The Carnage of Substandard Research During the COVID-19 Pandemic: a Call for Quality Katrina a Bramstedt 1,2
Total Page:16
File Type:pdf, Size:1020Kb
Brief report J Med Ethics: first published as 10.1136/medethics-2020-106494 on 1 October 2020. Downloaded from The carnage of substandard research during the COVID-19 pandemic: a call for quality Katrina A Bramstedt 1,2 1Luxembourg Agency for ABSTRACT issued (tables 2 and 3).ii The source of most of these Research Integrity, Esch- sur- Worldwide there are currently over 1200 research studies incidents is Asia (n=19; 57.6%), with China the Alzette, Luxembourg 2 largest Asian subgroup (n=11; 57.9%). For three Bond University Faculty of being performed on the topic of COVID-19. Many of Health Sciences and Medicine, these involve children and adults over age 65 years. papers, the reason for the removal is unknown; Gold Coast, Queensland, There are also numerous studies testing investigational however, for the others, a range of problems Australia vaccines on healthy volunteers. No research team exist, including data falsification, methodological is exempt from the pressures and speed at which concerns, and concerns about interpretation of data Correspondence to COVID-19 research is occurring. And this can increase and conclusions, as well as authorship and research Professor Katrina A Bramstedt, participant privacy issues (table 4). To date, there Bond University Faculty of the risk of honest error as well as misconduct. To date, Health Sciences and Medicine, 33 papers have been identified as unsuitable for public have been no identified reports of plagiarism or Gold Coast, Queensland, use and either retracted, withdrawn, or noted with data fabrication. Another paper, a preprint from the Australia; concern. Asia is the source of most of these manuscripts USA about COVID-19 antibody seroprevalence, has txbioethics@ yahoo. com (n=19; 57.6%) with China the largest Asian subgroup come under scrutiny for an undisclosed conflict of 6 7 Received 23 May 2020 (n=11; 57.9%). This paper explores these findings and interest, but no official findings have been issued. Revised 1 August 2020 offers guidance for responsible research practice during Accepted 12 August 2020 pandemics. COMPLICATIONS OF PUBLISHING PROBLEMATIC RESEARCH INTRODUCTION There has been a surge of almost 4000 papers As of 7 May 2020, the international clinical trial related to COVID-19 placed on preprint servers registry site, ClinicalTrials.gov , reported 1221 regis- recently.8 These platforms provide a lower level tered studies focused on COVID-19 (table 1).i Of of quality checks compared with a full peer review these, 214 (17.5%) include children as research process by journals. Accordingly, preprint plat- participants, and 1155 (94.6%) include adults age forms do routinely advise their readers not to use 65 years and older. Healthy volunteers are the their content for clinical decision- making, but the cohort for 243 (19.9%) studies, and of these, 13 latter cannot be ruled out, especially in the situa- (5.4%) are vaccine clinical trials. Most studies are tion of a pandemic with high rates of morbidity and being performed in Europe and North America mortality. Patient harm that is significant, perma- where research ethics regulations are robust; nent and irreversible could result from using faulty however, no location is exempt from the pressures research results from preprints as well as published http://jme.bmj.com/ and speed at which COVID-19 research is occur- papers. Published papers are interpreted as vetted ring. This can increase the risk of honest error as via peer review, with an inference of quality and well as misconduct.1 accuracy and clinicians are likely to refer to them According to WHO, the first cases of COVID-19 for guidance. occurred in China in December 2019.2 Other The act of retraction follows a thorough inves- sources indicate the first case could have been in tigation and concern about a paper’s integrity (eg, on September 29, 2021 by guest. Protected copyright. November 2019.3 As the virus rapidly spread from methods, data analysis, data interpretation, data an epidemic to a pandemic, researchers worldwide reporting). While not as severe as a retraction, began the search for treatments and vaccines. Jour- an expression of concern announces to potential nals have been flooded with manuscript submis- readers that a paper should be read with caution sions4 and more journals are providing open access,5 due to potential integrity issues. Both retraction allowing the research to be readily viewed by clini- and expression of concern are markers of integrity cians who can potentially then use the research for issues that reflect on the authors and their insti- clinical decision- making. tutional affiliation. Retracted research can have direct funding costs due to