Confidentiality During Editorial Evaluation and Peer Review and After Publication
Confidentiality During Editorial Evaluation and Peer Review and After Publication
Strict confidentiality regarding the review and evaluation of submitted manuscripts and all relevant correspondence and other forms of communication is essential to the integrity of the editorial process (see 6.1, Editorial Assessment and Processing, Editorial Assessment). Authors must feel free to submit manuscripts that contain their unique ideas and information that may affect their reputations or careers or that may be proprietary. Thus, editors and reviewers have an ethical duty to keep information about a manuscript confidential, and authors have a right to expect that confidentiality will be maintained.3-7 Policies supporting the confidential nature of the peer review and editorial processes are well described by the International Committee of Medical Journal Editors,3 the Council of Science Editors,4 the World Association of Medical Editors,5 and the UK Committee on Publication Ethics.6 The very existence of a submission should not be revealed (by either confirmation or denial) to anyone other than the editors, editorial staff, peer reviewers, and necessary publishing staff (ie, those essential to producing the journal but not others such as sales and marketing staff), unless and until the manuscript is released for publication (see also 5.13, Release of Information to the Public and Journal/Author Relations With the News Media). In addition, editors should refrain from discussing any aspect of the peer review process of a particular manuscript or any unpublished manuscripts with anyone except authors, reviewers, and editorial staff. Even after publication, information and communications about a manuscript, its review (including reviewers' comments), or the editorial process should not be made public without consent of the author, editor, or reviewer (see also “Record Retention Policies for Journals” in 5.6.1, Intellectual Property: Ownership, Access, Rights, and Management: Ownership and Control of Data, and 5.6.7, Intellectual Property: Ownership, Access, Rights, and Management, Copying, Reproducing, Adapting, and Other Uses of Content).
To maintain confidentiality, editors should deny requests or demands for confidential information during editorial evaluation, during peer review, and after publication from any third party, including readers, authors of other manuscripts, owners of the journal, publishing staff other than those essential to producing the journal in print/online, news media, advertisers, governmental agencies, academic institutions, commercial entities, and representatives of those seeking information for use in actual or threatened legal proceedings (see 5.7.3, Confidentiality in Legal Petitions and Claims for Privileged Information). Exceptions to this policy may be made in specific circumstances provided that disclosures are limited and that anyone else given access to confidential information agrees to keep the information confidential. Examples of exceptions include the following:
▪ A prospective author who is invited by an editor to write an editorial commenting on a paper that has not yet been published (Note: such authors should be reminded about the confidential nature of the unpublished paper and not to consult anyone about the paper without prior approval of the editor, including the author of the unpublished paper)
▪ A governmental agency representative consulted by the editor or author on a matter considered a public health emergency or a matter that by regulation requires notification (eg, serious adverse drug event)
▪ An attorney who is asked to advise an editor if legal concerns are raised or who represents the journal in legal proceedings
▪ An institutional or funding authority requested by the editor to investigate an allegation of scientific misconduct related to a manuscript under consideration or a published article (for additional information, see 5.7.2, Confidentiality in Allegations of Scientific Misconduct, and 5.4.4, Scientific Misconduct, Editorial Policy and Procedures for Detecting and Handling Allegations of Scientific Misconduct)
▪ An author’s violation of public journal policy, such as prohibition of covert duplicate publication or failure to disclose conflicts of interest (see also 5.3.2, Duplicate Publication, Editorial Policy for Preventing and Handling Allegations of Duplicate Publication, and 5.5.8, Conflict of Interest, Handling Failure to Disclose Financial Interest)
▪ An author’s refusal to address an editor’s questions about serious ethical concerns, such as whether research participants provided appropriate informed consent or whether a study was appropriately reviewed and approved, or waived for approval, by an independent ethics committee (see also “Reports of Unethical Studies” in 5.8.1, Protecting Research Participants' and Patients' Rights in Scientific Publication, Ethical Review of Studies and Informed Consent)
Journals do not own or have licenses to unpublished works (because copyright and publication licenses are typically transferred in the event of publication); thus, editors should not keep print or electronic copies of rejected manuscripts. Copies should be returned to the author or destroyed. However, a journal may choose to keep a copy of a rejected manuscript for a predetermined, limited period if it has a policy that allows for author appeals of editorial decisions (see also “Record Retention Policies for Journals” in 5.6.1, Intellectual Property: Ownership, Access, Rights, and Management, Ownership and Control of Data). Similarly, reviewers should not keep copies of the manuscripts they are asked to assess. Reviewers should destroy any print and digital copies of manuscripts they have reviewed. Reviewers should not use others' manuscripts as teaching tools or in journal club discussions because doing so would violate confidentiality.
Journals should publish details about the confidential nature of the editorial, peer review, and publication processes in their instructions for authors, and editors should inform all reviewers of the confidential nature of peer review in correspondence to and instructions for reviewers3 (see 5.11, Editorial Responsibilities, Roles, Procedures, and Policies).
