Scientific Misconduct and Medical Journals

According to the US Department of Health and Human Services, “Research misconduct means fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results.”¹ Other important irregularities involving the biomedical research process include, but are not limited to, ethical issues (eg, failure to obtain informed consent, failure to obtain appropriate approval from an institutional review board, and mistreatment of research participants), issues involving authorship responsibilities and disputes, duplicate publication, and failure to report conflicts of interest. When authors are found to have been involved with research misconduct or other serious irregularities involving articles that have been published in scientific journals, editors have a responsibility to ensure the accuracy and integrity of the scientific record.^2,3

Although not much is known about the prevalence of scientific misconduct, several studies with limited methods have estimated that the prevalence of scientists who have been involved in scientific misconduct ranges from 1% to 2%.^4-6 During the last 5 years, JAMA and the JAMA Network journals have published 12 notices of Retraction about 15 articles (including recent Retractions of 6 articles by the same author)⁷ and 6 notices of Expression of Concern about 9 articles. These notices were published primarily because the original studies were found to involve fabrication or falsification of data that invalidated the research and the published articles; in some cases, postpublication investigations could not provide evidence that the original research was valid. Since 2015, JAMA and the JAMA Network journals also have retracted and replaced 12 articles for instances of inadvertent pervasive error resulting from incorrect data coding or incorrect analyses and without evidence of research misconduct.⁸ During the same period, 1021 correction notices have been published in these journals. The JAMA Network policies regarding corrections and retraction with replacement have been published previously.^8,9 In this Editorial, the focus is on a more complex and challenging issue—scientific misconduct involving fabrication, falsification, and plagiarism in the reporting of research.¹

JAMA and the JAMA Network journals receive numerous communications from readers, such as letters to the editor and emails, that are critical of the published content. Most of the critiques involve matters of interpretation, the need for clarification of content, or differences of opinion; some address ethical concerns, some are frivolous complaints, and some include calls for retraction. However, typically 10 to 12 times each year these journals receive allegations of scientific misconduct. All matters related to allegations of scientific misconduct for articles published in JAMA and the JAMA Network journals are evaluated and managed by the senior staff of JAMA including the editor in chief of JAMA, executive editor, executive managing editor, and the editorial counsel. This provides a consistent process for dealing with potential scientific misconduct. If the allegation involves an article published in a network journal, the editor in chief of that journal is involved and kept informed about the progress of the investigation. In addition, when necessary, additional expertise is obtained.

Allegations of scientific misconduct brought to journals are challenging and time-consuming for the authors, for editors, and potentially for institutions. The first step involves determining the validity of the allegation and an assessment of whether the allegation is consistent with the definition of research misconduct. In some cases, when authors are accused of misconduct, the criticism represents a different interpretation of the data or disagreement with the statistical approach used, rather than scientific misconduct. This initial step also involves determining whether the individuals alleging misconduct have relevant conflicts of interest. In some cases, it appears that financial interests and strongly held views (intellectual conflict of interest) may have led to the allegation. This does not mean that potential conflicts of interest on the part of the persons bringing the allegations preclude the possibility of scientific misconduct on the part of the authors, but rather, evaluation of conflict of interest is part of the assessment process.

If scientific misconduct or the presence of other substantial research irregularities is a possibility, the allegations are shared with the corresponding author, who, on behalf of all of the coauthors, is requested to provide a detailed response. Depending on the nature of the allegation, it can take months for some authors to respond to the concerns. After the response is received and evaluated, additional review and involvement of experts (such as statistical reviewers) may be obtained. In the majority of cases, the authors’ responses and additional information provided regarding the concerns raised are sufficient to make a determination of whether the allegations raised are likely to represent misconduct. For cases in which it is unlikely that misconduct has occurred, clarifications, additional analyses, or both, published as letters to the editor, and often including a correction notice and correction to the published article are sufficient. To date, JAMA has had very few disagreements with individuals making allegations of scientific misconduct, although some have been critical of the time it has taken for JAMA and other journals to resolve an issue of alleged scientific misconduct.^10-12

