suite (JATS) to define a set of XML elements and attributes that describe the content and metadata of journal articles.⁵⁹ JATS provides guidance on how to tag authors, nonauthor collaborators, and group authors. Consistent use of tagging within the full-text article will also enable publishers to deposit the correct metadata to databases such as PubMed and Web of Science and will improve data mining and discoverability through various search engines (see 21.0, Editing, Proofreading, Tagging, and Display).

Principal Author: Annette Flanagin, RN, MA

Acknowledgment

I thank the following for review and helpful comments: Howard Bauchner, MD, JAMA and JAMA Network; Carissa Gilman, American Cancer Society, Atlanta, Georgia; Timothy Gray, PhD, JAMA Network; Iris Y. Lo, JAMA Network; Lou Knecht, formerly of the National Library of Medicine, Bethesda, Maryland; Ana Marušić, MD, PhD, Journal of Global Health and University of Split School of Medicine, Croatia; and Fred Rivara, MD, MPH, JAMA Network Open and University of Washington, Seattle.

References

1.Pope A. An Essay on Criticism. 1711:part II, lines 412-413.

2.Hewitt RM. Exposition as applied to medicine: a glance at the ethics of it. JAMA. 1954;156(5):477-479.

3.Huth EJ. Guidelines on authorship of medical papers. Ann Intern Med. 1986;104(2):269-274.

4.International Committee of Medical Journal Editors. Guidelines on authorship. BMJ. 1985;291(6947):722.

5.International Committee of Medical Journal Editors. Recommendations for the conduct, reporting, editing, and publication of scholarly work in medical journals. Updated December 2018. Accessed July 7, 2019. http://www.icmje.org/recommendations

6.US National Library of Medicine. Fact Sheet. Authorship in MEDLINE. Updated June 22, 2018. Accessed July 7, 2019. https://www.nlm.nih.gov/bsd/policy/authorship.html

7.Council of Science Editors. Scientific Style and Format: The CSE Manual for Authors, Editors, and Publishers. 8th ed. University of Chicago Press/Council of Science Editors; 2014.

8.Council of Science Editors. White Paper on Publication Ethics. Updated May 2018. Accessed July 7, 2019. https://www.councilscienceeditors.org/resource-library/editorial-policies/white-paper-on-publication-ethics/

9.Fontanarosa P, Bauchner H, Flanagin A. Authorship and team science. JAMA. 2017;318(24):2433-2437. doi:10.1001/jama.2017.19341

10.Rennie D, Yank V, Emanuel L. When authorship fails: a proposal to make contributors accountable. JAMA. 1997;278(7):579-585. doi:10.1001/jama.1997.03550070071041

11.Rennie D, Flanagin A, Yank V. The contributions of authors. JAMA. 2000;284(1):89-91. doi:10.1001/jama.284.1.89

12.Instructions for Authors. JAMA. Updated January 17, 2019. Accessed January 28, 2019. https://jamanetwork.com/journals/jama/pages/instructions-for-authors

13.Brand A, Allen L, Altman M, Hlava M, Scott J. Beyond authorship: attribution, contribution, collaboration, and credit. Learned Publishing. 2015;28(2):151-155.

14.CASRAI. CRediT. Accessed October 11, 2019. https://casrai.org/credit/

15.ORCID. Distinguishing yourself in three easy steps. Accessed October 11, 2019. https://orcid.org

16.Taichman DB, Sahni P, Pinborg A, et al. Data sharing statements for clinical trials: a requirement of the International Committee of Medical Journal Editors. JAMA. 2017;317(24):2491-2492. doi:10.1001/jama.2017.6514

17.Rennie D, Flanagin A. Authorship! authorship! guests, ghosts, grafters, and the two-sided coin. JAMA. 1994;271(6):469-471. doi:10.1001/jama.1994.03510300075043

18.Flanagin A, Rennie D. Acknowledging ghosts. JAMA. 1995;273(1):73. doi:10.1001/jama.1995.03520250089041

19.Yank V, Rennie D. Disclosure of researcher contributions: a study of original research articles in The Lancet. Ann Intern Med. 1999;130(8):661-670.

20.Flanagin A, Carey LA, Fontanarosa PB, et al. Prevalence of articles with honorary authors and ghost authors in peer-reviewed medical journals. JAMA. 1998;280(3):222-224. doi:10.1001/jama.280.3.222

21.Wislar J, Flanagin A, Fontanarosa PB, DeAngelis CD. Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey. BMJ. 2011;343:d6128. doi:10.1136/bmj.d6128

22.Davidoff F, for the Council of Science Editors Taskforce on Authorship. Who’s the author? problems with biomedical authorship, and some possible solutions. February 2000. Accessed December 21, 2018. https://www.councilscienceeditors.org/wp-content/uploads/v23n4p111-119.pdf

23.Marušić A, Bates T, Anic A, Marusic M. How the structure of contribution disclosure statements affects validity of authorship: a randomized study in a general medical journal. Curr Med Res Opin. 2006;22(6):1035-1044.

24.Greenland P, Fontanarosa PB. Ending honorary authorship. Science. 2012;337(6098):1019. doi:10.1126/science.1224988

25.Smith J. Gift authorship: a poisoned chalice? BMJ. 1994;309(6967):1456-1457.

26.Brennan TA. Buying editorials. N Engl J Med. 1994;331(10):673-675.