waste and payback requirements, as well as decreased future funding.9 PROBLEMATIC PANDEMIC RESEARCH Sometimes, the integrity of the journal can also be © Author(s) (or their PUBLICATIONS implicated if the matter is associated with a less employer(s)) 2020. No As of 31 July 2020, 19 published articles and 14 commercial re-use . See rights preprints about COVID-19 have been retracted, and permissions. Published ii by BMJ. withdrawn, or an expression of concern has been Retraction Watch (https://retractionwatch.com/ retracted-coronavirus-covid-19-papers) was used as To cite: Bramstedt KA. the source of these articles as they catalogue articles J Med Ethics Epub ahead of from numerous journal indexing databases beyond print: [please include Day MEDLINE, as well as preprint servers. The source i Month Year]. doi:10.1136/ Search terms COVID-19, SARS- CoV-2, severe also mentioned one COVID-19- related conference medethics-2020-106494 acute respiratory syndrome coronavirus 2, 2019- paper that was withdrawn before presentation with nCoV, 2019 novel coronavirus, Wuhan coronavirus unclear explanation. Bramstedt KA. J Med Ethics 2020;0:1–5. doi:10.1136/medethics-2020-106494 1 Brief report J Med Ethics: first published as 10.1136/medethics-2020-106494 on 1 October 2020. Downloaded from can potentially be harmed by a colleague’s misconduct. Research Table 1 COVID-19 clinical trials summary (data from ClinicalTrials. has shown that in such situations, the citations of prior collab- gov, 7 May 2020), n=1221 orators can fall by 8–9%.10 There can be additional harm to Study characteristic No of studies personal relationships amid and across teams as well, and unhap- 11 Study type piness can negatively impact productivity. Patient registry 84 Observational 485 Interventional 736 PREVENTING SUBSTANDARD RESEARCH Research normally occurs at the speed of a marathon, but during Early phase I 15 a pandemic, the pace is more like a sprint.1 The prior slower Phase I 66 pace gave space for reflection on quality, as well as time for Phase II 290 researchers to rest their minds and bodies. During the pandemic Phase III 200 there is the potential for the race of research to occur with fewer Phase IV 47 reflective and rest periods. Less reflection can facilitate missed Funding opportunities for quality checks such as robust reviews of study applications by research ethics committees (RECs) and quality US Federal Government 3 checks by researchers and their supervisors. Less rest can facil- NIH* 11 itate fatigue and mistakes of human error or judgement (eg, Industry 138 taking short cuts rather than performing standard practice).12 Other† 585 During a pandemic, research teams should build in time for Location (study site)‡ reflection such as scheduled spot- checking of data.13 Rejuvena- Europe 459 tion time can also be scheduled such as a team exercise break for North America 294 a brisk walk around the campus (respecting social distancing) or East Asia 87 even a fun musical tune blasted across the public announcement system signalling a 3 min brain break. Middle East 55 RECs cannot be expected to routinely have membership Africa 36 of immunologists, microbiologists and pulmonologists, for South America 34 example, yet these are key topic experts for research protocols South Asia 16 dealing with COVID-19. In an effort to provide high- quality Southeast Asia 11 reviews of these protocols, RECs should create a list of go-to North Asia 6 experts who can be retained as consultants for these reviews. These experts can be readily found through a search of PubMed Pacifica 6 or international scientific professional societies. Some universi- Central America 2 ties also provide experts collated by topic on their websites. It is *US National Institutes of Health. not enough to have a fast-track review process that speeds the †Individuals, universities, organisations. ‡Some studies have more than one study site within and across countries. protocol submissions through the REC; a robust review process that reflects on both science and ethics is needed. Considering http://jme.bmj.com/ that many COVID-19 projects involve epidemiology and contact than robust peer review process or a deficient conflict of interest tracing, reviewers should also pay special attention to the privacy assessment. protection of human participants. Even if only one author’s behaviour is confirmed as the source Research ethics and integrity training should be mandatory for of an article’s retraction or concern, the author’s collaborators all researchers so that they have a foundational understanding of on September 29, 2021 by guest. Protected copyright. Table 2 Problematic COVID-19 preprint articles Location of corresponding Source Issue author’s institution Finding SSRN21 Dataset is linked to two other retracted papers22 23 USA Retracted SSRN24 Dataset is