However, if the authors’ responses to the allegations raised are unsatisfactory or unconvincing, or if there is any doubt as to whether scientific misconduct has occurred, additional information and investigation are usually necessary, and the appropriate institution is contacted with a request to conduct a formal evaluation. At that time, and depending on the nature of the allegations, the journal may publish a notice of Expression of Concern about the published reports in question, indicating that issues of validity or other concerns have arisen and are under investigation.²

Involving institutions is done with great care for several reasons. First, even just an allegation of misconduct can harm the reputation of an individual. Individuals involved in such allegations have expressed this concern and notification of an institution increases the level of scrutiny directed toward the involved person. In these cases, institutions are responsible for ensuring appropriate due process and confidentiality, based on their policies and procedures. Second, just as JAMA receives allegations of scientific misconduct and research irregularities, so too do institutions. It simply is not possible for every institution to conduct a detailed investigation of every allegation received; thus, JAMA and the JAMA Network journals ensure that institutions are only asked to be involved after a determination has been made that scientific misconduct is a possibility and for which the authors have not adequately responded to the concerns raised.

Institutions are expected to conduct an appropriate and thorough investigation of allegations of scientific misconduct. Some institutions are immediately responsive, acknowledging receipt of the letter from the journal describing the concerns, and quickly begin an investigation. In other cases, it may take time to identify the appropriate institutional individuals to contact, and even then, many months to receive a response. Some institutions appear well-equipped to conduct investigations, whereas other institutions appear to have little experience in such matters or fail to conduct adequate investigations¹³; these institutions can take months to years to provide JAMA with an adequate response. In some cases involving questions of misconduct from outside of the United States, institutions have indicated that further investigation must wait until numerous legal issues are resolved, further delaying a response.

The type of investigation an institution conducts depends on the specific allegations and the institutional policies and procedures. In some cases, the investigation has involved reviewing the data, the article and related articles, and the analysis. In other cases, the investigation has involved reanalysis by the authors, or independent statistical analysis by a third party not involved in the initial study. Other cases have involved investigation of ethical issues related to the research, such as appropriate ethical review and approval of the study, informed consent for study participants, and notification of study participants about information related to risks of an intervention. No single approach is appropriate in all cases, but rather it depends on the specific allegation. In 2017, a group of representatives who deal with scientific misconduct, including university and institutional leaders and research integrity officers, federal officials, researchers, journal editors, journalists, and attorneys representing respondents, whistle-blowers, and institutions, examined best and failed practices related to institutional investigation of scientific misconduct.¹⁴ The group developed a checklist that can be used by institutions to follow reasonable standards to investigate an allegation of scientific misconduct and to provide an appropriate and complete report following the investigation.¹⁴

JAMA editors request institutions to provide periodic updates on the status of an investigation, and once the investigation is completed, institutions are expected to provide the editors with a detailed report of their findings. For cases in which misconduct has been identified, the institution and the authors may recommend and request retraction of the published article. In other cases, based on the report of the investigation from the institution, the journal editors make the determination of what actions are needed, such as whether an article should be retracted; or when a notice of Expression of Concern had been posted, whether it should be subsequently followed by a notice of Retraction. In each case, the notices are linked to and from the original article, and retracted articles are clearly watermarked as retracted so that readers and researchers are properly alerted to the invalid nature of the original articles.²

Allegations of scientific misconduct are challenging. Not all such allegations warrant investigation, but some require extensive evaluation. JAMA reviews its approach to allegations of scientific misconduct on a regular basis to ensure that the process is timely, objective, and fair to authors and their institutions, and results in evidence that will directly address the allegations of misconduct. Ultimately, authors, journals, and institutions have an important obligation to ensure the accuracy of the scientific record. By responding appropriately to concerns about scientific misconduct, and taking necessary actions based on evaluation of these concerns, such as corrections, retractions with replacement, notices of Expression of Concern, and Retractions, JAMA and the JAMA Network journals will continue to fulfill the responsibilities of ensuring the validity and integrity of the scientific record.

Corresponding Author: Howard Bauchner, MD, Editor in Chief, JAMA, 330 N Wabash Ave, Chicago, IL 60611 (howard.bauchner@jamanetwork.org).

Published Online: October 19, 2018. doi:10.1001/jama.2018.14350

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Ms Flanagin reports serving as an unpaid member of the board of STM: International Association of Scientific, Technical, and Medical Publishers. No other disclosures were reported.