27.Ross JS, Hill KP, Egilman DS, Krumholz HM. Guest authorship and ghostwriting in publications related to rofecoxib: a case study of industry documents from rofecoxib litigation. JAMA. 2008;299(15):1800-1812. doi:10.1001/jama.299.15.1800

28.DeAngelis CD, Fontanarosa PB. Impugning the integrity of medical science: the adverse effects of industry influence. JAMA. 2008;299(15):1833-1835. doi:10.1001/jama.299.15.1833

29.Mowatt G, Shirran L, Grimshaw JM, et al. Prevalence of honorary and ghost authorship in Cochrane reviews. JAMA. 2002;287(21):2769-2771. doi:10.1001/jama.287.21.2769

30.DeBakey L. Rewriting and the by-line: is the author the writer? Surgery. 1974;75(1):38-48.

31.Hamilton CW, Royer MG; for the AMWA 2002 Task Force on the Contributions of Medical Writers to Scientific Publications. AMWA position statement on the contributions of medical writers to scientific publications. AMWA J. 2003;18(1):13-16.

32.Information for Authors. Anesthesiology. 2018. Accessed October 21, 2018. http://anesthesiology.pubs.asahq.org/public/instructionsforauthors.aspx

33.Information for Authors. Neurology. Accessed October 21, 2018. http://www.neurology.org/authorship-and-disclosures

34.Information for Authors. The Lancet. Updated April 2018. Accessed October 21, 2018. http://www.thelancet.com/lancet-information-for-authors

35.Morgan P. An Insider’s Guide for Medical Authors and Editors. ISI Press; 1986.

36.Lock S. Signed editorials. BMJ. 1981;283(6296):876.

37.The Lancet. Signed—The Lancet. Lancet. 1993;341(8836):24.

38.Fye WB. Medical authorship: traditions, trends, and tribulations. Ann Intern Med. 1990;113(4):317-325.

39.US National Library of Medicine. Number of authors per MEDLINE/PubMed citation. Updated September 7, 2018. Accessed October 11, 2019. https://www.nlm.nih.gov/bsd/authors1.html

40.Aboukhalil R. The rising trend in authorship. Winnower. 2014;3:e141832.26907. doi:10.15200/winn.141832.26907

41.Burman KD. “Hanging from the masthead”: reflections on authorship. Ann Intern Med. 1982;97(4):602-605.

42.Epstein RJ. Six authors in search of a citation: villains or victims of the Vancouver convention? BMJ. 1993;306(6880):765-767.

43.US National Library of Medicine. MEDLINE/PubMed data element (field) descriptions. Updated July 23, 2018. Accessed October 21, 2018. https://www.nlm.nih.gov/bsd/mms/medlineelements.html

44.Schmidt RH. A worksheet for authorship of scientific articles. Bull Ecol Soc Am. 1987;68(1):8-10.

45.Davis PJ, Gregerman RI. Parse analysis: a new method for the evaluation of investigators’ bibliographies. N Engl J Med. 1969;281(18):989-990.

46.Chambers R, Boath E, Chambers S. The A to Z of authorship: analysis of influence of initial letter of surname of order of authorship. BMJ. 2001;323(7327):1460-1461.

47.Riesenberg D, Lundberg GD. The order of authorship: who’s on first? JAMA. 1990; 264(14):1857. doi:10.1001/jama.1990.03450140079039

48.Authorship & contributorship. BMJ. Accessed October 21, 2018. http://www.bmj.com/about-bmj/resources-authors/article-submission/authorship-contributorship

49.Committee on Publication Ethics. Accessed October 11, 2019. https://publicationethics.org/

50.Martinson BC, Anderson MS, de Vries R. Scientists behaving badly. Nature. 2005;435:737-738. doi:10.1038/435737a

51.Reinisch JF, Li WY, Yu DC, Walker JW. Authorship conflicts: a study of awareness of authorship criteria among academic plastic surgeons. Plast Reconstr Surg. 2013;132(2):303e-310e. doi:10.1097/PRS.0b013e3182958b5a

52.Dance A. Authorship: who’s on first? Nature. 2012;489:591-593. doi:10.1038/nj7417-591a

53.Marušić A, Bošnjak L, Jerončić A. A systematic review of research on the meaning, ethics and practices of authorship across scholarly disciplines. PLoS ONE. 2011;6(9):e23477. doi:10.1371/journal.pone.0023477

54.Flanagin A, Fontanarosa PB, DeAngelis CD. Authorship for research groups. JAMA. 2002;288(24):3166-3168. doi:10.1001/jama.288.24.3166

55.Dickersin K, Scherer R, Suci EST, Gil-Montero M. Problems with indexing and citation of articles with group authorship. JAMA. 2002;287(21):2772-2774. doi:10.1001/jama.287.21.2772

56.Errors in citation statistics. Nature. 2002;415(6868):101.

57.Cherfas J. With missing citations reported: Nature genome paper jumps. Sci Watch. 2002;13(1):8.

58.Flanagin A, Wrobel P, Barbour V, et al. CSE recommendations for group-author articles in scientific journals and bibliometric databases. Accessed October 21, 2018. http://www.councilscienceeditors.org/resource-library/editorial-policies/cse-policies/approved-by-the-cse-board-of-directors/cse-recommendations-for-group-author-articles-in-scientific-journals-and-bibliometric-databases/

59.National Center for Biotechnology Information. Journal Article Tag Suite. Accessed October 11, 2019. https://jats.nlm.nih.gov/faq.html