Cooperation between research institutions and journals on research integrity cases: guidance from the committee on publication ethics (cope).

Institutions and journals both have important duties relating to research and publication misconduct. Institutions are responsible for the conduct of their researchers and for encouraging a healthy research environment. Journals are responsible for the conduct of their editors, for safeguarding the research record, and for ensuring the reliability of everything they publish. It is therefore important for institutions and journals to communicate and collaborate effectively on cases relating to research integrity. To achieve this, we make the following recommendations. Institutions should: have a research integrity officer (or office) and publish their contact details prominently; inform journals about cases of proven misconduct that affect the reliability or attribution of work that they have published; respond to journals if they request information about issues, such as disputed authorship, misleading reporting, competing interests, or other factors, including honest errors, that could affect the reliability of published work; initiate inquiries into allegations of research misconduct or unacceptable publication practice raised by journals; have policies supporting responsible research conduct and systems in place for investigating suspected research misconduct.Journals should: publish the contact details of their editor-in-chief who should act as the point of contact for questions relating to research and publication integrity; inform institutions if they suspect misconduct by their researchers, and provide evidence to support these concerns; cooperate with investigations and respond to institutions' questions about misconduct allegations; be prepared to issue retractions or corrections (according to the COP E guidelines on retractions) when provided with findings of misconduct arising from investigations; have policies for responding to institutions and other organizations that investigate cases of research misconduct.

INTRODUCTION

Research institutions (such as universities) and scholarly journals have important duties and common interests in terms of research and publication integrity. Institutions are responsible for the conduct of their researchers and for encouraging a healthy research environment that fosters research integrity. Journals are responsible for the conduct of their editors, for safeguarding the research record, and for ensuring the reliability of everything they publish. Ensuring research and publication integrity requires institutions and journals to cooperate with each other on all aspects of research and publication integrity. Institutions and journals should also promote best practice among researchers, authors, reviewers, and editors (e.g. by policy development and training). Journals should make efforts to detect misconduct before publication (e.g. by screening for plagiarism). Institutions should investigate possible misconduct and journals should correct or retract findings that are invalid or unreliable (because of misconduct or honest errors) to prevent readers from being misled by them.

COPE (the Committee on Publication Ethics) provides a Forum for its member journals to seek advice from other members on troublesome cases. Based on this experience, we have become aware that editors sometimes face difficulties when trying to work with institutions on cases of possible misconduct (1, 2). From the literature, and discussions with institutions, we are also aware that editors do not always respond appropriately when informed by institutions about research misconduct findings (3).

Given the importance of journal editors and research institutions collaborating effectively on cases relating to publication ethics and research integrity, COPE has developed the following guidelines in consultation with the individuals and institutions listed at the end of the document.

SCOPE

This document focuses on the investigation of possible misconduct. However, this is not to belittle the important roles of prevention, education, etc. outlined above. Ideally, journal and institutional policies should cover all aspects. Journal policies should cover not only their responses to misconduct but also to genuine errors; this is described in more detail in the COPE guidelines on retractions (4).

Similarly, we recognise that other parties, notably funders, have an important role in fostering research integrity and should be informed about research or publication misconduct relating to research they have funded. These guidelines focus on the roles of institutions and journals but we hope they may help funders to develop their own policies to foster research integrity and responsible conduct of research in collaboration with institutions and journals.

BACKGROUND PRINCIPLES

The COPE Code of Conduct for Journal Editors (Clause 11.4) notes that, in cases of suspected or alleged research or publication misconduct ‘editors should first seek a response from those suspected of misconduct. If they are not satisfied with the response, they should ask the relevant employers, or institution … to investigate’ (5).

COPE therefore advises that investigations into possible misconduct should generally be undertaken by the researcher’s institution and not by editors. If a journal has published unreliable or fraudulent information, the editor has a duty to correct or retract this. However, responsibility for disciplining researchers and ensuring they do not commit further misconduct lies with their institution/employer. Therefore, even when faced with apparently strong evidence of misconduct (e.g. plagiarism or inappropriate image manipulation), and a clear need to correct the published record, editors should liaise with institutions and ensure they are informed.

