Response: Letter to the Editor regarding International multi‐stakeholder consensus statement on post‐publication integrity issues in randomized clinical trials by Cairo Consensus Group

Abstract

We thank Professors Mol, Schroll, and ter Riet for their engagement with our international Consensus statement on post-publication integrity issues in randomized clinical trials.1 Our ongoing initiative is intended to foster constructive dialogue on an issue of growing importance. We wish to reassure colleagues that our work is continuing, and we welcome constructive suggestions that may help with our endeavor. This letter refers to its intention of highlighting what it considers “missing”. We appreciate this, as our paper stated that “clarification is a continuous exercise”.1 We take this opportunity to address the points made in the letter to the Editor in the same sequence as presented. The Cairo Consensus document is a formal research study, not an opinion article. The strength of its statements should be judged by considering the agreement data presented in the paper. When screening complaints, there is a responsibility for investigators to examine the validity or veracity of the allegations.2, 3 The Committee on Publication Ethics recommends that editors should “consider getting a second opinion from another reviewer” and “query the author of the critique and ask what support/evidence they have for the claims made”.2, 3 The “red flags” have previously been published.4 The consensus concerning the relevant statement was at a 69% agreement level, passing close to the threshold in the first Delphi round. The need for distinguishing between frivolous or ill-intended complaints and bona fide allegations is well recognized in the literature.2-5 Anonymity request by the complainant has relevance within institutional whistleblowing, particularly where employees with power can adversely affect the careers of vulnerable colleagues. The matter is quite different when dealing with issues related to the correction of the scientific record, where the complainants are often powerful and the trialists vulnerable. The consensus concerning the relevant statement was at a 75% agreement level, passing well above the threshold in the second Delphi round. The statement advises that the merit of any anonymity request by a complainant is best judged by the journal Editor-in-Chief receiving the complaint. The Consensus covers integrity assessment validation in four statements, all of which passed the agreement threshold in the first Delphi round. Two statements are concerned with future research, and one statement each concerning complaints and investigations. Validation of integrity tests is key to differentiating between true and false positives and negatives.6 If the tests, individually or in combination, are of unknown accuracy/validity in demonstrating misconduct, how are editors and investigators to scrutinize the complaint's veracity? The general sentiment expressed by the authors of the letter aligns with the validation-related statements agreed upon in the Consensus. However, it is unclear why they wish for academics, who aim for scientific record correction through their complaints or correspondence, to hide behind anonymity and avoid scientific scrutiny when making allegations. The Consensus encourages transparency and accountability. The letter highlights the Consensus statement that “peer reviewers and other relevant stakeholders, including authors and readers, need to continuously undertake specific RCT integrity training”, and in the same breath, it states that “Not a single statement appears to apply to authors” [quoted text italicized to emphasize similarity]. We regard this remark not as misguided criticism, which in fact it is, but as an attempt to be constructive. The term “authors” appears at least a dozen times within the Consensus statement. It concurs with the letter's emphasis on authorial responsibility, which applies to both trialists and complainants. The Consensus statement presents a whole section on journals' instructions, which covers instructions to authors. The suggestion by the letter's authors that “some statements seem redundant, e.g. prospective registration” overlooks the fact that even well-publicized standards remain inconsistently applied, such as adherence to the prospective registration standard.7 Reaffirming these principles through the Cairo Consensus will hopefully help to embed them as a universal research practice going forward. The letter points out that one of the co-authors has a retracted paper. Their claim that our assertion concerning the contributor retraction record within our Consensus is “incorrect” is false. For accuracy, it is necessary to outline the chronology. The Consensus was registered on 24 August 2022, and its final development meeting concluded on 14 February 2023. The first submission of the resulting manuscript was made on 10 December 2023 to BJOG, where it was not accepted. The retraction in question was not issued until 28 February 2024, more than a year after the Consensus had been finalized. When the retraction became known, the Consensus manuscript had already been submitted to BJOG on 10 December 2023 with its authorship finalized. Following rejection, it was transferred as is to IJGO. At the time of both consensus development and its first submission, therefore, our statement that “none of the authors nor those acknowledged have retracted trials for research integrity flaws” was entirely correct. For the avoidance of any doubt, the dates of the Consensus registration and final meeting are publicly known, as they are documented within the published paper, and the date of retraction is publicly known too. This information was available to the authors of the letter. We also want readers to note that, following the retraction, the concerned co-author has stepped down from the Cairo Consensus and is not part of the authorship of the subsequent Consensus concerning randomized trial ethics and consent.8 Finally, as the letter highlights the importance of retraction records in assessing credibility, we believe transparency should apply equally to all participants in this scientific dialogue. We note that the letter's lead author has also recently incurred a retraction regarding a randomized clinical trial,9 and another trial has a published commentary indicating potential trustworthiness issues.10 We hope that all stakeholders, regardless of their position in this debate, continue to strive toward the rigorous standards necessary for correcting the scientific record. In conclusion, the issues raised in the letter and our response rebutting them demonstrate precisely why complaints and correspondence need to be assessed for their veracity. We welcome the correspondents' contribution to this ongoing discussion and view it as part of the iterative process by which research integrity will be strengthened. All authors contributed equally in drafting and reviewing this letter. Yacoub Khalaf, Khalid S. Khan, and Mohamed Fawzy have no conflicts of interest to declare beyond those listed in the Supporting Information of the original consensus publication.