Evaluating Quality of Retraction Notices of Health and Life Science Articles Published by Authors Affiliated to Brazilian Institutions: a Systematic Review
To present and discuss the results of a systematic review that aimed to investigate the main reason for retractions of medical and life science research of authors affiliated to Brazilian academic institutions. For it, reason for retractions, citation of retracted publications and quality, availability and accessibility to data regarding retracted papers from the publishers are described considering the Committee on Publication Ethics – COPE recommendations.
Two independent reviewers searched for retracted articles since 2004 at PUBMED, Web of Science, BVS and Google Scholar databases. Indexed keywords from MeSH and DeCS in Portuguese, English or Spanish were used. Data was also collected from the Retraction Watch website (www.retractionwatch.com). This study was registered at PROSPERO systematic review database (CRD42017071647) and was published at PLOS ONE this year (1).
Results and Discussion
A final sample of 65 articles was retrieved; 20 were published at national and 35 international journals. Among the retrieved articles, plagiarism was the main reason for retraction (60%). Missing data were found in 57% of the retraction notices representing a limitation to this review. Missing information regarding reason for retraction (7%), who requested the retraction (3%) and endorsement by the authors (38.4%) were part of the retrieved articles. Retraction warnings such as a withdrawn/retracted red sign over the article were also nonexistent (37%). Measures to ensure research integrity has been widely discussed due to its social, economic and scientific impact (1,2,3,4). Most of the investigations on this subject focuses on the role of authors and institutions to promote research transparency, however research integrity is reinforced by editorial politics for publication of articles or retraction notices when needed. Hence, scientific journals play an important role to promote and reassure integrity. Complete information on retraction notices contributes to scientific transparency, avoids reproduction of unreliable research (5). Still, some retractions are vague and don’t comply with COPE recommendations. Despite the reasons that lead to the publication of incomplete retractions, we found complete and clear information at several retraction notices. This illustrates the disparities of policies and attitudes towards research mistakes. Regarding reasons for retraction, previous studies have shown that fraud and error have accounted for most of the retractions of biomedical articles (6,7); however, the present review revealed a larger number of retractions due to plagiarism. It is known by previous studies (8) that articles are still cited after been retracted. This systematic review found that 63% of the retracted articles was still cited as legit. More investigations are needed to understand the underlying causes of research misconduct and the continue citation of retracted articles as reliable. Brazil is a member of the BRICS (Brazil, Russia, India, China, South Africa) cooperative group that is responsible for some of the 1% most cited publications in the world (8). The rise on retracted publications accompanied increase on scientific production of the country with international partners. The majority of the retracted articles retrieved by this study were from public university and therefore, the researchers were public funded. This systematic review was the first to address the theme for this group considering the Brazilian context. Its results contributes for the discussion of means to improve research quality and reliability in the country, as well as to support more research on research.
The majority of the retracted publications retrieved by this systematic review did not complied COPE recommendations. It’s important to engage all scientific community, including publishers and editors at research integrity discussions in order to promote transparency at all levels of research: from it’s idealization, planning, execution, report, to possible retractions. Additionally, plagiarism was the main cause of retractions of health and life science publications of authors affiliated to Brazilian institutions. Coordinated action between institutions, funders, publishers and researchers is important to avoid scientific flaws. These results will pave the way for future research on the theme in order to understand the underlying causes of research integrity breaches and therefore, will contribute for a scientific production with higher quality and reliability.