Journals also need to work with institutions when disputes arise between researchers (e.g. about authorship). As with research misconduct, journals are not in a position to investigate or resolve such disputes, but should refer them to the relevant institution(s).

The following guidelines are therefore based on the principle that institutions have responsibilities for the conduct of their researchers, which include investigating possible misconduct and applying appropriate sanctions, while journals are responsible for what they publish.

While these guidelines encourage exchange of information between institutions and journals regarding cases of possible and proven misconduct, we recognize that full disclosure may sometimes be restricted by considerations of confidentiality (e.g. to protect the identity of a whistleblower), conventions about confidential communication (e.g. peer review comments), and legal considerations.

DEFINING MISCONDUCT

Several definitions of research misconduct are available and are used by different organizations for various purposes. Difficulties sometimes arise when an institution adopts a narrow or strict definition of serious misconduct which does not include practices that, while falling short of this definition, may nevertheless distort the research record. In such instances an institution may find a researcher not guilty of misconduct yet a journal may consider that a correction or retraction is warranted to safeguard readers (e.g. to alert them to redundant publication).

In these guidelines we do not attempt to define serious or lesser types of misconduct, or so-called ‘questionable practices’, but we use the term misconduct in its broadest sense to include any practice that may affect the reliability of the research record in terms of findings, conclusions, or attribution.

RECOMMENDATIONS FOR COOPERATION BETWEEN RESEARCH INSTITUTIONS AND JOURNALS

Points of contact

To facilitate communication, research institutions should designate an individual or office with responsibility for research integrity and dealing with misconduct allegations. Contact details of the research integrity officer(s) should be published prominently on the institution’s website. This person (or office) should be free from conflicts of interest in relation to individual cases (i.e. have no involvement with any researcher or project being investigated). If a suitable individual without conflicts of interest cannot be found, it may be necessary to involve an external person in investigations.

Likewise, journals should publish contact details of their editor(s)-in-chief who should act as the point of contact for questions relating to research and publication integrity. COPE also recommends that journals should appoint an ombudsperson to adjudicate in complaints that cannot be resolved internally or that relate to the conduct of the editor (5).

Informing each other about cases of research and publication misconduct

If an institution investigates a case of misconduct by one of its researchers and finds misconduct that affects the reliability or attribution of published work (e.g. fabrication or plagiarism), the institution should inform the editor(s) of any journal that has published the affected work. If a case is investigated by another organization (e.g. a national body), the institution should pass on the findings to the relevant journal(s). Institutions should also be prepared to answer editors’ questions about any investigation or its findings that are necessary to determine the appropriate response (e.g. retraction or Expression of Concern).

Institutions should also notify journal editors and answer their questions in cases of inappropriate publication practices such as authorship misattribution, redundant publication, duplicate submission, failure to disclose competing interests, or misleading reporting (even if these fall outside the institution’s definition of research misconduct). Institutions should also encourage researchers to inform journals about honest errors likely to affect the reliability of published work.

Editors should cooperate with investigations and respond to institutions’ questions about misconduct allegations.

Communication between institutions and journals

Institutions should:

* acknowledge receipt of communications from journal editors and respond promptly to allegations of research misconduct;

* inform editors (or respond to enquiries from editors) about on-going investigations into misconduct likely to affect the validity of publications (e.g. to confirm that formal investigations are underway – following initial assessment of the allegation – and state the likely duration) so that editors can issue an Expression of Concern if necessary;

* share the findings of misconduct investigations with journals so that the editor(s) can determine whether retractions or corrections are required;

* ensure that all communications relating to misconduct investigations (such as press briefings and notifications to journals) are clear, accurate and complete.

Both institutions and editors should generally ensure that communications relating to ongoing misconduct investigations are kept confidential between parties; however editors may use an Expression of Concern to inform readers about serious allegations likely to affect the reliability or integrity of a publication.

Journals should:

acknowledge receipt of communications from institutions and respond promptly to findings of research misconduct;

inform institutions about possible misconduct and provide evidence to support these concerns (e.g. analysis of text similarity in cases of suspected plagiarism, or evidence of inappropriate image manipulation);

investigate allegations of misconduct by researchers acting as peer reviewers for the journal (e.g. that reviewers plagiarized another researcher’s work), follow the COPE flowchart on such cases, and liaise with the institution as required;6

follow the COPE guidelines on retractions (4).