Requirements During a Blinded (Masked) Peer Review Process.
Journals should inform reviewers in explicit terms what they mean by “confidentiality,” “confidential information,” and “privileged information” (ie, that not subject to disclosure).8 Journals should also inform reviewers and authors if the review process is single-blinded (ie, only the reviewers' identities are not disclosed), double-blinded (ie, both the reviewers' and the authors' identities are blinded), or open (ie, all author and reviewer identities are disclosed to all). For a detailed discussion of the various mechanisms of peer review (eg, single-blinded, double-blinded, open), see 6.1, Editorial Assessment and Processing, Editorial Assessment. JAMA and the Archives Journals and many other medical and scientific journals use a single-blind review process.
Peer reviewers should receive instructions reminding them to maintain confidentiality when they are invited to review and also after they agree to review (see, for example, the instructions in the Box and also 6.1, Editorial Assessment and Processing, Editorial Assessment). Reviewers should be instructed not to keep copies of manuscripts they have reviewed and to refrain from discussing the information in the manuscript with others. Reviewers should never contact authors directly to discuss their review without explicit permission from the editor.
In some circumstances, a reviewer may wish to enlist the aid of a colleague to assist with the review. Some journals prohibit such consultation, and other journals require that editorial permission be sought in advance of the consultation. If a reviewer is uncertain of a journal’s policy, the reviewer should contact the editorial office. For example, JAMA informs reviewers that they may enlist the aid of colleagues to assist with the review as long as confidentiality is maintained and all other review policies (such as those pertaining to conflicts of interest) are followed. JAMA reviewers are required to inform editors if such consultation has occurred.
After an initial editorial decision (eg, rejection or revision) has been made about a reviewed paper, JAMA provides the corresponding author with copies of the unnamed reviewers' comments. JAMA reviewers are also asked to provide confidential comments to the editor, which include recommendations of acceptance, revision, or rejection; these reviewer-specific recommendations generally are not shared with the authors. However, comments directed to the editor may be summarized or excerpted and included in a letter to the author if necessary.
To provide reviewers with constructive feedback, journal editors should send to reviewers copies of other unnamed reviewers' comments.3 Editors should inform reviewers how their reviews will be used and who will have access to the reviews and to the identities of the reviewers (see 6.1, Editorial Assessment and Processing, Editorial Assessment). In blinded peer review, reviewers have a right to expect that their identities will be protected. Thus, names and identifiers (eg, e-mail addresses, fax numbers, and initials or names) should be removed from reviewers' comments before they are disseminated to the authors or other reviewers.
Occasionally an editor may choose not to send a reviewer’s comments to the author, for example, when comments are considered libelous or hypercritical. Similarly, an editor may choose to remove or mask any unhelpful or derogatory comments from an otherwise valuable review.
Occasionally, reviewers will intentionally identify themselves in their reviews or sign their reviews, even though they know the journal’s peer review process is blinded. Although such identification might imply that the reviewer has waived the right to anonymity, it does not relieve the editor or the reviewer of the duty to maintain confidentiality. If the editor of a journal with a blind review process wishes to disclose the identity of a reviewer who has signed a review, the editor should first contact the reviewer to verify that the reviewer actually intended for her or his identity to be revealed. The editor should remind the reviewer and the author that any communication about the manuscript should occur through the editorial office. If the editor does not want to disclose any reviewer identities, the editor may inform the reviewer that her/his identity or signature will be removed from the review.
Disclosure of Reviewer Identities During Open Review and With Publication.
Some journals, such as the BMJ, have an open review process that encourages reviewers to identify themselves to the authors and other reviewers.9,10 Other journals, such as those published by BioMed Central, also publish signed comments from the reviewers with accepted papers.11 Here again, authors and reviewers should be informed of policies regarding open review and publication of reviewer comments and identities and be reminded that all communications about the peer review and editorial process should be directed to the editor and editorial staff. Journals should clearly describe such policies in instructions for authors and reviewers and in relevant correspondence to authors and reviewers.
Acknowledging and Crediting Reviewers.
An author may want to credit the help of peer reviewers in an acknowledgment. Public acknowledgment of anonymous reviewers is not necessary or informative. However, some journals will honor authors' requests to thank anonymous reviewers.
Many journals also publish the names of individuals who reviewed for the journal during the previous year to thank them publicly. Journals can notify reviewers of this plan in their instructions for reviewers or in relevant correspondence.
Rarely, an editor may receive a request from an author, who has made substantial suggestions for a complete revision, to include a peer reviewer as a coauthor. If the author’s request appears justified, the editor should contact the reviewer to discuss the author’s request and, if appropriate, the author and the reviewer should communicate directly. If such an arrangement is to occur, the request must be made early in the process (ie, before the major revision or complete rewrite) and the reviewer would then need to participate fully in the revision and to meet authorship criteria (see also 5.1.1, Authorship Responsibility, Authorship: Definition, Criteria, Contributions, and Requirements). Such a scenario is unlikely to occur with reports of original research.