ABSTRACT
Background: Studies related to the coronavirus disease 2019 (COVID-19) were frequently published as pre-prints prior to undergoing peer-review. However, several publications were later retracted due to ethical concerns or study misconduct. Although these studies have been retracted, the availability of their corresponding pre-prints has never been formally investigated, and may result in the spread of misinformation if they are being used to inform decision-making. Methods: Our objective was to conduct a systematic survey of retracted COVID-19 publications listed on the Retraction Watch database as of August 15th, 2021. We assessed the availability of corresponding pre-prints for retracted publications, and documented the number of citations and online views. Results: Our study included 140 retracted COVID-19 publications, and we could not retrieve corresponding pre-prints for 132 retracted publications in our study (94%). Although we were unable to find the majority of pre-prints, they had already been disseminated, with a maximal citation count of 593 and Altmetric score of 558,928. Conclusion: While it is reassuring that most corresponding pre-prints could not be retrieved, our study highlights the need for online platforms and journals to employ quality assurance methods to prevent the spread of misinformation through citation of retracted papers.
ABSTRACT
BACKGROUND: The torrent of research during the coronavirus (COVID-19) pandemic has exposed the persistent challenges with reporting trials, open science practices, and scholarship in academia. These real-world examples provide unique learning opportunities for research methodologists and clinical epidemiologists-in-training. Dr. David Moher, a recognized expert on the science of research reporting and one of the founders of the Consolidated Standards of Reporting Trials (CONSORT) statement, was a guest speaker for the 2021 Hooker Distinguished Visiting Professor Lecture series at McMaster University and shared his insights about these issues. MAIN TEXT: This paper covers a discussion on the influence of reporting guidelines on trials and issues with the use of CONSORT as a measure of quality. Dr. Moher also addresses how the overwhelming body of COVID-19 research reflects the "publish or perish" paradigm in academia and why improvement in the reporting of trials requires policy initiatives from research institutions and funding agencies. We also discuss the rise of publication bias and other questionable reporting practices. To combat this, Dr. Moher believes open science and training initiatives led by institutions can foster research integrity, including the trustworthiness of researchers, institutions, and journals, as well as counter threats posed by predatory journals. He highlights how metrics like journal impact factor and quantity of publications also harm research integrity. Dr. Moher also discussed the importance of meta-science, the study of how research is carried out, which can help to evaluate audit and feedback systems and their effect on open science practices. CONCLUSION: Dr. Moher advocates for policy to further improve the reporting of trials and health research. The COVID-19 pandemic has exposed how a lack of open science practices and flawed systems incentivizing researchers to publish can harm research integrity. There is a need for a culture shift in assessing careers and "productivity" in academia, and this requires collaborative top-down and bottom-up approaches.
Subject(s)
COVID-19 , Communication , Humans , Pandemics , Publishing , Research PersonnelABSTRACT
IntroductionTreatment of bipolar disorder is the focus of several clinical trials, however the understanding of the outcomes for establishing treatment effectiveness within these trials is limited. Further, there is limited literature which reports on the outcomes considered to be important to patients, indicating that patient perspectives are often not considered when selecting outcomes of effectiveness within trials. This protocol describes a systematic review which aims to describe the outcomes being used within trials to measure treatment effectiveness, commenting on the inclusion of patient-important outcomes within previous trials.Methods and analysisThis protocol is reported using the Preferred Reporting Items for Systematic Reviews and Meta-analyses Protocols statement. OVID MEDLINE, OVID Embase, OVID APA PsycINFO, Web of Science, the Wiley Cochrane Library, ClinicalTrials.gov and the International Clinical Trials Registry Platform databases will be searched for eligible studies. Screening, full-text and data extraction stages will be completed in duplicate using the Covidence platform for systematic reviews. Eligible studies will include clinical trials of interventions in bipolar disorder, in order to identify outcomes used to assess treatment effectiveness, and qualitative studies, to determine which outcomes have been reported as important by patients. Risk of bias for included studies will be assessed using the Cochrane Risk of Bias Tool for randomised controlled trials, and the Newcastle-Ottawa Scale for observational research.Ethics and disseminationThis review will involve dissemination to key stakeholders, including primary end users such as patients, clinicians and trialists. Knowledge translation tools will be generated to share the relevant conclusions of this review. Results will be communicated to the scientific community through peer-reviewed publications, conferences and workshops. No ethics approval will be sought as this study is based on literature.PROSPERO registration numberCRD42021214435.
ABSTRACT
BACKGROUND: Psychiatric disorders increase risk of neuropsychiatric disease and poor outcomes, yet little is known about the neuropsychiatric manifestations of COVID-19 in the psychiatric population. The primary objective is to synthesize neuropsychiatric outcomes of COVID-19 in people with preexisting psychiatric disorders. METHODS: Data were collected during an ongoing review of the impact of pandemics on people with existing psychiatric disorders. All study designs and gray literature were included. Medline, PsychInfo, CINAHL, EMBASE, and MedRx were searched from inception to September 1 2020. Risk of bias was assessed using a published tool that can accommodate all study types. Two independent authors screened the studies and extracted data. Data were narratively synthesized, as there were insufficient data to meta-analyze. Evidence was appraised according to GRADE. RESULTS: Four case reports were included, comprising 13 participants from three countries. Many large-sample, relevant papers were omitted for not reporting psychiatric history, despite reporting other comorbidities. Included participants (n = 13) were hospitalized with COVID-19 and appeared to meet criteria for delirium. Myoclonus, rigidity, and alogia were also reported. The most commonly reported preexisting psychiatric diagnoses were mood disorders, schizophrenia, and alcohol use disorder. CONCLUSIONS: People with preexisting psychiatric disorders may experience delirium, rigidity, myoclonus, and alogia during COVID-19 infection; although higher quality and longitudinal data are needed to better understand these phenomena. Relevant COVID-19 literature does not always report psychiatric history, despite heightened neuropsychiatric vulnerability within this population. TRIAL REGISTRATION: PROSPERO (CRD42020179611).