Responding to journal concerns about research integrity or publication practices

Institutions should initiate inquiries into allegations of research misconduct or unacceptable publication practice raised by journal editors.

Where possible, journals should provide evidence to support allegations of misconduct or questionable practices (e.g. copies of overlapping publications, evidence of plagiarism). However, editors may be obliged to protect the identity of whistleblowers or of peer reviewers.

Institutions should respond promptly and constructively to editors’ requests for clarification of authorship or data ownership. Editors have to rely on the honesty of researchers in declaring their contributions to a project. Journals cannot be expected to adjudicate in authorship disputes and therefore rely on institutions to arbitrate in such matters. Editors should respond to authorship adjudications supplied by institutions and, where necessary, issue corrections (i.e. by adding or removing authors from the by-line of published or submitted articles). Editors should follow the relevant COPE flowcharts in such cases.7

Cases involving multiple institutions or journals

In collaborative research involving multiple institutions, one institution should be nominated to coordinate investigations and act as the point of contact unless there is an obvious lead institution (e.g. that administers the grant or employs the researchers). Disputes between institutions over authorship or data ownership may require adjudication by an independent arbitrator agreeable to all parties.

Cases of plagiarism, breach of copyright or redundant publication usually involve several journals who should therefore cooperate with each other and share information as required (e.g. about submission dates and copyright transfer agreements) to resolve the issues.

Ensuring the reliability of the published research record

If an institutional investigation or disciplinary hearing into research misconduct recommends that a researcher seeks a retraction or correction, the institution should inform the editor(s).

Similarly, journals should be prepared to issue retractions or corrections when provided with findings of misconduct arising from appropriate investigations (4).

Publications should be retracted if they prove unreliable (for whatever reason), but if only a small part of the publication is affected (while the majority of findings and conclusions are valid) then a correction should be published.

Expressions of Concern may be published to alert readers to an ongoing investigation into actions likely to affect the reliability of published findings; they should be followed by a retraction, exoneration or correction when the investigation has concluded. Expressions of Concern should not be viewed as ‘milder’ versions of retractions.

Journals should also be prepared to publish corrections or retractions when honest errors are admitted. Retraction statements should include the reasons for the retraction and should distinguish between cases of misconduct and honest error to encourage researchers to report errors when they occur and ensure no stigma is attached to this (4).

Journal and institutional policies

Institutions should have policies supporting research integrity and good practice (e.g. for authorship), describing research misconduct (e.g. data fabrication and plagiarism) and unacceptable publication practices (e.g. redundant publication, inappropriate authorship, and use of confidential material by reviewers), and how these are handled (8). Such policies should be publicised and enforced within the institution.

Institutions should encourage researchers to inform journals if errors are discovered in published work.

Journals should have policies about how they handle suspected misconduct and how they respond to institutions and other organizations that investigate cases of research misconduct (e.g. national bodies).

Encouraging good practice

Journals should provide clear advice to authors and reviewers and have appropriate policies for editors and staff relating to all aspects of publication ethics (9, 10). Journals should inform authors and readers how they handle cases of suspected research misconduct or unacceptable publication practices.

Institutions should include training in good publication practices as part of their programmes of education in research integrity.

Institutional leaders and journal editors should aim to create research environments that encourage good practice and should lead by example in their own publication practices.

Institutions should ensure that their systems for appointments and assessing research productivity do not create incentives for unacceptable practices, such as redundant publication and guest authorship.

Investigating previous publications

Research and publication misconduct may not be an isolated incident. In many cases, when serious misconduct comes to light, investigation of the researcher’s earlier work reveals further problems. Therefore, when a researcher is found to have committed serious misconduct (such as data fabrication, falsification or plagiarism) the institution should review all the individual’s publications, including those published before the proven misconduct took place. In such cases, it may be necessary to alert previous employers to enable them to review work carried out by the discredited researcher when working at their institution, to determine the reliability of publications arising from that work (for an example of this see reference 11).